Tailored Cellulosic Materials By Physical Adsorption Of

Transcript

TAILORED CELLULOSIC MATERIALS BY PHYSICAL ADSORPTION OF POLYELECTROLYTES Andrew Marais Doctoral Thesis KTH Royal Institute of Technology School of Chemical Science and Engineering Department of Fibre and Polymer Technology Stockholm, 2015 ISBN 978-91-7595-500-1 TRITA-CHE Report 2015:14 ISSN 1654-1081 AKADEMISK AVHANDLING Som med tillstånd av Kungliga Tekniska Högskolan i Stockholm framlägges till offentlig granskning för avläggande av teknisk doktorsexamen fredagen den 22:a maj 2015, kl. 10:00 i Kollegiesalen, Brinellvägen 8, KTH, Stockholm. Avhandlingen försvaras på engelska. Copyright © Andrew Marais, 2015 Stockholm 2015 -ii- ABSTRACT The growing interest in using bio-based resources has made forest-based cellulose (used as fibre and nanofibril building elements) a good candidate for the development of new materials. In order to be used in commercial applications, cellulose must however be processed and/or functionalized to provide the final material with specific properties. This thesis presents (a) a way to improve the mechanical properties of traditional cellulose-based materials (paper), (b) an investigation into the structural and adhesive properties of the self-assembled hyaluronic acid thin films used to tailor the mechanical properties of paper materials, and (c) the preparation and functionalization of cellulose aerogels. In the first part of this thesis, the adsorption of polyelectrolytes onto pulp fibres (either as a monolayer or as multilayers) was studied as a way to improve the mechanical properties of paper materials. It was found that low amounts of adsorbed cationic amines were able to significantly improve the tensile properties of sheets made from treated fibres. Tensile testing of fibre crosses and microtomography revealed that this improvement in mechanical properties was due to an increase in both the interfibre joint strength and the interfibre contact area. By building up polyelectrolyte multilayers of hyaluronic acid (HA) and polyallylamine hydrochloride (PAH) onto the fibres, a threefold increase in both strain at break and tensile strength was achieved. In the second part, the structural and adhesive properties of (PAH/HA) thin films were investigated. Such films showed adhesive features stronger than those reported for bone structures containing mineralised collagen. Finally, wet-resilient porous cellulose aerogels were developed by freezedrying and crosslinking cellulose nanofibrillar gels. This material with high porosity and a high specific surface area was then used as a template to build three dimensional (3D) energy-storage devices using the Layer-by-Layer approach. Thin films of conductive materials were -iii- deposited into the bulk of the material, and 3D-interdigitated supercapacitors and batteries were built. The devices showed high capacitance and operated under extreme conditions of compression and bending, opening up numerous possibilities in the field of flexible electronics. -iv- SAMMANFATTNING Det växande behovet av att använda biobaserade resurser har dramatiskt ökat intresset för att använda cellulosa ifrån ved för utvecklingen av nya material. Cellulosa, i form av fibrer eller fibriller med cellulosa som huvudbeståndsdel, måste funktionaliseras att för naturligtvis kunna bearbetas användas i nya och/eller kommersiella tillämpningar. Denna avhandling omfattar (a) en metodik att förbättra pappers mekaniska egenskaper mha av adsorberade polyelektrolyter, (b) en undersökning av strukturen hos polymera multiskikt av PAH och hyaluronsyra (HA), deras adhesiva egenskaper, och hur de kan användas för att förbättra mekaniska egenskaper hos papper som tillverkats från multiskiktsbehandlade fibrer, och (c) tillverkning och funktionalisering av porösa cellulosa aerogeler. I den första delen av avhandlingen studerades adsorption av polyelektrolyter till massafibrer (i monolager eller multilager) och hur denna kan förbättra pappersmaterialets mekaniska egenskaper. Dragprovning av fiberkors samt röntgen mikrotomografi visade att förbättringen i de mekaniska egenskaperna kan förklaras både av en ökning i fogstyrkan och av kontaktytan mellan fibrerna. Genom att bygga upp multilager av hyaluronsyra och PAH på fibrerna ökade båda töjningen och draghållfastheten med en faktor tre. I den andra delen av avhandlingen undersöktes den övermolekylära strukturen och adhesions egenskaperna hos multilager av PAH och HA. Tunna filmer av (PAH/HA) uppvisar bättre adhesion egenskaper än vad som rapporteras för mineraliserade kollagenstrukturer i ben. Slutligen utvecklades metoder för att preparera en våtstabil, porös cellulosa aerogel. Materialet karakteriserades av en hög porositet (99%) och en hög specifik yta. Den våtstabila aerogelen användes sedan som templat för att kräddarsy olika 3D energilagrings material med multiskiktstekniken. Tunna filmer av ledande material byggdes upp i aerogelen, vilket gjorde det möjligt -v- att sedan tillverka 3D superkondensatorer och batterier. De tillverkade komponenterna uppvisade hög kapacitans och fungerar under extrem kompression och böjning, vilket öppnar nya möjligheter för cellulosabaserade material inom flexibel elektronik. -vi- användning av LIST OF PUBLICATIONS This thesis is based on the following papers, referred to in the thesis by roman numerals, and appended in the final section of the thesis. I. The use of polymeric amines to enhance the mechanical properties of lignocellulosic fibrous networks A. Marais and L. Wågberg Cellulose (2012), 19, 1437-1447 II. New insights into the mechanisms behind the strengthening of lignocellulosic fibrous networks with polyamines A. Marais, M.S. Magnusson, T. Joffre, E.L.G. Wernersson and L. Wågberg Cellulose (2014), 21, 3941-3950 III. Towards a super-strainable paper using the Layer-by-Layer technique A. Marais, S. Utsel, E. Gustafsson and L. Wågberg Carbohydrate Polymers (2014), 100, 218-224 IV. Robust and tailored wet adhesion in hyaluronic acid thin films T. Pettersson, S. Pendergraph, S. Utsel, A. Marais, E. Gustafsson, and L. Wågberg Biomacromolecules (2014), 15, 4420-4428 V. Nanometer-thick hyaluronic acid self-assemblies with outstanding adhesive properties A. Marais, S. Pendergraph and L. Wågberg Manuscript VI. Nanocellulose aerogels functionalized by rapid Layer-by-Layer assembly for high charge storage and beyond M. Hamedi, E. Karabulut, A. Marais, A. Herland, G. Nyström, and L. Wågberg Angewandte Chemie International Edition (2013), 52, 12038-12042 VII. Self-assembled three dimensional and compressible interdigitated thin film supercapacitors and batteries G. Nyström, A. Marais, E. Karabulut, L. Wågberg, Y. Cui and M. Hamedi Nature Communications (2015), accepted for publication -vii- 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. Part of the experimental work and preparation of the manuscript III. All the experimental work and preparation of the manuscript IV. Part of the experimental work and part of the manuscript V. All the experimental work and preparation of the manuscript VI. Part of the experimental work, and part of the manuscript VII. Most of the experimental work and part of the manuscript -viii- RELATED MATERIAL In addition to the appended papers, the work has resulted in the following presentations:  The use of polymeric amines to enhance the mechanical properties of fibrous networks A. Marais, L. Wågberg International Paper and Coating Chemistry Symposium (PCCS), Stockholm, Sweden (2012)  Tailoring of the LbL structures for optimized adhesion A. Marais, S. Pendergraph and L. Wågberg American Chemical Society (ACS) meeting, New Orleans, USA (2013)  The build-up of structured polyelectrolyte multilayers on pulp fibres to prepare stretchable paper materials A. Marais, S. Utsel, E. Gustafsson, L. Wågberg International Symposium on Wood, Fibre and Pulping Chemistry (ISWFPC), Vancouver, Canada (2013)  Layer-by-Layer films of polyallylamine and hyaluronic acid with superadhesive properties A. Marais, S. Pendergraph and L. Wågberg American Chemical Society (ACS) meeting, Dallas, USA (2014)  Superadhesive properties of LbL thin films of polyallylamine and hyaluronic acid A. Marais, S. Pendergraph and L. Wågberg Layer-by-Layer Assemblies: Science and Technology Conference, Hoboken, USA (2014) -ix- LIST OF ABBREVIATIONS ADS2000P Sodium poly[2-(3-thienyl)ethyloxy-4-butylsulfonate] AFM Atomic force microscopy BTCA 1,2,3,4-Butanetetracarboxylic acid CAT Contact adhesion testing CNF Cellulose nanofibrils CNT Carbon nanotubes CuHCF Copper hexacyanoferrate CV Cyclic voltammetry DPI Dual polarization interferometry HA Hyaluronic acid KPVS Potassium polyvinyl sulphate LbL Layer-by-Layer OTB Ortho-toluidine blue PAA Polyacrylic acid PAH Polyallylamine hydrochloride PDMS Polydimethylsiloxane PEDOT:PSS Poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) PEI Polyethyleneimine PVAm Polyvinylamine hydrochloride QCM-d Quartz crystal microbalance with dissipation SEM Scanning electron microscopy SHP Sodium hypophosphite X-μCT X-ray micro-computed tomography -x- TABLE OF CONTENTS 1. 2. INTRODUCTION .................................................................................... 1 1.1. Context ............................................................................................... 1 1.2. Purpose of the study ........................................................................ 2 BACKGROUND ....................................................................................... 3 2.1. From wood to fibres/nanofibres ..................................................... 3 2.2. Materials from wood ....................................................................... 5 2.2.1. Paper.......................................................................................... 6 2.2.2. Aerogels .................................................................................... 8 2.3. 3. 2.3.1. Polyelectrolyte adsorption ..................................................... 9 2.3.2. Layer-by-Layer assembly ..................................................... 11 EXPERIMENTAL ................................................................................... 17 3.1. Materials .......................................................................................... 17 3.1.1. Fibres / Nanofibres ................................................................ 17 3.1.2. Model surfaces ....................................................................... 17 3.1.3. Chemicals................................................................................ 18 3.2. 4. Surface modification of materials .................................................. 9 Methods ........................................................................................... 18 3.2.1. Fibres and paper (Papers I-III) ............................................. 18 3.2.2. Layer-by-Layer assemblies of (PAH/HA) (Papers IV-V) . 23 3.2.3. Wet-resilient cellulose aerogels (Papers VI-VII) ................ 26 RESULTS & DISCUSSION .................................................................... 31 4.1. Lignocellulosic fibre networks with tailored properties (Papers I-III) ................................................................................................ 31 4.1.1. Adsorption of polyamines monolayers on fibres for the preparation of strong paper materials ................................................. 31 -xi- 4.1.2. Build-up of polyelectrolyte multilayers onto fibres for strong and stretchable paper materials ............................................... 37 4.2. Layer-by-Layer assemblies of (PAH/HA) (Papers III-V) .......... 39 4.2.1. Build-up .................................................................................. 39 4.2.2. Adhesive properties .............................................................. 42 4.3. Nanocellulose aerogels as a platform for a variety of applications (Papers VI-VII) ...................................................................... 48 4.3.1. Preparation and characterization of the aerogels .............. 48 4.3.2. LbL functionalization of the aerogels.................................. 49 4.3.3. Preparation of three dimensional and compressible energy storage devices ........................................................................................ 51 5. CONCLUSIONS ..................................................................................... 57 6. FUTURE WORK ..................................................................................... 59 7. ACKNOWLEDGEMENTS .................................................................... 61 8. REFERENCES ......................................................................................... 63 -xii- INTRODUCTION 1. INTRODUCTION 1.1. Context Both ecological and economic crises have recently dramatically increased the need for a more sustainable society, and industries are facing new demands aiming at reducing production costs as well as utilising renewable resources to develop new functional materials. In this respect, one renewable raw material of major interest is wood and its main component: cellulose. Cellulose is the most abundant biopolymer on Earth and its annual worldwide production is estimated1 to be of the order of 1010-1011 t. Since its discovery by Payen in the nineteenth century2, cellulose has been widely investigated and used industrially to produce materials such as paper. Industries producing materials from forest resources are nowadays facing new challenges and it is necessary to produce traditional paperbased materials in a more cost-effective way, for example by decreasing the weight of packaging materials without influencing the mechanical properties of the end-product. In order to maintain the properties of such materials, the strength of its components must be enhanced, using monolayers or multilayers of polyelectrolytes for instance, and it is important to understand the structural mechanisms governing the strength enhancement of paper materials at the microscopic level. Moreover, paper can be envisioned as a substitute for oil-based plastics in 3D-shaped packaging, provided that its stretchability can be increased to allow 3D forming. A further challenge is to develop a new class of materials from wood that can be functionalized for use in a broad range of applications. -1- INTRODUCTION 1.2. Purpose of the study The overall aim of the work presented in this thesis has been to address some of the challenges faced by the forest, paper and cellulose industries, and to develop new functional materials from cellulose. More specifically, the objectives have been: (a) To prepare paper materials with enhanced dry strength by the adsorption of polyamines, and to investigate the microscopic mechanisms behind the strength enhancement. (b) To prepare stretchable paper materials by building up polyelectrolyte multilayers on the fibres by Layer-by-Layer deposition of polyallylamine hydrochloride (PAH) and hyaluronic acid (HA). (c) To investigate the structural and adhesive properties of Layerby-Layer assemblies containing PAH and HA. (d) To prepare and utilize porous materials from cellulose nanofibrils: characterization and functionalization via Layer-byLayer to prepare energy storage devices. -2- BACKGROUND 2. BACKGROUND 2.1. From wood to fibres/nanofibres Wood consists of an assembly of fibres held together by lignin, and the fibre wall is an ingenious nanocomposite where cellulose nanofibrils are embedded in a matrix of hemicellulose and lignin. The main constituents of the wood fibres are cellulose (about 50%), hemicelluloses (about 25%) and lignin (about 25%). The middle lamella holding the fibres together has a very high lignin content, but there is a distribution of lignin, hemicellulose and cellulose across the fibre wall and the secondary wall is dominated by cellulose and hemicellulose3. The wood fibres are typically a few mm long and about 20-50 μm in diameter3. The cylindrical fibre wall is composed of concentric layers, each consisting of fibril aggregates that are oriented in a certain direction along the fibre axis, usually called the fibril angle. The difference in fibril angle in the different layers leads to the characteristic mechanical properties of the fibres. The fibril aggregates are bundles of aggregated cellulose nanofibrils, which are in turn composed of cellulose chains (consisting of β-(1-4)-D-glucose repeating units) packed into cellulose crystals (a few Å), and are typically a few nm in diameter and over 1 μm in length3,4. The hierarchical structure of wood is depicted in Figure 1. Wood fibres can be extracted from wood chips by a chemical treatment, a mechanical treatment, or a combination of the two. The fibres used in the work presented in this thesis were extracted by a chemical process called the kraft process, in which the wood chips are reacted in an alkaline medium (sodium hydroxide and sodium sulphide) at elevated temperature (typically around 150 °C for several hours)5. Under such cooking conditions, the lignin is dissolved to an extent that allows the fibres to be extracted from the wood matrix. The wood chips are cooked until a desired yield (the ratio of the mass of pulp to the initial mass of wood) is reached. The remaining amount of lignin in the fibres is evaluated as the kappa number6 of the pulp, and this value is usually -3- BACKGROUND used to follow the cooking process and to characterize the pulp. After the kraft process, the liberated fibres can be used to prepare various paper materials and depending on the end application of the fibres, the pulp can be bleached in various sequences. If the extraction process is taken one step further, it is possible to extract nano-sized fibres, or cellulose nanofibrils (CNF), from wood fibres. HO HO HO O HO O O HO O OH HO O O HO OH Figure 1. The hierarchical structure of wood: from the tree to the fibre, nanofibril, and cellulose chain The liberation of CNF was first introduced in the 1980s by Turbak et al.7, but due to the high energy consumption and a lack of applications at that time, the process did not reach commercial application in the 1980s. -4- BACKGROUND Later, methods were developed to reduce the energy consumption by enzymatic treatment, beating and/or charging of the fibres prior to the mechanical treatment8–10. For example, the fibres can be chemically treated (a carboxymethylation is performed to increase the charge) before being mechanically treated in a high-pressure homogenizer. If shearing forces are applied to the carboxymethylated fibres, the fibre wall delaminates and it is possible to extract fibrils with dimensions in the nanometer range. Such fibrils have remarkable properties, among which their high charge (in the range of 500 μeq/g9), high aspect ratio and high modulus11 can be specially mentioned. Due to their high charge and high aspect ratio, CNF dispersions form gels at very low concentrations and their colloidal behaviour in solution is naturally directly influenced by the pH and salt concentration12. Significant efforts have been made to make possible the production of CNF at the industrial scale13. Nowadays, several plants around the world are producing CNF in industrial quantities and the scaling-up of CNF production paves the way for future applications in which the fibrils can be used to produce materials for packaging, electronic and magnetic applications, to mention only a few14–16. 2.2. Materials from wood Although a large variety of materials can be produced from wood, the focus in the work presented in this thesis is on fibrous networks (paper) and porous nanofibrous networks (aerogels), as depicted in Figure 2. -5- BACKGROUND Figure 2. Materials from wood: fibres used to prepare paper materials, and nanofibrils (extracted from the wood fibres) used to prepare aerogels 2.2.1. Paper Paper is a fibrous network formed during a process in which a fibre suspension is first dewatered, to form a wet web that is then pressed and dried. As the water is removed, the fibres are brought into intimate contact (due mainly to capillary forces) where van der Waals forces and hydrogen bonds can form, which are crucial for the development of paper strength17. The structure of a typical paper material is shown in Figure 3. The mechanical properties of paper materials are governed by three main parameters: the individual fibre strength, the interfibre joint strength, and the number of efficient interfibre joints per unit volume18–20. The endless need for fibrous materials with higher performances has led to the development of techniques aiming to increase the mechanical properties of paper. -6- BACKGROUND Figure 3. (Left) photograph of paper sheets and (right) SEM image of a paper sheet, showing the microscopic structure of a lignocellulosic fibrous network One of the most common methods, with low specificity, is to beat (or refine) the fibres21. The shear forces in the presence of water during the beating process produce hydrated/swollen fibres (which increases their flexibility) and promote both internal and external fibrillation (which increases the mechanical entanglements) as well as the creation of fines (increasing the specific surface area and hence favouring the contact between fibres), giving rise to a paper material with stronger mechanical properties22. However, beating also increases the density of the network and slows down the dewatering at the wet-end and in the press sections of the paper machine, and thus decreases the productivity of the papermaking process. In addition, beating is an energy-consuming process. A further drawback of beating is the shrinkage of the fibres during drying which leads to a shrinkage of the entire web during drying, which lowers both the productivity and the mechanical properties of the paper. Another possible way to modify the mechanical properties of a cellulosic network is to use chemical additives to modify the interfacial interactions between the fibres. Cationic starch23 and polyacrylamide resins24 have been widely used for decades as dry-strength additives, and more recently, other polyelectrolytes have been investigated as an easy and inexpensive way to improve the mechanical properties of paper25–28. The focus in the present study is on two polyamines: polyallylamine -7- BACKGROUND hydrochloride (PAH) and polyvinylamine (PVAm), used as dry-strength and, to some extent, wet-strength additives. Apart from conventional macroscopic tensile tests, the mechanical properties of a fibrous network can be investigated at the microscopic scale by measuring the tensile strength of the interfibre joints. This has been challenging in the past because of the difficulty in handling individual fibre crosses, and also because of the heterogeneity of the fibres, giving rise to measurements with large variations. In the present study, a direct method was used, where individual fibre crosses are formed and mechanically tested. This was first shown in the 1960s29, and was followed by a range of similar studies30,31. Nowadays, new techniques are available to test single fibre crosses in various modes of loading following a protocol which reduces the probability of sample failure during preparation32–34. 2.2.2. Aerogels A significant amount of research has recently been undertaken in order to develop new types of materials from CNF, including porous materials such as foams or aerogels35–37. An aerogel can be defined as a material initially in a gel state in which the liquid has been replaced by a gas. The main characteristics of an aerogel are: very low density, high porosity, high specific surface area and insulation properties. A cellulose aerogel can be prepared by freeze-drying a CNF gel. The freeze-drying method consists in freezing the sample in liquid nitrogen (i.e. at -196 °C) followed by sublimation of the frozen water. The sublimation prevents the structure from collapsing, and a porous network can be obtained36. The macroscopic and microscopic structures of a CNF aerogel are shown in Figure 4. The porosity of a cellulose aerogel can reach 99% and the density is typically 0.02 g/cm3. Despite many interesting properties, the use of cellulose aerogels is somewhat limited because they are not stable in water and decompose. -8- BACKGROUND Figure 4. (Left) Photograph and (right) SEM image of a cellulose aerogel One way of solving this problem is to chemically crosslink the CNF network. Zhang et al. 38 suggested a route to crosslink the aerogels using polyamide-epichlorhydrin (PAE). PAE resins have been extensively used in the paper industry as wet strength agents for hygienic papers, filter papers and wallpaper17. However, the toxicity of PAE is a strong hindrance to wider use, and it is therefore important to develop other crosslinking routes for wet-resistant cellulose aerogels where the negative charges of the aerogel are preserved or increased. 2.3. Surface modification of materials Although wood-based materials display many interesting features, it is often necessary to modify them in order to give them specific functionalities. There are two ways of modifying the surface of a material: a chemical way (by covalently grafting species onto the surface for example) and a physical way, relying on non-covalent, physical adsorption. The focus in the present work is on the physical adsorption of charged species such as polyelectrolytes. 2.3.1. Polyelectrolyte adsorption A polyelectrolyte is by definition a polymer containing one (or more) potentially charged groups in its repeating unit. Polyelectrolytes are ubiquitous and play a major role in our lives: from our body (e.g. DNA) -9- BACKGROUND to the food we eat everyday (e.g. carboxymethyl cellulose, used as a thickener). Polyelectrolytes may be either strong or weak. The former are not affected by the pH whereas the charge density of the latter is directly linked to the pH. Due to the presence of charges, polyelectrolytes are soluble in water. The adsorption of polyelectrolytes onto oppositely charged surfaces has been extensively investigated over the past decades and the driving force for this phenomenon is the gain in entropy following the release of counter-ions upon adsorption39–42. The following parameters are crucial to an understanding of the behaviour of the polyelectrolyte/substrate system: - The pH, which controls the degree of dissociation and thereby the charge density of the polymer (in the case of weak polyelectrolytes) and the substrate; - The surface charge of the substrate, since there is a 1:1 ratio between the amount of adsorbed charge and the surface charge of the substrate (also termed pure electropsorption)43; - The salt concentration and salt type, since an increased salt concentration decreases the entropy gain upon adsorption, which in turn decreases the adsorption, but also reduces the intramolecular repulsion, making the polymer coil-up and increasing the adsorbed amount in the case of a porous substrate; - The porosity of the surface (or the molecular mass of the polymer), which determines whether the polyelectrolyte is adsorbed exclusively onto the surface or whether it penetrates into the substrate; - The solvent: in a so-called good solvent or poor solvent, the polymer chains have respectively a strong or weak affinity for the solvent and a tendency to adopt an extended or coiled conformation. The quality of the solvent may also influence the -10- BACKGROUND adsorption if there is a considerable non-ionic contribution to the adsorption process. Since pulp fibres are negatively charged due to the carboxylic groups on the hemicelluloses and oxidized lignin4, it is possible to adsorb cationic polymers onto their surfaces. Polyelectrolytes are a key feature of the papermaking process, since they are used to tailor the properties of paper from the wet state (e.g. retention, flocculation) to the dry final material (e.g. wet and dry strength)44–47, and they are nowadays used in the production of almost all paper grades. Polyelectrolytes can be deposited onto charged substrates as monolayers, as multilayers or as complexes to physically modify the surface properties of the material onto which they are adsorbed48,49. 2.3.2. Layer-by-Layer assembly Principle The Layer-by-Layer (LbL) technique involves building up a film by depositing alternating cationic and anionic polyelectrolytes and/or particles onto a charged substrate, as shown in Figure 5. The principle of LbL assembly was first introduced for colloids in the 1960s50 and was then developed and extended to a broad range of polymers and particles by Decher in the 1990s49,51,52. This work was a breakthrough in physical surface modification and the interest in this technique has continuously increased since then. - - - - - - Substrate - - P+ Rinse + + + + + + + + + + + + + - - - - - - - - Substrate PRinse - - - - - - + + + + + + + + + + + - - - - - - Substrate - - + + + - - Figure 5. Simplified scheme of the Layer-by-Layer deposition technique, starting from a negatively charged substrate and depositing consecutively cationic (P+) and anionic polymers (P-), forming the first bilayer -11- BACKGROUND LbL films can be assembled either by dipping the substrate in a solution or dispersion, or by spraying the solution onto the surface. The spraying technique is much faster than the dipping technique because of the shorter time necessary to deposit each layer and also because the rinsing step can be avoided. However, the thickness of the LbLs assembled by dipping is greater than that of films obtained by spraying53,54. The fibres and model surfaces investigated in the present work were LbL-coated using the dipping approach, but a new filtration method was introduced for the LbL deposition on aerogels. The assembly of LbL films is a very versatile technique that can be used to tailor the properties of a surface, and specific functionalities can be achieved with only a few layers, i.e. with nanometre-thick films. The protocol for the buildup of LbL structures is very simple and environment-friendly, since water is most commonly used as solvent. Moreover, the LbL technique allows a large variety of combinations since the only requirement is to use charged species. It is even possible to use uncharged species, provided that other driving forces, of enthalpic or entropic origin, can ensure interactions with the second LbL component, as is the case for LbL films of polyethylene oxide (PEO) and polyacrylic acid (PAA)55. For all these reasons, the LbL technique is nowadays used in a broad range of applications including gas barrier56,57, biomimicking of natural structures58, superhydrophobic surfaces59, electronics60, drug delivery61–64, and other biomedical applications65–69. Linear vs. super-linear growth As with monolayer polyelectrolyte systems, the build-up of polyelectrolyte multilayers is extremely dependent on the pH and on the salt concentration. Together with the intrinsic properties of the polymers (such as molecular weight and charge), these parameters control the growth of the multilayer structure, which can be either linear or superlinear70–73. Although the mechanisms governing the transition from linear -12- BACKGROUND to super-linear growth are still debated in the literature, two models can be mentioned: the first relying on the interdiffusion of one species through the LbL film71, and the second being based on the so-called Island model74, where the adsorbed polymer forms islands on the substrate, increasing the surface area available for further deposition. The thickness of the LbL films can thereby be controlled and tailored by adjusting the number of layers deposited, the pH of the solutions, the salt concentration, and also the nature of the species deposited (polymers or particles). LbL on wood fibres The LbL technique has already been used for wood fibres as a way to functionalize75,76 or to increase the strength of fibrous materials, typically using PAH and PAA77, or cationic and anionic starch78. The deposition of polyelectrolyte multilayers onto the surface of lignocellulosic fibres has been shown to have a strong impact on the interfibre joints, which is the molecular/microscopic reason why differences can be seen on a macroscopic scale79. Adhesive properties of LbL thin films Different approaches can be used to determine the adhesive properties of solid materials, where different length scales are involved. In the present work, two approaches were used to characterize the adhesive properties of nanometer-thick LbL films: a colloidal probe atomic force microscopy (AFM) technique, and a contact adhesion testing (CAT) technique. AFM is a microscopy technique that can be used to image the topography of a substrate with a nanometre resolution. It is based on a cantilever scanning the substrate and detecting the force of interaction between the tip of the cantilever and the substrate. This technique can also be used to measure adhesion between films/particles80. A common procedure consists of attaching a colloidal probe (typically a silica particle with a diameter in the micrometre range). The probe can be LbL-coated and -13- BACKGROUND allowed to approach a similarly coated surface. The force profile is recorded throughout the measurement. The force necessary to separate the two surfaces brought into contact is called the pull-off force, and is one parameter describing the adhesive properties of a material. The AFM colloidal probe technique allows the measurement of pull-off forces for films with thicknesses down to the nanometre level and is therefore adapted to LbL studies. The number of layers, the outermost layer, the molecular mass of the two compounds and the contact time have been shown to be crucial parameters for the adhesive properties of LbL films81. Colloidal probe AFM have been used to investigate LbL systems such as (PAH/PAA)81 and (PAH/PSS)82 to mention a few. For CAT measurements, model surfaces such as polydimethylsiloxane (PDMS)83, glass and silicon wafers can be used as substrates for LbL deposition and evaluation of the adhesive properties of LbL films. The JKR technique84, based on the adhesion between a sphere/hemisphere and a flat surface, can be used to provide information about the properties of LbL films, as shown by Nolte et al.85, who investigated adhesive interactions of (PAH/PAA) films. However, it was shown that the roughness and stiffness of LbL films in the dry state hinders the establishment of intimate contact between the coated substrates, and subsequently the adhesion between the LbL thin films. LbL films of hyaluronic acid One of the polymers often used in LbL systems for biomedical applications is hyaluronic acid (HA), a linear biopolymer present in the human body (eyes, synovial liquid etc.). It is commonly used in medical applications such as biomedicine, surgery and drug delivery, and it is a polysaccharide of major interest due to some of its properties (hydrophilicity, rheological behaviour, lubrication ability and so on, to mention only a few)86–88. The presence of a negatively charged group in its repeating unit makes it possible to use hyaluronic acid in LbL films. Often associated to poly-L-lysine (PLL), HA has been used in LbL films -14- BACKGROUND for drug delivery and cell adhesion, for example61–64. Biological structures can also be mimicked using HA, as shown for example by investigations where a skin tissue replica was designed by LbL89. Other examples of tissue engineering using LbL have also been reported90. Although well studied in the above-mentioned applications, the use of HA in biomedical applications could be extended. In this thesis, we investigate HA as a strong bio-adhesive by studying the adhesive properties of LbL films where HA is associated with PAH. LbL and energy storage Among the several approaches that can be used to store electrical energy, the most common ones are batteries and electrochemical supercapacitors91,92. The main difference between these two types of device is that batteries store energy as chemical reactants, whereas supercapacitors store energy as charge. Both batteries and supercapacitors are composed of two electrodes (anode and cathode) separated by an electrolyte, which ensures the ion transport between the two electrodes and prevents direct contact between them. The functioning of a battery is based on redox reactions (faradaic processes) at the interface between the electrodes and the electrolyte. On the other hand, supercapacitors store charge in an electric double layer at the interface between the electrode and the electrolyte. These operating differences give rise to devices with different properties. The processes involved in the charge storage in supercapacitors are much faster than the faradaic processes in batteries. One consequence is that the charge/discharge timescale is seconds for supercapacitors, and can be up to days for batteries. This translates into lower energy densities but higher power densities for supercapacitors than for batteries. Upon operation, the stability of the electrode materials in a battery is affected (as the molecular structure can be changed with the redox reactions), and this makes the lifetime of a battery (in terms of number of charge/discharge cycles) shorter than that of a supercapacitor. -15- BACKGROUND Among the conductive materials used in emerging storage energy technologies, carbon nanotubes (CNT) have shown promising features as electrode material93,94. CNT are nanometer-sized cylindrical structures made of graphite sheets, characterized by a high aspect ratio, and displaying high mechanical and electrical conductivity properties95,96. One way of incorporating charges in the CNT structures is to functionalize them with carboxylic acid groups. The structure of a singlewall carbon nanotube functionalized with carboxylic acid groups is schematically shown in Figure 6. The presence of charged group makes CNT suitable for LbL deposition63, and LbL films containing CNT have already been assembled for strong composites60 or energy storage applications97,98. O C OH Figure 6. The structure of a carboxymethylated CNT, used as active material in the 3D energy storage devices Traditionally, energy storage devices are two-dimensional and although several studies have been carried out to build 3D energy storage devices99, these are limited due to the design complexity of such devices, and the difficulty in depositing material in a porous template. The ability to assemble three-dimensional devices with interdigitated electrodes using a simple and fast procedure such as LbL would pave the way for a range of potential applications. -16- EXPERIMENTAL 3. EXPERIMENTAL The main procedures and chemicals used in the different studies are described in the following section. For more information, the reader is referred to the relevant section in the corresponding paper. 3.1. Materials 3.1.1. Fibres / Nanofibres The fibres used in Paper I were from specially prepared, laboratorycooked kraft pulps (SCA Research AB, Sundsvall, Sweden). They were never-dried, unbleached and unbeaten softwood fibres (pure spruce) cooked to three different kappa numbers: 34, 75 and 107 (denoted K34, K75 and K107 respectively). The K75 fibres were used in Paper II. In Paper III, a totally chlorine-free bleached kraft pulp from softwood was used (SCA Forest Products, Östrand Mill, Sundsvall, Sweden). The pulp was received as dry sheets which were disintegrated, washed and converted to the sodium form prior to use, following a procedure described earlier23. The CNF used in Papers III, VI and VII were provided as a 2 wt% gel by Innventia AB, Stockholm, Sweden. The material was produced and characterized following procedures described by Wågberg et al.9 3.1.2. Model surfaces The silicon wafers (p-type, MEMC Electronics Materials, Inc., Novara, Italy) used in Papers III, IV and V as substrates for the LbL deposition were first rinsed with Milli-Q water, ethanol and Milli-Q water, and then blown dry with N2. They were then plasma-treated (Model PDC 002, Harrick Scientific Corporation, NY, USA) for 2 min at 30 W. -17- EXPERIMENTAL 3.1.3. Chemicals PAH (with molecular masses 15 kDa and 56 kDa), PAA (240 kDa), HA (1.6 MDa), 1,2,3,4-Butanetetracarboxylic acid (BTCA) and sodium hypophosphite (SHP) were provided by Sigma Aldrich and used without further treatment. Polyvinylamine (PVAm) polymers with molecular masses 45 kDa and 340 kDa and a rate of hydrolysis greater than 90% were provided by BASF (Ludwigshafen, Germany) under the commercial names Lupamin 5095 and Lupamin 9095 respectively. These commercial products were dialysed against Milli-Q water and freeze-dried prior to use. Branched PEI (60 kDa) was provided as a 50 wt% aqueous solution by Acros Organics (U.S.) and was used as received. The Sylgard 184 and curing agent used to prepare the PDMS probes were purchased from Dow Corning (U.S.). Single Wall carbon nanotubes (CNT) functionalized with carboxyl groups were purchased from Carbon Solutions. CNT dispersions were prepared by ultrasonication and centrifugation to remove the largest nanoparticle aggregates. 3.2. Methods 3.2.1. Fibres and paper (Papers I-III) Polyelectrolyte titration The polyelectrolyte titration method was used to determine the surface charge of the fibres and the charge density of the cationic polymers, and to determine the adsorption isotherms of polyelectrolytes on the fibres100. In this method, a solution is titrated with potassium polyvinyl sulphate (KPVS) in the presence of a cationic indicator, ortho-toluidine blue (OTB). Since KPVS is negatively charged, it neutralizes the cationic charges present in the solution. When all the cationic charges have been neutralized, KPVS accumulates and forms a coloured complex with OTB. By recording the absorbance of -18- the solution throughout the EXPERIMENTAL measurement, it is possible to detect the equivalent point and thereby determine the amount of charges initially present in the solution101,102. Polyelectrolyte adsorption A polyamine monolayer was adsorbed by pouring the desired amount of polymer into a 5 g/L fibre suspension under stirring with a background salt concentration of 5·10-4 M of NaHCO3. The pH was close to 8 and the adsorption time was 10 min. The LbL build-up on the pulp fibres was carried out on a larger scale since a sheet was made after each deposited layer. Adsorption isotherms were first determined in order to evaluate the amount of polyelectrolyte to be added as the first layer. To predict the amount of polymer to be added in the next layer, dual polarization interferometry was used103. The cationic polymer (i.e. the first layer) was added to a 3.6 g/L fibre suspension under stirring. The adsorption time was 10 min and, for the systems with added salt, the salt concentration was set to 10 mM of NaCl. After the adsorption, 1L of the suspension was taken out and used to prepare a sheet. P- S0 V0 S2i+1 [ (V0-2i-1) S2i+2 (V0-2i-2) [ P+ n Figure 7. Schematic of the protocol to build LbL films on fibres, for i ϵ ⟦0;n-1⟧, where V0 is the initial volume of fibre suspension, Sk is the k-layer treated sheet, P+/P- is the cationic/anionic polymer, and n is the number of bilayers The remaining suspension was dewatered in a Büchner funnel in order to remove the excess polymer. Water was thereafter added to form a 3.6 g/L -19- EXPERIMENTAL suspension to which the next polymer was added. This sequence was repeated until the desired number of layers (i.e. 10 in this study) was achieved, as depicted in Figure 7. Handsheet preparation and evaluation A Rapid Köthen instrument (PTI, Vorchdorf, Austria) was used to prepare the handsheets. Each sheet was dried under restrained conditions at 93 °C under a reduced pressure of 95 kPa for 15 min. All the sheets were then conditioned at 23 °C and 50% RH before further testing. Uniaxial tensile testing was conducted according to the SCAN-P 67:93 Standard using a horizontal tensile tester (PTI, Vorchdorf, Austria). The tensile index, as well as the stress and strain at break, were used to evaluate the mechanical properties of the paper. Nitrogen analysis was used to determine the nitrogen content of paper samples, and thus to assess the amount of polymer adsorbed. This technique consists in burning the sample at 1050 °C in an oxygen-poor medium. All the nitrogen-containing groups are then converted to nitrogen oxides, which in turn react with ozone to form excited nitrogen oxide molecules. As these excited molecules decay, light is emitted. A photomultiplier detects the emitted light and the instrument returns a number of counts, corresponding to the amount of emitted light and thus to the amount of nitrogen. For each polymer, a calibration curve was established in order to link the number of counts to the amount of polymer. The tests were performed with the aid of an ANTEK MultiTek (by PAC, Houston, Texas, USA). Interfibre joint strength evaluation The fibre crosses for interfibre joint strength measurements were prepared from untreated and PAH-saturated K75 fibres, according to the following procedure: a few fibres were suspended in droplets of deionized water on a Teflon-coated steel surface. A similar steel plate was then placed on top of the former and a weight was added to reach a -20- EXPERIMENTAL nominal drying pressure of 2.6 kPa. The system was heated for 2 hours at 108°C. After drying and conditioning at 23°C and 50% RH, overlapped fibres were selected and tested. The individual fibre crosses were thereafter fixed on specially designed sample holders using liquid adhesives as shown in Figure 8. Crossed fibre Loaded fibre 0.5 mm Glue X Y Figure 8. Photo micrograph showing the fibre cross mounted on the sample holder prior to tensile testing The fibre crosses were subjected to either a shearing or a peeling mode of loading (only shearing is discussed in the present thesis). The mechanical testing was performed with the aid of a commercial tensile testing machine with an in-house constructed grip-system. The conventional shearing type of loading, commonly considered to be the interfibre joint strength, was performed by fixing both ends of one fibre while one end of the loaded fibre was subjected to a prescribed displacement at a rate of 2 μm/s in the direction of the fibre axis. The recorded force at rupture (Pmax) was normalised with respect to the macroscopic overlapping area (Aoverlap) of the crossing fibres, measured in a high magnification transmission microscope, in order to take into account differences in fibre dimensions, as wider fibres are expected to carry more load since a larger area is available for bonding. The interfibre joint strength is defined as: -21- EXPERIMENTAL  overlap  Pmax Aoverlap Since the strength distribution of interfibre joints prepared and tested using this method is skewed34, the median is the best measure of the central tendency of the distribution. The median can be used to compare the populations of modified and unmodified specimens. Moreover, since the direct testing of interfibre joints is difficult, time-consuming and subject to large variations, relatively small sample sizes are usually obtained. A univariate bootstrap analysis was therefore performed using 10000 resamplings of the data to provide an evaluation of the stability of the measurements. The median was then determined as the arithmetic mean of the resampled median calculations. X-ray micro-computed tomography (X-μCT) and image analysis In order to acquire information about the network structure, X-μCT scans were performed using a Bruker Skyscan 1172 at constant room temperature (23°C) and a relative humidity of 30% (±2%). 1200 projections were acquired and the resulting images had 4000 x 2664 pixels, 0.8 μm in size. 3D images with a voxel size of 0.8 μm were reconstructed from the images using NRecon 1.6.9 (Bruker, Belgium). A resolution of about 1 μm is too poor to resolve molecular interactions on a joint level, but it can provide information about the changes in the sheet structure on a submicrometre level when chemical additives are added. A segmentation of the fibres was performed using the method described by Wernersson et al.104, in which the user manually follows the centre line of a fibre introducing control points using different cutting planes, as illustrated in Figure 9. By following the demarcated fibre and analysing its neighbourhood, it is possible to estimate whether the fibre is free or bonded, as well as the cross-sectional area and circumference of the fibres. It is thus possible to segment the marked fibre and count the number of contacts per unit length. At the intersection of two segmented fibres, the contact area of the interfibre joint can also be evaluated. -22- EXPERIMENTAL Figure 9. Illustration of the different planes used to mark the centre line of a fibre: (a) top view, (b) section used to indicate the depth of the centre line, (c) section perpendicular to the centre line of the fibre, used to manually ensure that the spline stays at the centre of the marked fibre 3.2.2. Layer-by-Layer assemblies of (PAH/HA) (Papers IV-V) LbL deposition on model surfaces The LbL deposition of the (PAH/HA) system onto model surfaces (silicon wafers or PDMS spheres) was achieved by dipping the surfaces consecutively in PAH and HA solutions (of concentration 0.1 g/L), with a rinsing step in Milli-Q water after each polymer deposition. The dipping times were 10 min in the polyelectrolyte solutions and 5 min in the rinsing bath. The pH of the PAH and HA solutions was 4.9 and 5.8 respectively. -23- EXPERIMENTAL Thickness and growth of (PAH/HA) assemblies Dual polarization interferometry (DPI) is an optical method used to study the build-up of a polymer film on a substrate103,105. A laser light beam is passed through waveguides composed of a reference and the substrate to be modified. Depending on the amount of material deposited onto the substrate, the output from the two waveguides generates interference fringes that can be traced back to the thickness of the adsorbed film103. A Dual Polarization Interferometer Analight Bio200 from Farfield Sensors Ltd (Manchester, UK) was used to study the (PAH/HA) assemblies with a continuous flow of 100 μL/min. All the samples had a concentration of 0.1 g/L during the adsorption step. Nitrogen-doped silica chips were used as deposition substrates. Quartz-crystal microbalance with dissipation (QCM-d) is a technique used to record the real time adsorption of a polymeric compound to a surface in terms of mass and viscoelastic properties. A crystal is subjected to an oscillation and the change in frequency due to the adsorption of a polymer can be recorded and related to the mass of adsorbed material, while the energy dissipation provides information about the viscoelasticity of the adsorbed layers . A QCM E4 (Q-Sense AB, Västra 106 Frölunda, Sweden) was used to study the multilayer build-up with a continuous flow of 100 μL/min. The substrates were silicon oxide crystals with an active surface of sputtered silica prepared in the same way as the silicon oxide wafers described earlier. AT-cut quartz crystals with a 5 MHz resonance frequency were used. The change in frequency is equivalent to a change in adsorbed mass according to the Sauerbrey model107. This model assumes rigidly attached layers, and the adsorbed amount determined contains both polymer and water coupled to the adsorbed layer. Imaging and adhesion properties by Atomic Force Microscopy An AFM MultiMode IIIa (Veeco Instruments Inc. Santa Barbara, CA) was used both for imaging and for adhesion measurements. For tapping -24- EXPERIMENTAL mode imaging in air, an EV scanner was employed using standard noncontact mode silicon cantilevers with a spring constant in the range of 32 to 70 N/m (TAP150, Bruker, Camarillo, CA). The force measurements were performed by capturing normal force curves in water. A silica particle (Thermo scientific, CA) with a diameter of approximately 10 μm was attached, with the aid of a manual micromanipulator and an Olympus reflection microscope, to the end of the tipless cantilever, using a small amount of a two-component epoxy adhesive (Strong epoxy rapid, Casco). For each probe, the exact diameter of the probe was determined. Adhesion properties of (PAH/HA) using contact adhesion testing (CAT) PDMS spheres with diameters of about 3 mm were prepared by injecting droplets of PDMS into a bath of water heated to 65 °C. The spheres were then cured by standing in this water bath for 2 hours. In order to remove the non-cured PDMS fractions, the spheres were thereafter washed in heptane. For both the adsorption steps and the adhesion measurements, the spheres were attached to a piece of glass slide using PDMS, as shown in Figure 10. The PDMS spheres were LbL-coated by dipping in alternate PAH and HA solutions, with a rinsing step between each step. Figure 10. Photograph of a probe consisting of a PDMS sphere attached to a glass slide A specially designed CAT instrument was used to characterize the adhesive properties of the LbL thin films. The instrument consists of a stage that can be moved up and down with a micrometer-controlled -25- EXPERIMENTAL translation stage with a horizontal aluminium holder on which the probe (PDMS sphere) is attached to a transparent glass slide. The flat surface (silicon wafer) is attached to a sample holder standing on a precision balance and the two surfaces are brought into contact until a certain trigger load is reached. The stage is then moved upwards to pull the surfaces apart and the evolution of the load as a function of time (and distance) is recorded. The instrument is connected to an optical microscope as well as a camera, allowing images of the contact zone between the two surfaces to be recorded throughout the measurement. The never-dried LbL coated surfaces were brought into contact in the wet state, dried overnight in contact, and thereafter pulled apart at a velocity of 100 μm/min, as depicted in the schematic shown in Figure 11. Contact Dry Pull-off Figure 11. Schematic of the adhesion measurements, where the LbL-coated PDMS sphere was brought into contact with the LbL-coated silicon wafer in the wet state. The surfaces were then dried in contact before being pulled-apart 3.2.3. Wet-resilient cellulose aerogels (Papers VI-VII) Preparation 1,2,3,4-butanetetracarboxylic acid (BTCA) and sodium hypophosphite (SHP) were added to a CNF gel (2 wt%) in ratios of 1:1 and 1:2 respectively. The mixture was then stirred for 20 min using an Ultra Turrax T25 (IKA, Germany) at 10 000 rpm. The gel was thereafter frozen in aluminium cups using liquid nitrogen, and freeze-dried. The gels were then cross-linked in the dry state by heating at 170°C for 4 min. The aerogels were thoroughly rinsed with -26- EXPERIMENTAL Milli-Q water prior to use in order to remove any unreacted crosslinker or catalyst. The procedure is depicted in Figure 12. O HO O O HO O HO HO O O O O O HO O O O HO HO O OH HO O OH O HO O O HO OH O O O O HO O OH OH Figure 12. Schematic of the preparation of the crosslinked cellulose aerogel by freeze-drying and heating the gel LbL functionalization of the aerogels The aerogels were functionalized by LbL deposition. Since aerogels are porous materials, the commonly-used dipping approach could not provide a homogeneous coating through the bulk aerogel in a reasonable time. A new method was therefore introduced, where the liquid polymer solution (or particle suspension) was forced through the aerogel by pouring the solution on top of the aerogel and applying suction to the aerogel, as schematically outlined in Figure 13. The concentration of the solutions/dispersions used was 1 g/L and no pH adjustment was made. Figure 13. Schematic of the filtration method used to build LbL assemblies within the aerogels by forcing the liquid through the porous structure -27- EXPERIMENTAL Characterization of the aerogels The total charge of the aerogels was determined using conductometric titration. An automatic titrator was used and HCl was added to protonate the carboxylic groups of the aerogel. Scanning electron microscopy (SEM) was used to image the structure of the aerogels, using a Hitachi S-4800 field emission scanning electron microscope. Compression tests were performed on wet samples using an Instron 5566. The samples were tested at a speed of 50%/min in compression and 30% in extension, and cycles were performed between 10% and 80%. Preparation of energy storage devices The vacuum-assisted LbL-method was used to build layers of active materials inside the aerogel structure to assemble 3D interdigitated supercapacitors. An aerogel sample was first cut to the desired shape, and a (PEI/CNT)5 multilayer was then formed in the aerogel to create the first electrode. After drying, the first electrode was contacted using silver paint and a copper wire, and the contact region was waxed using paraffin wax to prevent any wetting of this region. The separator was then built using an LbL system composed of (PEI/PAA)30. Finally, the second electrode was assembled and contacted in a similar way as the first electrode, as outlined in Figure 14. Hybrid batteries were also formed by incorporating copper hexacyanoferrate nanoparticles (CuHCF) as a cathode material. -28- EXPERIMENTAL Aerogel substrate Substrate Electrode1 LbL Electrode 1 Masking + contact LbL Separator LbL Electrode 2 Masking + contact Separator Electrode 2 Figure 14. Schematic of the step-by-step procedure to build 3D interdigitated energy storage devices in the aerogels at the (top) macro-, (middle) micro- and (bottom) nano-level Electrochemical characterization of the energy storage devices The LbL-assembled 3D supercapaci tors and batteries were characterized electrochemically using a PGSTAT302N potentiostat (Eco Chemie, The Netherlands). The supercapacitors were tested in a 1 M Na2SO4 electrolyte, and cyclic voltammetry was performed using scan rates of 5– 50 mV/s. Constant current charging and discharging was carried out using currents between 25 and 200 μA, corresponding to current densities between 0.07–0.5 A/g based on the total carbon mass of the device. The voltage window was 0–0.8 V. The copper hexacyanoferrate cathode and hybrid 3D batteries were tested using cyclic voltammetry at a scan rate of 1 mV/s in a (1 M KNO3; 0.01 M HNO3) electrolyte. -29- EXPERIMENTAL For measurements under compression, a setup using a digital caliper was used, as shown in Figure 15, and CVs were recorded at each compression step using a scan rate of 10 mV/s. Figure 15. Photograph of the setup used to characterize the electrochemical behaviour of the devices under compression -30- RESULTS & DISCUSSION 4. RESULTS & DISCUSSION 4.1. Lignocellulosic fibre networks with tailored properties (Papers I-III) 4.1.1. Adsorption of polyamines monolayers on fibres for the preparation of strong paper materials Adsorption isotherms In order to estimate the amount of polyelectrolyte that could be adsorbed onto the fibres, adsorption isotherms were determined for different combinations of fibre/polyelectrolyte using polyelectroyte titration of the filtrates obtained from dewatered fibres suspensions with different amounts of added polyelectrolyte. Fibres from pulps with different kappa numbers were used and the charge characterization of these is displayed in Table 1. As expected, the pulp yield had a strong influence on both the surface and total charge of the fibres, since charged species such as oxidized lignin and charged hemicelluloses are removed during the cooking process. A maximum in the surface charge was observed due to a competition between the accessibility of charges within the fibres (which increases during cooking) and the total charge of the fibres (which decreases during cooking)108. The adsorption isotherms for the K75 pulp are shown in Figure 16. The typical behaviour (an increase followed by a levelling-off of the adsorbed amount, corresponding to the saturation adsorbed amount) can be detected, and lower molecular mass polymers gave rise to higher adsorbed amounts, which shows that, in addition to adsorption at the surface, there can also be penetration of polymer into the fibre wall to a certain extent. In the case of the K75 pulp, the pore radius (11 nm) is in the same range as the radius of gyration of low molecular mass PAH (10 nm)109, making it possible for polymer chains to penetrate the fibre wall. -31- RESULTS & DISCUSSION Table 1. Surface charge, total charge, lignin content and average pore size of the fibres from the K34, K75 and K107 pulps Pulp Yield Surface Total Lignin Pore radiusd chargea chargeb contentc (%) (μeq/g) (μeq/g) (w%) K34 49.7 5.6 74 3.3 17 K75 56.4 9.3 137 10.0 11 K107 60.4 8.5 177 15.1 10 (nm) Determined by polyelectrolyte titration Determined by conductometric titration cDetermined by chemical analysis dFrom Andreasson et al. (2003)108 a b Figure 16. Adsorption isotherms for PAH and PVAm with different molecular masses, for pulp K75, measured in 5·10-4 M NaHCO3 -32- RESULTS & DISCUSSION Mechanical properties of the handsheets Hansheets were prepared from fibres treated with different amounts of polyelectrolyte, and the tensile indices of the sheets prepared from pulp K75 are shown in Figure 17 as a function of the adsorbed amount of polymer. The density of the networks was not affected by the adsorption of polymer, so that a direct comparison can be made between the different configurations. Each fibre/polyelectrolyte combination gave rise to an increase in the tensile index, with a tensile index increasing from about 80 Nm/g (unmodified fibres) to 120 Nm/g (at saturation adsorption, i.e. about 2 mg/g) in the case of PAH 56 kDa. Figure 17. (Left) Tensile index as a function of the adsorbed amount of polyelectrolyte (determined by nitrogen analysis), and (right) stress-strain curves of handsheets adsorbed with different polymers at saturation levels, both for pulp K75 The stress-strain curves of the samples, also displayed in Figure 17 for paper samples made from the K75 pulp, show that each polymer, when adsorbed at saturation level, gave rise to a significant improvement in the mechanical properties, although the improvement was slightly greater for the combination K75/PAH 56 kDa. Handsheets prepared from the K34 and K107 pulps showed lower relative improvements in the mechanical properties when polyelectrolytes were adsorbed. This is probably because fibres from the K75 pulp are the most swollen fibres110, and consequently more flexible, increasing the probability for a fibre to -33- RESULTS & DISCUSSION achieve intimate contact with another fibre to form a strong interfibre joint, which is even stronger after adsorption of strength additives such as PAH or PVAm. Interfibre joint strength To obtain a better understanding of the microscopic mechanisms involved in the dry strengthening of paper materials following the adsorption of polyamines, the strength of interfibre joints was evaluated for the K75 fibres, untreated and treated with PAH 56 kDa at saturation adsorption, which was the combination for which the largest relative increase (47%) in tensile index was observed. The results are presented in Table 2, and show that an increase of 18% in the interfibre joint strength was achieved when PAH was adsorbed. Table 2. Median and standard deviation of the interfibre joint strength in shearing mode Median Standard deviation (MPa) (MPa) Untreated 1.94 0.60 40 PAH 56 kDa 2.28 0.45 35 Fibres Number of specimens Another important feature of the interfibre joint strength is the distribution of the joint strength values. As can be seen in Figure 18, treatment with PAH shifted the interfibre joint strength distribution towards stronger joints, which means that the number of weak joints decreased and/or the number of strong joints increased after PAH adsorption onto the fibres. The variation in interfibre joint strength is generally high, but the strength variation found here was comparable to that reported in previous studies21,30, and a clear trend for the joint strength to increase as a result of the adsorption of PAH was evident. However, the difference in joint strength accounts for only 18% of the -34- RESULTS & DISCUSSION increase in median strength and cannot solely explain the 47% increase in strength at the macroscopic level. 0.35 Unmodified Modified Distribution unmodified Distribution modified 0.3 Density 0.25 0.2 0.15 0.1 0.05 0 0 5 σ overlap /MPa 10 15 Figure 18. Interfibre joint strength distribution and the fitted Weibull distribution for the unmodified and PAH-modified specimens respectively tested in shear Interfibre joints, contact area and number X-μCT was used to provide information on the microstructure of the fibrous network, and to investigate the influence of the adsorption of PAH on the number and area of the interfibre joints. 3D images of the fibrous network and of one isolated interfibre joint are shown in Figure 19. Using image analysis, the contact area of the interfibre joints as well as the number of interfibre joints in the fibrous network were estimated, and the results are presented in Table 3. The contact area and the number of joints increased by 14% and 26% respectively when PAH was added. The relatively large standard deviations show the large variety of interfibre joints formed during the consolidation process. -35- RESULTS & DISCUSSION Figure 19. 3D visualisations of (left) the untreated paper (sample width 0.8 mm) and (right) one interfibre joint from the same sample, with a contact area of 557 µm2. The fibre on top is semi-transparent in order to make the optically joined area visible Table 3. Mean contact area of interfibre joints and number of interfibre joints per mm fibre for untreated and PAH-treated samples, evaluated from X-µCT Mean contact area (μm2) Standard deviation (μm2) Number of joints / mm fibre Untreated 214 95 14.7 PAH 56 kDa 245 110 18.5 The strength enhancement of a fibrous network upon adsorption of a polyamine monolayer can therefore be explained at the microscopic level by an increase in the interfibre joint strength together with an increase in both the number of joints and the contact area. This is the first time, to the knowledge of the author, that these different factors have been evaluated separately and combined to explain the mechanism behind the dry strengthening of paper materials using chemical additives. The results demonstrate how important it is to separate the factors that are influenced by the additives in order to further improve their performance. In future investigations, it would be of interest to use X-ray -36- RESULTS & DISCUSSION tomography with a higher resolution in order the better to evaluate the molecular interactions across fibre surfaces. 4.1.2. Build-up of polyelectrolyte multilayers onto fibres for strong and stretchable paper materials Polyelectrolyte multilayer films of (PAH/HA) were assembled on the surface of bleached pulp fibres under two different salt conditions: no salt, and 10 mM NaCl. Handsheets were thereafter prepared from the unmodified and modified fibres. The mechanical properties (tensile index and strain at break) of these handsheets are presented in Figure 20. A major difference can be seen between systems without added salt, and systems in which the LbLs were assembled in 10 mM NaCl. Neither tensile index nor strain at break showed any significant change when the fibres were treated with (PAH/HA) in the absence of salt, regardless of the number of deposited layers. On the other hand, the mechanical properties increased after each added layer when the multilayers were deposited in a 10 mM NaCl environment, and both tensile index and strain at break increased by a factor of 3 after 5 bilayers of (PAH/HA) had been deposited. Figure 20. (Left) Tensile index and (right) strain at break of handsheets prepared from fibres coated with layers of (PAH/HA), assembled with 10 mM NaCl and without salt -37- RESULTS & DISCUSSION In order to explain how the multilayers influence the mechanical properties of the paper, the molecular organisation in the interfibre joints must be considered. In order to form a joint with properties providing both strength and stretchability to the paper, two features are crucial: (a) the polymers must be anchored at the surface of the fibre, and (b) the LbL film must have ductile properties, i.e. the film must have a yield stress/strain lower than that of the fibre wall, and the different layers in the formed film must interpenetrate to create a film with a thickness of 10-20 nm. Figure 21 depicts these aspects. A solid anchoring of the first polymeric layer to the fibre surface can be achieved both by charge interactions and by partial penetration of the polymer chains into the fibre wall. PAH with low molecular mass (15 kDa) was used as the anchoring layer. The radius of gyration of the polymer chains in a 10 mM NaCl medium was calculated to be 7 nm109, and the pore radius of a bleached kraft pulp fibre is of the order of 10 nm108. It is thus possible for these polymer chains to penetrate the outer layers of the fibre wall, ensuring a good anchoring of the layer from which the build-up of the structure is initiated. Fibre + LbL film + - + - Fibre - + + - - + - + - Fibre Figure 21. Schematic of an interfibre joint between two LbL-treated fibres with an enlargement of the fibre/LbL film interface, showing the anchoring of the polymer to the fibre surface, and the intermixing within the LbL film In order the better to understand the behaviour of the (PAH/HA) LbL films leading to the significant improvements at the paper level presented above, a more fundamental study of the build-up and adhesive properties of (PAH/HA) films was carried out. -38- RESULTS & DISCUSSION 4.2. Layer-by-Layer assemblies of (PAH/HA) (Papers III-V) 4.2.1. Build-up LbL films of (PAH/HA) were assembled onto silicon wafers to study the build-up of the multilayers. The influence of the number of layers deposited and of the addition of 10 mM NaCl were investigated. QCM-d was used to provide information on the film build-up and the results are shown in Figure 22, in terms of the frequency shift (which translates into the amount of material adsorbed at the solid/liquid interface) and the dissipation (which translates into the visco-elastic behaviour of the film). Without added salt, the LbL system showed a very low build-up with increasing number of layers, indicated by a more or less constant frequency shift regardless of the number of layers, and stable low dissipation values, indicating a rigid film with low viscous losses. Figure 22. QCM-d data for (PAH/HA) assemblies with and without added salt: (left) frequency shift and (right) dissipation as a function of the layer number In the presence of 10 mM NaCl during the assembly of the layers, the behaviour was totally different. The frequency shift decreased in a superlinear manner and an odd-even effect was observed, with a greater frequency shift and thus more adsorbed material when HA was in the -39- RESULTS & DISCUSSION outermost layer. The dissipation data showed an increased viscoelasticity of the film when the number of adsorbed layer increased, and the odd-even effect was also observed. The thickness of the (PAH/HA) films was estimated by DPI, and the results are shown in Figure 23. As observed in QCM-d, the addition of salt changed the build-up behaviour from a linear to a super-linear growth, and an odd/even effect was again observed in the system with added salt, where films with HA in the outermost layer were thicker than those with PAH in the outermost layer. Figure 23. Thickness of (PAH/HA) films as a function of the number of layers deposited for systems assembled with and without added salt, obtained by DPI measurements AFM images after drying of the structures formed upon LbL deposition are displayed in Figure 24. The images show a major difference in the build-up of the assemblies depending on whether or not salt was added. In the system without added salt, small structures were formed and their number increased with increasing number of layers deposited. On the other hand, with added salt, nano-sized structures were formed, and these structures increased in size and number as the number of deposited layers increased, until they coalesced to form even larger structures, -40- RESULTS & DISCUSSION giving rise to a rough film composed of structures reaching typically 30 nm in height. As was observed in DPI and QCM-d, an odd-even effect was observed, where structures with PAH in the outermost layer appeared to be flatter while structures with HA in the outermost layer showed a surface structure with greater height differences, and thus a greater roughness. 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 10 mM NaCl No added salt 1 Figure 24. AFM images of (PAH/HA) assemblies deposited on silicon wafers (top) without added salt and (bottom) with 10 mM NaCl. The numbers indicate the number of layers deposited. Images are 2x2 µm2, and the z-range is 30 nm Such structural changes can be explained by different salt conditions between the assembly (10 mM NaCl) and washing steps (pure water)111. Furthermore, Kovacevic et al.112 suggested that multilayers can be in a -41- RESULTS & DISCUSSION glassy state at low ionic strength and that an increase in ionic strength can lead to a liquid state where the molecules have more freedom to diffuse in the layer to reach an equilibrium conformation. The increased mobility of the polymer chains in the presence of salt promotes the interdiffusion within the LbL films, and it has been suggested that this is an explanation for the super-linear growth of some LbL systems71. The increase in film roughness shown in the AFM images gives rise to a larger surface area available for further deposition of polymer, and this, also known as the Island model74, can explain the super-linear growth observed in the QCM-d and DPI studies. 4.2.2. Adhesive properties Two different approaches were used to investigate the adhesive properties of (PAH/HA) films: AFM colloidal probe, providing information about the adhesive behaviour at the nano-level, and CAT, at the micro-level. AFM colloidal probe AFM force measurements were performed on LbL films of (PAH/HA) assembled in-situ in the AFM, simultaneously on the silicon wafer and on the colloidal probe. The force-displacement curves are shown in Figure 25 for systems tested under conditions where the following parameters were varied:  Number of layers deposited  Outermost layer  Dwell time  Presence of added salt -42- RESULTS & DISCUSSION No added salt Added salt a b c d e f Figure 25. Force-displacement curves during separation of the colloidal probe and flat surfaces. Graphs (a) and (b) show the influence of contact time for 5 layers in the systems, i.e. with PAH in the outermost layer, (a) without and (b) with salt. Graphs (c) and (d) show the influence of contact time for 6 layers in the systems, i.e. with HA in the outermost layer, (c) without and (d) with 10 mM NaCl. Graphs (e) and (f) show the influence of layer number at 10 seconds contact time in the systems (e) without and (f) with 10 mM NaCl -43- RESULTS & DISCUSSION As was observed in previously shown AFM, QCM-d and DPI results, the systems assembled with and without added salt behaved totally differently: an increase by one order of magnitude in the pull-off force was observed when 10 mM NaCl was added for 10 layers and a contact time of 10 s. The addition of salt promotes chain mobility, and the chain mobility promotes the interactions between the films in contact to an extent that is dependent on the contact time. Increasing the contact time between the two surfaces increased the pull-off forces, regardless of whether or not salt was added. This trend is expected since it is possible for the polymer chains to establish a greater contact with the other surface when the contact time is increased. When the number of layers deposited was increased, the pull-off force also increased. The pull-off forces were similar for the systems with PAH and the systems with HA in the outermost layer, but the separation distances were greater when HA was in the outermost layer, and separation forces were still detected at separation distances of about 8 μm for the system with added salt and a contact time of 10 s, showing that differences in the molecular weight of the polyelectrolyte in the outermost layer allow the molecular interactions to act over different distances, due to the unique properties of the supramolecular structures containing long HA chains and short PAH chains. The work of adhesion was obtained by calculating the area under the force-displacement curves, and the results are shown in Figure 26 as functions of contact time and number of layers. The work of adhesion values followed a similar trend to the pull-off forces, i.e. the work of adhesion increased when salt was added, when the contact time was increased, and when the number of layers deposited was increased. In the films with up to six layers, the work of adhesion appeared to correlate with the pull-off force of the interface. However, when subsequent layers were added, the increase in work of adhesion was related to the ability of the film to undergo a large displacement before total release, which suggests that there is a transition between thin films where the work of adhesion is governed mainly by the pull-off force and thicker films where -44- RESULTS & DISCUSSION the critical force approaches a limit where the microscopic displacement of the interfaces controls the work of adhesion. The work of adhesion for 9 and 10 layers formed in the presence of 10 mM NaCl was more than 20 times greater than the maximum work of adhesion reported earlier for sacrificial bonds in bone structures113. Furthermore, adhesive chain pulling interactions were recorded over a distance twice as long (> 8 μm) as the greatest distance measured for the chains to rupture with collagen and bone114. This shows the potential of the supramolecular structures formed in the (PAH/HA) LbL system. Figure 26. Work of adhesion as a function of (left) the contact time and (right) the number of layers, for (PAH/HA) systems with and without added salt The results suggest a mechanism involving two structural levels: an intimate contact is achieved by bringing the nanoscopic rough (PAH/HA) surfaces towards each other; molecular interdiffusion then leads to the formation of an interpenetrating network. The supramolecular structure of the (PAH/HA) assemblies, in combination with polymer chain disentanglement, leads to the high separation forces and large separation distances observed upon separation of the two surfaces. The suggested mechanism is illustrated in Figure 27. -45- RESULTS & DISCUSSION Figure 27. Schematic description of the approach and separation of two (PAH/HA)-coated surfaces CAT measurements Adhesion measurements using a CAT equipment were conducted to provide information on the adhesive properties of the (PAH/HA) system formed at 10 mM NaCl on a micro-scale. In order to promote the contact between the LbL-coated surfaces and thus optimize the adhesion, the treated surfaces were brought in contact in the wet state, and dried in contact before they were pulled apart. The force-displacement curves of are shown in Figure 28. A dramatic increase in the pull-off force was obtained for (PAH/HA) films from 1 to 3 bilayers, corresponding to a greatest film thickness of 12 nm. The separation distances also significantly increased. Images of the surfaces after the pull-off experiments are shown in Figure 28, and these images reveal a transition from a purely adhesive failure regime to a mixed adhesive and cohesive failure regime, indicated by the presence of small fragments of PDMS remaining on the silicon wafer for systems coated with 2 bilayers and more. This means that the adhesion was strong enough to cause cohesive failure of the surface of the PDMS probe. -46- RESULTS & DISCUSSION Reference (PAH/HA)1 (PAH/HA) (PAH/HA) 2 3 Figure 28. (Left) Force-displacement curves for the reference system and the system with 1, 2 and 3 bilayers of (PAH/HA) formed at 10 mM NaCl, and (right) images of the corresponding surfaces after the pull-off experiments (the scale bars are 250 µm) The area under the force-displacement curves was calculated, and the work of adhesion values were obtained for the different systems, and the results are plotted in Figure 29. As envisioned in the force-displacement curves, the work of adhesion increased by approximately one order of magnitude for each bilayer deposited. Figure 29. Work of adhesion as a function of the number of bilayers, formed in the presence of 10 mM NaCl, for model surfaces coated with (PAH/HA) and (PAH/PAA) -47- RESULTS & DISCUSSION To put these values into context, similar experiments were performed using a system commonly used in LbL studies, and known for its adhesive properties: (PAH/PAA). The results are plotted on the same graph and the increase in adhesion with an LbL coating of (PAH/PAA) is much lower than that of the (PAH/HA) system. This shows the unique adhesion features of HA films, and supports the trends seen in the AFM force measurements. 4.3. Nanocellulose aerogels as a platform for a variety of applications (Papers VI-VII) 4.3.1. Preparation and characterization of the aerogels Cellulose aerogels were prepared by freeze-drying a CNF gel. BTCA was used to crosslink the material and render it solvent-stable. The BTCA crosslinking leads to the incorporation of one carboxyl group per ester linkage to the cellulose hydroxyls, resulting in an increase in the charge (up to 2300 μeq/g for a 1:1 weight ratio of CNF and BTCA), which is beneficial in a LbL perspective. The material has a porosity close to 99% and the porous structure of the aerogel is shown in Figure 30. 100 µm 200 µm Figure 30. (Left) X-ray microtomography and (right) SEM images of the aerogel showing the porous structure of the material -48- RESULTS & DISCUSSION The aerogels were tested in compression in the wet state by compressing an aerogel in a step-by-step manner, with a relaxation between each compressive step. The results, displayed in Figure 31, show that the material undergoes shape-recovery up to very high compressive strains. Figure 31. Compressive stress-strain cycling of a wet aerogel up to 80% strain 4.3.2. LbL functionalization of the aerogels LbL was used to coat the aerogels with active materials and provide them with specific functionalities. The layering was achieved by a new way of forming LbLs, by forcing a polyelectrolyte suspension through the aerogel, ensuring complete coverage of the bulk of the material, achieved in only a few seconds, which is impossible using common dipping methods where diffusion processes in this type of low density material take a long time. To show that the method allows full coverage of the material in the bulk, confocal microscopy was used to analyse a cross section of an aerogel that was LbL-coated with 5 bilayers of a fluorescent polymer, Poly[2-(3-thienyl)ethyloxy-4-butylsulfonate], also termed ADS2000P. The confocal image, shown in Figure 32, reveals the structure of the aerogel with full coverage of the structure by the fluorescent polymer. -49- RESULTS & DISCUSSION 100 µm Figure 32. Confocal image reconstructed from slice images of the cross section of an aerogel LbL coated with 5 bilayers of ADS2000P The versatility of LbL makes it possible to use the aerogel as a platform for a variety of applications where active polymers and/or nanoparticles can be incorporated. To show the variety of possibilities offered by this template, LbL was used to incorporate a biopolymer (HA), a conductive polymer, poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS), and conductive nanoparticles, carbon nanotubes (CNT), as depicted in Figure 33 and Figure 34. O O OH OH C OH O O HO HO O O OH NH O n Figure 33. CNF aerogels, LbL-functionalized with (from left to right) HA, PEDOT:PSS and CNT -50- RESULTS & DISCUSSION Among the possible applications of these LbL-functionalized CNF aerogels, the preparation of energy-storage devices is particularly interesting, and the next section focuses on supercapacitors and batteries prepared in this manner. Figure 34. Illustration of the porous aerogel coated with CNT 4.3.3. Preparation of three dimensional and compressible energy storage devices LbL assembly of the electrodes and separator Supercapacitors were built inside the aerogel using the cellulose aerogel as a template by sequentially adsorbing (PEI/CNT)5 to assemble the first electrode, followed by (PEI/PAA)30 to assemble the separator and finally (PEI/CNT)5 again to form the second electrode. The electrodes were contacted immediately after they had been assembled using copper wires and silver paint. The contacts were protected from water by masking the contact region with paraffin wax. A photograph of an actual device is shown in Figure 35. -51- RESULTS & DISCUSSION Figure 35. Photograph of a full device Electron microscopy was used to show the sequential build-up of the full device (Electrode 1 - Separator - Electrode 2). The SEM images shown in Figure 36 show the build-up of CNT networks separated by a polymeric film on both sides of a cellulose sheet constituting the aerogel. The thickness of the CNT electrodes can be estimated to be 150 ±30 nm and that of the separator 1.70 ±0.46 μm. Figure 36. SEM images showing the sequential build-up of a full device at different magnifications: (left) Electrode 1, (middle) Electrode 1 – Separator, and (right) full device: Electrode 1 – Separator – Electrode 2. The scale bars are (top) 50 µm and (bottom) 2 µm -52- RESULTS & DISCUSSION Electrochemical characterization Cyclic voltammetry (CV) and galvanostatic cycling of the supercapacitors were carried out in 1 M Na2SO4 electrolyte, and the plots are shown in Figure 37. The CV graph shows square-shaped curves, characteristic of carbon supercapacitors and associated with the charging of a double layer at the interface between a carbon electrode and the electrolyte. The capacitive behaviour of the device is also reflected in the charge/discharge curves with a linear dependence of the potential with time. From these data, the specific capacitance of the devices was estimated to be 25 F/g, using a current corresponding to a 1 min discharge rate, and the devices showed stable cycling behaviour after 400 charge/discharge cycles. Although lower than some recently reported values, the capacitance is still comparable to that of other traditional carbon-based supercapacitors, showing that the concept of building 3D interdigitated energy-storage devices inside porous soft materials, using Current (A g-1) 0.2 0.1 0 5 mV s-1 10 mV s-1 25 mV s-1 50 mV s-1 -0.1 -0.2 0 0.2 0.4 0.6 Cell voltage (V) 30 20 10 25 μA 50 μA 0.8 0.6 0.4 0.2 0 0 200 400 600 800 Time (s) 0 0.8 Cell voltage (V) 0.3 Specific capacactiance (F g-1) LbL, is possible. 0 100 200 Cycle number 300 400 Figure 37. (Left) Cycling voltammograms at different scan rates and (right) galvanostatic cycling of the 3D supercapacitors To take full advantage of the flexibility of the porous aerogels used as a template, the devices were tested under compression. CVs were recorded for a device compressed to different levels, up to 75% of its initial thickness. The results shown in Figure 38 indicate that the device can operate under extreme conditions of compression without any significant -53- RESULTS & DISCUSSION change in its performance, and that the device recovers its initial shape 0.05 0 20 40 60 80 Compression (%) 0 0% 20% 40% 60% 75% -0.05 -0.1 0 25 μA 25 μA (after compression) 30 0.2 0.4 Cell voltage (V) 0.6 0.8 20 Cell voltage (V) Cuurent (A g-1) 0.1 c 1 0.8 0.6 0.4 0.2 0 Specific capacitance (F g-1) b 0.15 Normalized capcity after the pressure is released. 10 0 0 0.8 0.6 0.4 0.2 0 10600 100 10800 Time (s) 11000 200 300 Cycle number 400 Figure 38. Characterization of a full device under compression: (a) stepwise compression and release of the device (scale bar 0.5 cm), (b) cyclic voltammograms recorded under different compressions, and (c) relative losses in capacity after compressing the device When the device is compressed, the voids in the porous structure are compressed, and even if two LbL-coated cellulose walls come into contact, it is the external parts of the coating that come into contact, i.e. the same electrode, and this is the reason why no short circuits were observed even under extreme compression. Being able to operate a device under such conditions paves the way for many potential applications in flexible electronics. Extension to batteries Having demonstrated the concept of building 3D supercapacitors in aerogels, and taking advantage of the versatility of the LbL technique, hybrid batteries were built by incorporating copper hexacyanoferrate (CuHCF) nanoparticles in the cathode. CuHCF is an open framework material with the crystal structure of Prussian blue, and it can be used as -54- RESULTS & DISCUSSION cathode for ultra-long-lifetime, large-scale cheap batteries115. To first validate the concept, a single electrode was built, associating the negatively charged CuHCF with PEI, and incorporating CNT to provide conductivity, so that the electrode was composed of (PEI/CuHCF)3(PEI/CNT)3(PEI/CuHCF)3. The CV of the single electrode is shown in Figure 39c (inset). The two characteristic peaks corresponding to the faradaic processes that the cathode undergoes are clearly seen, showing that the CuHCF particles were successfully incorporated into the material. The single electrode was also shown to be stable over days of charge/discharge cycling, showing that the LbL coated electrodes are stable and that the incorporated CuHCF particles do not detach or dissolve in the electrolyte. A full hybrid battery was then built using this cathode and CNT as the active material in the anode. SEM images of the device as well as electrochemical characterisation are displayed in Figure 39. Small redox peaks can be seen in the CV, although these were not as pronounced as for the single electrode. One reason for this could be an asymmetry of the amount of active materials in the anode and cathode, giving an unmatched capacity of the electrodes. The loss in capacity upon cycling shows that the device design needs to be optimized. However, the general trends indicate that the battery operates as such and that the incorporation of CuHCF into the cathode was successful, opening up a range of possibilities where the properties of the device can be tailored by incorporating specific particles. -55- RESULTS & DISCUSSION d 10 1 0 Normalized capacity (C/C0) 0.8 -10 0.7 0.9 1.1 0.6 Potential vs. SHE (V) 8 0.4 4 0.2 0 Anode CNT Anode CNT -4 0.4 0.5 0.6 0.7 0.8 Cell voltage (V) 10 5 voltage (V) Cuurent (μA) 16 12 Cuurent (μA) c 0.6 10 μA 20 μA 10200 0 10800 Time (s) 0 0.9 1 0.8 50 100 150 Cycle number 200 Figure 39. SEM images of (a) the cross section of the hybrid battery and (b) zoom in the CuHCF cathode, and (c) cyclic voltammograms of the 3D battery with 5 and 10 bilayers of (PEI/CNT) in the anode, with the single electrode inset, recorded with a scan rate of 1 mV/s. (d) Galvanostatic charging and discharging of the CNT10 device and zoom in on one discharge cycle using 10 µA and 20 µA charging current and device cutoff voltage 0.5 V and 1 V -56- CONCLUSIONS 5. CONCLUSIONS The aim of the present thesis was to use physical adsorption of polyelectrolytes and nanoparticles to enhance the properties of existing cellulosic materials as well as to develop and functionalize a new class of nanocellulosic material. In this respect, the following results were achieved:  The strength of unbleached paper materials was enhanced by about 50% by adsorbing monolayers of polyamines onto the surface of the fibres in an amount of about 2 mg/g. The mechanisms governing this strength enhancement on the microscale were shown to be related to an increase in the interfibre joint strength together with an increase in the interfibre contact area and in the number of interfibre contacts.  The strength and strain at break of bleached paper materials were increased by a factor of three by building up multilayers of (PAH/HA) onto the surface of the fibres.  The fundamental study of (PAH/HA) LbL films revealed a transition from linear to super-linear growth upon addition of salt, as well as very high adhesive features on the nano- and micro-scales. The adhesion was shown to be stronger than the adhesion in bone structures.  Solvent-stable porous nanocellulosic soft aerogels were prepared. The material had a high specific area, porosity, and charge. A new method was introduced to functionalize this class of material using LbL in a fast and robust way by forcing polymeric or nanoparticle suspensions through the aerogel. -57- CONCLUSIONS  The CNF aerogels were used as a template to build 3D interdigitated energy storage devices by LbL deposition inside the aerogel. The devices operate under extreme conditions of compression and their shapes can be controlled. Recently developed high-performance cathode nanoparticles were incorporated to build a hybrid battery and demonstrate the versatility of the method, paving the way for the development of a new class of materials where high capacitance and flexibility can be achieved. -58- FUTURE WORK 6. FUTURE WORK As a continuation of the work presented in this thesis, a number of specific tracks can be suggested to provide an even more complete picture of cellulosic materials functionalized by polyelectrolyte adsorption.  Polyamines have been shown to be efficient dry-strength additives for paper materials, and the microscopic mechanisms of the dry-strengthening of lignocellulosic networks have been identified. At the molecular level, however, the mechanisms are still unclear and further investigations are necessary to identify the molecular interactions between polyamines and the polymers constituting the pulp fibres.  LbL assemblies of (PAH/HA) were characterized and its adhesive properties were studied on different length scales. An interesting future development of HA-containing LbL films would be to use them as an adhesive for biomedical applications such as tissue engineering. It would also be desirable to identify polyelectrolytes that could provide properties similar to those of HA, but with a greater availability and a lower cost.  CNF aerogels were developed and a simple, cheap and fast method to functionalize them has been introduced. This constitutes a totally new materials platform, where the aerogel can be used as a template for many potential applications. In addition to the energy-storage devices presented in this work, other initial activities in our research team have shown that antibacterial materials, as well as fire-retardant insulation materials can be prepared and it will probably also be possible to prepare materials for controlled drug release, to mention a few new routes, using a LbL approach similar to that demonstrated in the present thesis. Another future step regarding the development of -59- FUTURE WORK functional porous cellulose-based materials is to find efficient and cost-effective ways to produce CNF aerogels and provide them with specific functionalities on a larger scale.  The concept of assembling 3D supercapacitors and hybrid batteries in a porous template via LbL was demonstrated. Taking advantage of the versatility of the LbL method, other particles could be incorporated to build high-performance batteries. Scaling up the process would also be a next step of great interest. -60- ACKNOWLEDGEMENTS 7. ACKNOWLEDGEMENTS First and foremost, I would like to thank my supervisor Lars Wågberg for giving me the opportunity to work in his group, and for his endless optimism, enthusiasm and support regarding both the results and the life in the group. BiMaC Innovation is acknowledged for financial support. I would also like to thank all my co-authors for their assistance and cooperation in the different steps of the projects. My office-mates, Caroline, Simon, Rebecca, Louise and Maryam are acknowledged for valuable scientific discussions, for ensuring a very nice atmosphere at work, and last but not least for helping me to improve my Swedish! I wish to thank the administrative and technical staff at the Department of Fibre and Polymer Technology, Mia, Inga and Mona. Thank you for answering questions and for making administrative procedures so fast and simple. Special thanks to Mahiar for helping me bring my research and my thesis to a level that I would not otherwise have been able to reach! All the colleagues at FPT (particularly Andreas, Christopher, Dongfang, Emil, Erdem, Gustav, Johan, Louise, Mahiar, Maryam, Oruc, Petri, Raquel, Rebecca, Rosana, Veronica, Yujia) are also acknowledged for all the good times spent together, from the laboratory to the floorball court, not to mention the after-works and conference trips! Cora Thibeaut is acknowledged for her kind assistance in designing Figure 1. I would also like to address special thanks to my French colleagues and friends Myriam and Thomas for sharing their Swedish experience as PhD students. Thank you for your support and for all the discussions on -61- ACKNOWLEDGEMENTS politics, sports, and science! And thank you Myriam for reviewing the thesis and being my critic!!! Anthony, Paul, Pia, Rey and Zhenya (aka the NT team) are also acknowledged for all the fun times and great dinners. Juan Pablo Parra Martinez (…to make it short!) is thanked for numerous training and laughing sessions, not to mention the stress-releasing sugarand fat-based food occasions and other beer collaborations! Pampi: thank you so much for all the good times spent together, at the gym or around a glass of wine! Franz L., Ludwig v. B., Frédéric C., Sergei R., Giacomo P. (non-exhaustive liszt!) are acknowledged for providing inspiration, strength and motivation throughout these years. Finally, I would like to thank my closest friends from France, Benoît, Florian, Frédéric, Julien, Karine, Margot, Michaël, Nathalie, Noémie, for your support, for all the nice times spent together, and for being so close after so many years and despite the distance! Parvenu au terme de cette thèse, il me reste à remercier ceux qui m’ont permis d’en arriver là! Maman, Papa, Jordan, Ophélie, Valentine, merci pour tout! -62- REFERENCES 8. REFERENCES 1. Hon, D. N.-S. Cellulose: a random walk along its historical path. Cellulose 1, 1–25 (1994). 2. Payen, A. Mémoire sur la composition du tissu propre des plantes et du ligneux. Comptes Rendus 7, 1052–1056 (1838). 3. Bristow, J. A. & Kolseth, P. Paper Structure and Properties. (Dekker, 1986). 4. Wågberg, L. & Annergren, G. Physico-chemical characterisation of papermaking fibres. Proc. 11th Fundam. Res. Symp. Cambridge 1–82 (1997). 5. Biermann, C. J. Essentials of pulping and papermaking. (Academic press, 1993). 6. Ek, M., Gellerstedt, G. & Henriksson, G. Pulping chemistry and technology (vol. 2). (2009). 7. Turbak, A. F., Snyder, F. W. & Sandberg, K. R. Microfibrillated cellulose, a new cellulose product: Properties, uses, and commercial potential. J. Appl. Polym. Sci. Appl. Polym. Symp. 37, 815 (1983). 8. Henriksson, M., Henriksson, G., Berglund, L. & Lindström, T. An environmentally friendly method for enzyme-assisted preparation of microfibrillated cellulose (MFC) nanofibers. Eur. Polym. J. 43, 3434– 3441 (2007). 9. Wågberg, L. et al. The build-up of polyelectrolyte multilayers of microfibrillated cellulose and cationic polyelectrolytes. Langmuir 24, 784–795 (2008). 10. Saito, T., Kimura, S., Nishiyama, Y. & Isogai, A. Cellulose nanofibers prepared by TEMPO-mediated oxidation of native cellulose. Biomacromolecules 8, 2485–91 (2007). -63- REFERENCES 11. Eichhorn, S. J. et al. Review: current international research into cellulose nanofibres and nanocomposites. J. Mater. Sci. 45, 1–33 (2009). 12. Fall, A. B., Lindström, S. B., Sundman, O., Ödberg, L. & Wågberg, L. Colloidal stability of aqueous nanofibrillated cellulose dispersions. Langmuir 27, 11332–8 (2011). 13. Ankerfors, M. Microfibrillated cellulose: Energy-efficient preparation techniques and applications in paper. (2015). 14. Lavoine, N., Desloges, I., Dufresne, A. & Bras, J. Microfibrillated cellulose - its barrier properties and applications in cellulosic materials: a review. Carbohydr. Polym. 90, 735–64 (2012). 15. Olsson, R. T. et al. Making flexible magnetic aerogels and stiff magnetic nanopaper using cellulose nanofibrils as templates. Nat. Nanotechnol. 5, 584–588 (2010). 16. Zheng, G. et al. Nanostructured paper for flexible energy and electronic devices. MRS Bull. 38, 320–325 (2013). 17. 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. 13th Fundam. Res. Symp. 457–562 (2005). 18. Davison, R. W. The weak link in paper dry strength. Tappi 55, 567 (1972). 19. Davison, R. W. Theory of dry strength development. Dry Strength Addit. TAPPI Press. Atlanta 1–31 (1980). 20. Page, D. H. A theory for the tensile strength of paper. Tappi 52, 674– 681 (1969). 21. Mohlin, U. B. Cellulose fibre bonding. Part 3. The effect of beating and drying on interfibre bonding. Sven. Papperstidn 78, 338–341 (1975). -64- REFERENCES 22. Dasgupta, S. Mechanism of paper tensile-strength development due to pulp beating. Tappi J 77, 158–166 (1994). 23. Wågberg, L. & Björklund, M. Adsorption of cationic potato starch on cellulosic fibres. Nord. Pulp Pap. Res. J. 8, 399–404 (1993). 24. Lindström, T. & Söremark, C. Adsorption of cationic polyacrylamides on cellulose. J. Colloid Interface Sci. 55, 305–312 (1976). 25. Miao, C., Leduc, M. & Pelton, R. The influence of polyvinylamine microgels on paper strength. J. Pulp Pap. Sci 34, 69–75 (2008). 26. Mocchiutti, P., Galvan, M. V, Inalbon, M. C. & Zanuttini, M. A. Improvement of paper properties of recycled unbleached softwood kraft pulps by poly(allylamine hydrochloride). 6, 570–583 (2011). 27. Pelton, R. On the design of polymers for increased paper dry strength: A review. Appita J 57, 181–190 (2004). 28. Rathi, M. S. & Biermann, C. J. Application of polyallylamine as a dry strength agent for paper. Tappi J 83, 62 (2000). 29. Mayhood, C. H., Kallmes, O. & Cauley, M. The mechanical properties of paper. Part II: Measured shear strength of individual fiber to fiber contacts. Tappi J 45, 69–73 (1962). 30. Schniewind, A. P., Nemeth, L. . & Brink, D. L. Fiber and Pulp Properties. I. Shear Strength of Single-Fiber Crossings. Tappi J 47, 244– 248 (1964). 31. Button, A. F. Fiber-fiber bond strength - A study of a linear elastic model structure. (1979). 32. Schmied, F. J., Teichert, C., Kappel, L., Hirn, U. & Schennach, R. Joint strength measurements of individual fiber-fiber bonds: an atomic force microscopy based method. Rev. Sci. Instrum. 83, 073902 (2012). 33. Magnusson, M. S., Fischer, W. J., Östlund, S. & Hirn, U. Interfibre joint strength under peeling, shearing and tearing types of loading. in -65- REFERENCES Adv. Pulp Pap. Res. Cambridge 2013 Trans. 15th Fundam. Res. 103–124 (2013). 34. Magnusson, M. S., Zhang, X. & Östlund, S. Experimental Evaluation of the Interfibre Joint Strength of Papermaking Fibres in Terms of Manufacturing Parameters and in Two Different Loading Directions. Exp. Mech. 53, 1621–1634 (2013). 35. Jin, H., Nishiyama, Y., Wada, M. & Kuga, S. Nanofibrillar cellulose aerogels. Colloids Surfaces A Physicochem. Eng. Asp. 240, 63–67 (2004). 36. Pääkkö, M. et al. Long and entangled native cellulose I nanofibers allow flexible aerogels and hierarchically porous templates for functionalities. Soft Matter 4, 2492 (2008). 37. Sehaqui, H., Zhou, Q. & Berglund, L. High-porosity aerogels of high specific surface area prepared from nanofibrillated cellulose (NFC). Compos. Sci. Technol. 71, 1593–1599 (2011). 38. Zhang, W., Zhang, Y., Lu, C. & Deng, Y. Aerogels from crosslinked cellulose nano/micro-fibrils and their fast shape recovery property in water. J. Mater. Chem. 22, 11642 (2012). 39. Wågberg, L. Polyelectrolyte adsorption onto cellulose fibres-A review. Nord. Pulp Pap. Res. J. 15, 586–597 (2000). 40. Wågberg, L. & Hägglund, R. Kinetics of polyelectrolyte adsorption on cellulosic fibers. Langmuir 17, 1096–1103 (2001). 41. Wågberg, L., Ödberg, L. & Lindström, T. Kinetics of Adsorption and Ion-Exchange Reactions during Adsorption of Cationic Polyelectrolytes onto Cellulosic Fibers. 31, 119–124 (1988). 42. Van de Steeg, H. G. M., Cohen Stuart, M. A., De Keizer, A. & Bijsterbosch, B. H. Polyelectrolyte adsorption: a subtle balance of forces. Langmuir 8, 2538–2546 (1992). -66- REFERENCES 43. Winter, L., Wågberg, L., Ödberg, L. & Lindström, T. Polyelectrolytes adsorbed on the surface of cellulosic materials. J. Colloid Interface Sci. 111, 537–543 (1986). 44. Bates, R., Beijer, P. & Podd, B. in Papermak. Sci. Technol. chapter 13 (Papermaking Chemistry Eds. Gullichen J Neimo L, P. H.) 4, 288–301 (1999). 45. Gimåker, M., Horvath, A. & Wågberg, L. Influence of polymeric additives on short-time creep of paper. Nord Pulp Pap Res J 22, 217 (2007). 46. Lindström, T. & Söderberg, G. On the Mechanism of Sizing with Alkylketene Dimers - Part I. Studies on the Amount of Alkylketene Dimer Required for Sizing of Different Pulps. Nord. Pulp Pap. Res. J. 1, 26–33 (1986). 47. Moeller, H. W. Cationic starch as a wet-end strength additive. Tappi 49, 211 (1966). 48. Ankerfors, C., Lingström, R., Wågberg, L. & Ödberg, L. A comparison of polyelectrolyte complexes and multilayers: Their adsorption behaviour and use for enhancing tensile strength of paper. Nord. Pulp Pap. Res. J. 24, 77–86 (2009). 49. Decher, G. & Schlenoff, J. B. Multilayer Thin Films. (Wiley-VCH, 2012). 50. Iler, R. K. Multilayers of colloidal particles. J. Colloid Interface Sci. 21, 569–594 (1966). 51. Decher, G. Fuzzy Nanoassemblies: Toward Layered Polymeric Multicomposites. Science. 277, 1232–1237 (1997). 52. Decher, G. & Hong, J. D. Buildup of Ultrathin Multilayer Films by a Self-Assembly Process .2. Consecutive Adsorption of Anionic and Cationic Bipolar Amphiphiles and Polyelectrolytes on Charged Surfaces. Berichte Der Bunsen-Gesellschaft-Physical Chem. Chem. Phys. 95, 1430–1434 (1991). -67- REFERENCES 53. Izquierdo, A., Ono, S. S., Voegel, J.-C., Schaaf, P. & Decher, G. Dipping versus spraying: exploring the deposition conditions for speeding up layer-by-layer assembly. Langmuir 21, 7558–67 (2005). 54. Schlenoff, J. B., Dubas, S. T. & Farhat, T. Sprayed Polyelectrolyte Multilayers. Langmuir 16, 9968–9969 (2000). 55. DeLongchamp, D. M. & Hammond, P. T. Highly ion conductive poly(ethylene oxide)-based solid polymer electrolytes from hydrogen bonding layer-by-layer assembly. Langmuir 20, 5403–5411 (2004). 56. Yang, Y.-H., Haile, M., Park, Y. T., Malek, F. & Grunlan, J. C. Super Gas Barrier of All-Polymer Multilayer Thin Films. Macromolecules 44, 1450–1459 (2011). 57. Priolo, M., Gamboa, D., Holder, K. M. & Grunlan, J. C. Super gas barrier of transparent polymer-clay multilayer ultrathin films. Nano Lett. 10, 4970–4 (2010). 58. Tang, Z. Y., Kotov, N. A., Magonov, S. & Ozturk, B. Nanostructured artificial nacre. Nat. Mater. 2, 413–U8 (2003). 59. Zhai, L., Cebeci, F. C., Cohen, R. E. & Rubner, M. F. Stable superhydrophobic coatings from polyelectrolyte multilayers. Nano Lett. 4, 1349–1353 (2004). 60. Mamedov, A. et al. Molecular design of strong single-wall carbon nanotube/polyelectrolyte multilayer composites. Nat. Mater. 1, 190–4 (2002). 61. Szarpak, A. et al. Designing hyaluronic acid-based layer-by-layer capsules as a carrier for intracellular drug delivery. Biomacromolecules 11, 713–20 (2010). 62. Smith, R. C., Riollano, M., Leung, A. & Hammond, P. T. Layer-byLayer Platform Technology for Small-Molecule Delivery. Angew. Chemie 121, 9136–9139 (2009). -68- REFERENCES 63. Hammond, P. T. Engineering Materials Layer-by-Layer : Assembly. 57, (2011). 64. Hammond, P. T. Building biomedical materials layer-by-layer. Mater. Today 15, 196–206 (2012). 65. Boudou, T., Crouzier, T., Ren, K., Blin, G. & Picart, C. Multiple functionalities of polyelectrolyte multilayer films: new biomedical applications. Adv. Mater. 22, 441–467 (2010). 66. Hammond, P. T. Form and function in multilayer assembly: New applications at the nanoscale. Adv. Mater. 16, 1271–1293 (2004). 67. Yoo, P. J. et al. Spontaneous assembly of viruses on multilayered polymer surfaces. Nat. Mater. 5, 234–240 (2006). 68. Johnston, A. P. R., Cortez, C., Angelatos, A. S. & Caruso, F. Layer-bylayer engineered capsules and their applications. Curr. Opin. Colloid Interface Sci. 11, 203–209 (2006). 69. Wang, Y., Angelatos, A. S. & Caruso, F. Template Synthesis of Nanostructured Materials via Layer-by-Layer Assembly. Chem. Mater. 20, 848–858 (2008). 70. Dubas, S. T. & Schlenoff, J. B. Factors controlling the growth of polyelectrolyte multilayers. Macromolecules 32, 8153–8160 (1999). 71. Picart, C. et al. Molecular basis for the explanation of the exponential growth of polyelectrolyte multilayers. Proc. Natl. Acad. Sci. U. S. A. 99, 12531–12535 (2002). 72. Porcel, C. et al. From exponential to linear growth in polyelectrolyte multilayers. Langmuir 22, 4376–4383 (2006). 73. Porcel, C. et al. Influence of the polyelectrolyte molecular weight on exponentially growing multilayer films in the linear regime. Langmuir 23, 1898–1904 (2007). -69- REFERENCES 74. Haynie, D. T., Cho, E. H. & Waduge, P. ‘In and Out Diffusion’ Hypothesis of Exponential Multilayer Film Buildup Revisited. Langmuir 27, 5700–5704 (2011). 75. Agarwal, M., Lvov, Y. & Varahramyan, K. Conductive wood microfibres for smart paper through layer-by-layer nanocoating. Nanotechnology 17, 5319–5325 (2006). 76. Renneckar, S. & Zhou, Y. Nanoscale coatings on wood: polyelectrolyte adsorption and layer-by-layer assembled film formation. ACS Appl. Mater. Interfaces 1, 559–66 (2009). 77. Eriksson, M., Notley, S. M. & Wågberg, L. The influence on paper strength properties when building multilayers of weak polyelectrolytes onto wood fibres. J. Colloid Interface Sci. 292, 38–45 (2005). 78. Lundström, L., Lindgren, J., Svensson-Rundlöf, E., Sennerfors, T. & Wågberg, L. The adsorption of polyelectrolyte multilayers (PEM) of starch on mechanical pulps for improved mechanical paper properties. Nord. Pulp Pap. Res. J. 24, 459–468 (2009). 79. Eriksson, M., Torgnysdotter, A. & Wågberg, L. Surface Modification of Wood Fibers Using the Polyelectrolyte Multilayer Technique: Effects on Fiber Joint and Paper Strength Properties. Ind. Eng. Chem. Res. 45, 5279–5286 (2006). 80. Ducker, W. A., Senden, T. J. & Pashley, R. M. Direct Measurement of Colloidal Forces Using an Atomic Force Microscope. Nature 353, 239– 241 (1991). 81. Johansson, E., Blomberg, E. & Wågberg, L. Adhesive interaction between polyelectrolyte multilayers of polyallylamine hydrochloride and polyacrylic acid studied using atomic force microscopy and surface force apparatus. Langmuir 25, 2887–2894 (2009). 82. Bosio, V., Dubreuil, F., Bogdanovic, G. & Fery, A. Interactions between silica surfaces coated by polyelectrolyte multilayers in aqueous environment: Comparison between precursor and -70- REFERENCES multilayer regime. Colloids Surfaces A Physicochem. Eng. Asp. 243, 147– 155 (2004). 83. Chaudhury, M. K. & Whitesides, G. M. Direct measurement of interfacial interactions between semispherical lenses and flat sheets of poly(dimethylsiloxane) and their chemical derivatives. Langmuir 7, 1013–1025 (1991). 84. Johnson, K. L., Kendall, K. & Roberts, A. D. Surface Energy and the Contact of Elastic Solids. Proc. R. Soc. A Math. Phys. Eng. Sci. 324, 301– 313 (1971). 85. Nolte, A. J., Chung, J. Y., Walker, M. L. & Stafford, C. M. In situ adhesion measurements utilizing layer-by-layer functionalized surfaces. ACS Appl. Mater. Interfaces 1, 373–80 (2009). 86. Fraser, J. R., Laurent, T. C. & Laurent, U. B. Hyaluronan: its nature, distribution, functions and turnover. J. Intern. Med. 242, 27–33 (1997). 87. Laurent, T. C. & Fraser, J. R. Hyaluronan. FASEB J. 6, 2397–2404 (1992). 88. Kogan, G., Soltés, L., Stern, R. & Gemeiner, P. Hyaluronic acid: a natural biopolymer with a broad range of biomedical and industrial applications. Biotechnol. Lett. 29, 17–25 (2007). 89. Monteiro, I. P., Shukla, A., Marques, A. P., Reis, R. L. & Hammond, P. T. Spray-assisted layer-by-layer assembly on hyaluronic acid scaffolds for skin tissue engineering. J. Biomed. Mater. Res. A 1–11 (2014). doi:10.1002/jbm.a.35178 90. Gribova, V., Auzely-Velty, R. & Picart, C. Polyelectrolyte Multilayer Assemblies on Materials Surfaces: From Cell Adhesion to Tissue Engineering. Chem. Mater. 24, 854–869 (2012). 91. Winter, M. & Brodd, R. J. What are batteries, fuel cells, and supercapacitors? Chem. Rev. 104, 4245–69 (2004). -71- REFERENCES 92. Abruna, H. D., Kiya, Y. & Henderson, J. C. Batteries and electrochemical capacitors. Phys. Today 61, 43–47 (2008). 93. Lee, S. W. et al. High-power lithium batteries from functionalized carbon-nanotube electrodes. Nat. Nanotechnol. 5, 531–7 (2010). 94. Niu, C., Sichel, E. K., Hoch, R., Moy, D. & Tennent, H. High power electrochemical capacitors based on carbon nanotube electrodes. Appl. Phys. Lett. 70, 1480 (1997). 95. Dresselhaus, M. S., Dresselhaus, G. & Avouris, P. Carbon Nanotubes Synthesis, Structure, Properties, and Applications. (Springer, 2001). 96. Baughman, R. H., Zakhidov, A. & de Heer, W. Carbon nanotubes--the route toward applications. Science 297, 787–92 (2002). 97. Kim, S. Y. et al. Rapid fabrication of thick spray-layer-by-layer carbon nanotube electrodes for high power and energy devices. Energy Environ. Sci. 6, 888 (2013). 98. Xiang, Y., Lu, S. & Jiang, S. P. Layer-by-layer self-assembly in the development of electrochemical energy conversion and storage devices from fuel cells to supercapacitors. Chem. Soc. Rev. 41, 7291– 321 (2012). 99. Notten, P. H. L., Roozeboom, F., Niessen, R. a. H. & Baggetto, L. 3-D Integrated All-Solid-State Rechargeable Batteries. Adv. Mater. 19, 4564–4567 (2007). 100. Wågberg, L., Ödberg, L. & Glad-Nordmark, G. Charge determination of porous substrates by poly electrolyte adsorption. Nord Pulp Pap Res J 2, 71–76 (1989). 101. Terayama, H. Method of colloid titration (a new titration between polymer ions). J Polym Sci 8, 243–253 (1952). 102. Mocchiutti, P. & Zanuttini, M. A. Key considerations in the determination of polyelectrolyte concentration by the colloidal titration method. BioResources 2, (2007). -72- REFERENCES 103. Swann, M. J., Peel, L. L., Carrington, S. & Freeman, N. J. Dualpolarization interferometry: an analytical technique to measure changes in protein structure in real time, to determine the stoichiometry of binding events, and to differentiate between specific and nonspecific interactions. Anal. Biochem. 329, 190–8 (2004). 104. Wernersson, E. L. G., Borodulina, S., Kulachenko, A. & Borgefors, G. Characterisations of fibre networks in paper using micro computed tomography images. Nord. Pulp Pap. Res. J. 29, 468–475 (2014). 105. Escorihuela, J. et al. Dual-Polarization Interferometry : A Novel Technique To Light up the Nanomolecular World. Chem. Rev. 265–294 (2015). 106. Marx, K. A. Quartz crystal microbalance: A useful tool for studying thin polymer films and complex biomolecular systems at the solution-surface interface. Biomacromolecules 4, 1099–1120 (2003). 107. Sauerbrey, G. The use of quartz oscillators for weighing thin layers and for microweighing. Zeitschrift fuer Phys. 155, 206–222 (1959). 108. Andreasson, B., Forsström, J. & Wågberg, L. The porous structure of pulp fibres with different yields and its influence on paper strength. Cellulose 10, 111–123 (2003). 109. Gimåker, M. & Wågberg, L. Adsorption of polyallylamine to lignocellulosic fibres: effect of adsorption conditions on localisation of adsorbed polyelectrolyte and mechanical properties of resulting paper sheets. Cellulose 16, 87–101 (2008). 110. Scallan, A. M. & Tigerström, A. C. Swelling and elasticity of the cell walls of pulp fibres. J Pulp Pap Sci 18, J188–J193 (1992). 111. Fery, A., Schöler, B., Cassagneau, T. & Caruso, F. Nanoporous Thin Films Formed by Salt-Induced Structural Changes in Multilayers of Poly(acrylic acid) and Poly(allylamine). Langmuir 17, 3779–3783 (2001). 112. Kovacevic, D., Burgh, S. Van Der, Keizer, A. De & Stuart, M. A. C. Kinetics of Formation and Dissolution of Weak Polyelectrolyte -73- REFERENCES Multilayers : Role of Salt and Free Polyions. Langmuir 18, 5607–5612 (2002). 113. Thompson, J. B. et al. Bone indentation recovery time correlates with bond reforming time. Nature 414, 773–776 (2001). 114. Fantner, G. E. et al. Sacrificial bonds and hidden length dissipate energy as mineralized fibrils separate during bone fracture. Nat. Mater. 4, 612–6 (2005). 115. Wessells, C. D., Huggins, R. a & Cui, Y. Copper hexacyanoferrate battery electrodes with long cycle life and high power. Nat. Commun. 2, 550 (2011). -74-