Modification of Paper Into Conductive Substrate for Electronic Functions

Faculty of Technology and Science Chemical Engineering

Elson Montibon

Modification of Paper into Conductive Substrate for Electronic Functions

Deposition, Characterization and Demonstration

DISSERTATION Karlstad University Studies 2010:28 Elson Montibon

Modification of Paper into Conductive Substrate for Electronic Functions

Deposition, Characterization and Demonstration

Karlstad University Studies 2010:28 Elson Montibon. Modification of Paper into Conductive Substrate for Electronic Functions - Deposition, Characterization and Demonstration

Dissertation

Karlstad University Studies 2010:28 ISSN 1403-8099 ISBN 978-91-7063-361-4

Distribution: Karlstad University Faculty of Technology and Science Chemical Engineering S-651 88 Karlstad Sweden +45 54 700 100 www.kau.se

Print: Universitetstryckeriet, Karlstad 2011

“Delight thyself also in the LORD: and he shall give thee the desires of thine heart.” Psalm 37:4

Abstract

The thesis investigates the modification of paper into an ion- and electron-conductive material, and as a renewable material for electronic devices. The study stretches from investigating the interaction between cellulosic materials and a conducting polymer to demonstrating the performance of the conductive paper by printing an electronic structure on the surface of the conductive paper. Conducting materials such as conducting polymer, ionic liquids, and multi-wall carbon nanotubes were deposited into the fiber networks.

Papers displaying electroconductive behavior were produced via dip coating and rod coating, and characterized. The Scanning Electron Microscopy (SEM) / Energy Dispersive Spectroscopy (EDS) images showed that the conducting polymer was deposited in the fiber and in fiber-fiber contact areas. The X-ray Photoelectron Spectroscopy (XPS) analysis of dip-coated paper samples showed PEDOT enrichment on the surface. The degree of washing of the dip-coated paper with dilute acid did not significantly affect the PEDOT enrichment on the surface. When the base paper was coated with a PEDOT:PSS blend, the conductivity increased by many orders of magnitude. The base papers normally had bulk conductivities in the region of 10 -12 S/cm, and after modification, the conductivity reached between 10 -3 and 10 -1 S/cm. The effects of fiber beating and paper formation were also investigated. The presence of organic solvents N-methyl-2-pyrrolidone (NMP) and dimethyl sulfoxide (DMSO) in the PEDOT:PSS dispersion enhanced the conductivity of the coated sheets, while sorbitol and isopropanol had no enhancement effect. This is due to the conformational changes of the PEDOT molecule in the presence of NMP and DMSO and to their plasticizing effect. The conductivity of paper did not increase until the line load reached 174 kN/m during calendering which is indicative of PEDOT:PSS being suspended in the fiber network. The effect of the presence of pigments such as titanium dioxide (TiO 2) and multi-wall carbon nanotubes (MWCNT) was also discussed, and also their effects on other properties such as tensile strength, surface property and wetting were discussed.

Ionic paper was produced by depositing an ionic liquid into the commercial base paper. The effect of humidity on the ionic conductivity was determined. Even at a humidity as low as 25% RH, the ionic paper demonstrated good ionic conductivity. The temperature- dependence of the ionic conductivity followed the simple Arrhenius equation, which can be attributed to the hopping of carrier ions. The ionic paper was surface-sized in order to reduce the roughness and improve its printability. The bulk resistance increased with increasing surface sizing. The electrochemical performance of the ionic paper was confirmed by printing PEDOT:PSS on the surface. There was change in color of the polymer when a voltage was applied. It was demonstrated that the ionic paper is a good ionic conductor that can be used as a component for a more compact electronic device construction.

Conductive paper has a great potential to be a flexible platform on which an electronic structure can be constructed. The conduction process in the modified paper is due to the density of charge carriers (ions and electrons), and their short range mobility in the material. The charge carrying is believed to be heterogeneous, involving many species as the paper material is chemically heterogeneous. Paper is a versatile material. Its bulk and surface properties can easily be tuned to a desired level by modifying the fibers, adding chemical additives, and surface treatment. Furthermore, the abundance of cellulosic material on this planet makes conductive paper a renewable and sustainable component for electronic devices.

List of Papers

I Characterization of Poly(3,4.ethylenedioxythiophene):Poly(4-styrenesulfonate) (PEDOT:PSS) on Cellulosic Materials Elson Montibon, Lars Järnström, and Magnus Lestelius Cellulose 2009 , 16, 807-815

II Electroconductive Paper Prepared by Coating with Blends of Poly(3,4- ethylenedioxythiophene)/Poly(4-styrenesulfonate) (PEDOT:PSS) and Organic Solvents Elson Montibon, Magnus Lestelius, and Lars Järnström Journal of Applied Polymer Science 2010 , 117, 3524-3532

III. Electroconductive Paper – a study of conductivity achieved through polymer deposition and papermaking properties Elson Montibon, Karen T. Love, Magnus Lestelius, and Lars Järnström Nordic Pulp and Paper Research Journal 2010 , 25 (4), 473-480

IV Conductivity of paper containing poly(3,4-ethylenedioxythiophene)/poly(4- styrenesulfonate) and multi-wall carbon nanotubes Elson Montibon, Magnus Lestelius, and Lars Järnström Submitted for publication in Journal of Applied Polymer Science

V Porous and Flexible Ionic Paper as a platform for compact electronic devices Elson Montibon, Magnus Lestelius and Lars Järnström Manuscript to be submitted for publication

Reprints of Papers I, II, and III have been made with permission from the publishers

Elson Montibon’s contribution to the papers Elson Montibon performed all the experimental work with the exception of XPS in Paper I; BET area in Papers I and II; SEM/EDS in Papers II, III and IV; and Hg-porosimetry in Paper III. Elson is the main author of all five papers included in this thesis.

iii

Related presentations by the same author

Montibon E., Järnström L., and Lestelius M., “On Investigation of Poly(3,4.ethylenedioxythiophene)/Poly(4-styrenesulfonate) (PEDOT:PSS) Adsorption on Cellulosic Materials”, Poster Presentation, Conference Proceeding of the Nordic Polymer Days 2008, Stockholm, Sweden, June 11-13, 2008.

Montibon E., Järnström L., and Lestelius M., “Characterization of Poly(3,4- ethylenedioxythiophene)/Poly(4-styrenesulfonate) (PEDOT:PSS) Adsorption on Cellulosic Materials”, Oral Presentation, Conference Proceeding of 13th IACIS International Conference on Surface and Colloid Science, and the 83rd ACS Colloid & Surface Science Symposium, Columbia University, New York City, USA. June 14-19, 2009.

Montibon E., Lestelius M., and Järnström L. “Preparation of Electroconductive Paper by Coating Blends of Poly(3,4-ethylenedioxythiophene)/Poly(4-styrenesulfonate) (PEDOT:PSS) and Organic Solvents,” Oral Presentation, Proceedings of the TAPPI Coating and Graphic Arts Conference (PaperCon ‘09), St. Louis, Missouri, USA, May 31 – June 03, 2009.

Lestelius M., Montibon E., and Järnström L. “Preparation of Electroconductive Paper by Deposition of Conducting Polymer,” Oral Presentation, Innovative Packaging Symposium, PTS, Munich, Germany, June 05-06, 2010.

Montibon E., Lestelius M., and Järnström L. “Modification of Paper into Electroconductive Material by Deposition of Conducting Polymer Blends”, Oral Presentation, Proceedings of the 3 rd International Symposium of Emerging Technologies in Pulping and Papermaking (ISETPP ), SCUT, Guangzhou, China, November 8-10, 2010.

List of Symbols and Abbreviations

Abbreviations

AU - absorbance unit BSE - Backscattered electron CD - cross machine direction CNT - carbon nanotubes DMSO - Dimethyl sulfoxide EDS - Energy Dispersive X-ray Spectroscopy ESCA - Electron Spectroscopy for Chemical Analysis FFT - fast Fourier transform HOMO - highest occupied molecular orbital IS - impedance spectroscopy LUMO - lowest unoccupied molecular orbital IL - ionic liquid MCC - microcrystalline cellulose MD - machine direction MWCNT - multi-wall carbon nanotubes NMP - N-methyl-2-pyrrolidone PEDOT:PSS - poly(3,4-ethylenedioxythiophene)/poly(4-styrene sulfonate) PSI - phase shift interferometer RBA - relative bonded area RTIL - room temperature ionic liquid SWCNT - single wall carbon nanotubes VSI - vertical scanning interferometer XPS - X-ray Photoelectron spectroscopy

Symbols

a - cross-section area (m 2) A - measured absorbance (AU) E - applied electric field (N/C) E1 - heat of adsorption of layer 1 (J) EL - heat of adsorption of higher layers (J) Ea - activation energy (eV or J) Eb - binding energy (eV or J) Ek - kinetic energy (eV or J) e - elementary charge (1.602176487 x 10 -19 C) ei - electronic charge of ion of species i (C) ε - molar absorptivity (m 2/mol) εrs - relative permittivity of the dispersion medium ε0 - permittivity of vacuum (F/m or C² N-1 m-2) v

η - dynamic viscosity (Pa s) Fa - resistance force (N) Fr - resistance force (N) γ - surface tension (N/m) h - Planck’s constant (6.626068(33) × 10 -34 J s) I - intensity of transmitted light I0 - intensity of the incident light at a given wavelength J - current density (A/m 2) kB - Boltzmann’s constant 8.617 343(15)×10 −5 (eV/K) lf - fiber length (m) µi - mobility of the charge carrier species i [cm 2/(V s)] ni - density charge of material i (cm -3) Ni - concentration of generating sites of ion species i (cm -3) υ - light frequency (1/s) φ - work function of spectrometer (eV or J) l - distance between electrodes (m) r - pore radius (m) R - resistance (Ω) σdc - direct current bulk conductivity (S/cm) σv - conductivity of material (S/cm) t - thickness of the paper (m) T - tensile strength (kN/m) τb - breaking stress (Pa) θ - contact angle (º) Ui - energy required to dissociate the ion i (eV) v - amount of gas adsorbed (g or cm 3) V - Voltage (V) vm - amount of monolayer adsorbed (g or cm 3) w - solid content of the active substance (g) wf - fiber width (m) y - number of electron-hole pairs Z - zero span strength of paper ζ - zeta potential (mV)

Table of Contents

Abstract ...... i List of Papers ...... iii List of Symbols and Abbreviations ...... v Table of Contents ...... vii 1 Introduction ...... 1 2 Conducting Materials ...... 3 2.1 Conducting Polymers ...... 3 2.1.1 Background...... 3 2.1.2 Doping of Conducting Polymers ...... 5 2.1.3 Applications of Conducting Polymers ...... 6 2.2 Multi-wall Carbon Nanotubes ...... 7 2.2.1 Background...... 7 2.2.2 Physical Properties ...... 7 2.2.3 Applications of Carbon Nanotubes ...... 8 2.3 Ionic Liquids ...... 9 2.3.1 Background...... 9 2.3.2 Room Temperature Ionic Liquids ...... 9 2.2.3 Applications of RTILs ...... 10 3 Paper as a Renewable Material ...... 11 3.1 The Cellulosic Fiber Network ...... 11 3.1.1 Wood Fibers ...... 11 3.1.2 Paper and Papermaking ...... 12 3.2 Surface Treatment of Paper ...... 13 3.3 Some Paper Properties ...... 15 3.3.1 Porosity and Surface Area ...... 15 3.3.2 Surface Roughness...... 15 3.3.3 Tensile Strength ...... 16 3.3.4 Wetting and Liquid Penetration ...... 17 3.3.5 Electrical Properties ...... 18 3.4 Some Characterization Techniques ...... 19 3.4.1 Spectrophotometry ...... 19 3.4.2 Scanning Electron Microscopy - Energy Dispersive Spectroscopy ...... 20 3.4.3 X-ray Photoelectron Spectroscopy ...... 22 3.4.4 Raman Spectroscopy ...... 23 3.4.5 Mercury Porosimetry ...... 24 3.4.6 BET Area ...... 25 3.4.7 Contact Angle Measurement ...... 26 3.4.8 Impedance Spectroscopy ...... 27 3.4.9 Four-Probe Technique ...... 28 3.4.10 Optical Profilometry ...... 29 4 Preparation of Conductive Paper...... 31 vii

4.1 Materials ...... 31 4.1.1 Substrates ...... 31 4.1.2 Conducting Materials ...... 31 4.1.3. Additives ...... 32 4.2 Adsorption of PEDOT:PSS onto MCC ...... 32 4.3 Fiber Beating and Papermaking ...... 32 4.4 Dip Coating of Base Paper into PEDOT:PSS Dispersion ...... 33 4.5 Deposition of PEDOT:PSS Blends into the Fiber Network ...... 33 4.6 Deposition of Ionic Liquid into the Fiber Network ...... 34 4.7 Surface Sizing ...... 34 4.8 Screen Printing and Coating of Electrochromic Polymer ...... 34 5 Interaction Between PEDOT:PSS and Cellulosic Material ...... 35 5.1 Adsorption Isotherm ...... 35 5.2 The Effect of pH and Salt Concentration ...... 36 6 Electroconductive Paper ...... 39 6.1. The Conducting Polymer in the Fiber Network ...... 39 6.1.1 Scanning Electron Micrographs ...... 39 6.1.2 PEDOT to PSS ratio ...... 40 6.2 Some Factors Affecting the Bulk Conductivity ...... 42 6.2.1 The Influence of Fiber Beating and Sheet Forming ...... 43 6.2.2 The Influence of Calendering ...... 45 6.2.3 The Influence of Organic Solvents ...... 45 6.2.4 The Influence of TiO 2 Pigments ...... 48 6.2.5 The Influence of Multi-Wall Carbon Nanotubes ...... 50 6.3 Some Changes in Surface Properties ...... 52 6.3.1 Contact Angle ...... 52 6.3.2 Porosity and Surface Area ...... 55 6.4 Tensile Strength...... 56 7 Ion-Conductive Paper ...... 59 7.1 Ionic Paper ...... 59 7.1.1 The Effect of Relative Humidity ...... 59 7.1.2 The Effect of Temperature ...... 60 7.2 Surface Sizing ...... 61 7.3 Demonstration of Electrochemical Performance ...... 64 8 Conclusion ...... 67 9 Future Aspects and Challenges ...... 69 Acknowledgements ...... 70 References ...... 71

viii

1 Introduction

The search for renewable materials, and the utilization of these materials in various industries, is currently the major concern of this planet. This is because those fossil-based materials which are being utilized are running out, and their production creates ecological imbalance since they cannot be easily replenished. Awareness of this fact promotes research, both in academia and in industry, to either completely remove fossil-based materials or to add renewable materials to its components.

The 21 st century marks the rapid advancements of information technology. One of the essential features of this development is electronics. Humanity has been utilizing various advanced electronic-based products and applications such as mobile phones, computers and the internet, TV, etc in the last decades. A new field of electronics is emerging too; flexible electronics. The accidental discovery of conducting polymers in the late 1970s by Heeger, Shirakawa, and McDiarmid gave birth to the field of organic electronics (Chiang et al 1977). Now, it is possible to construct electronic components on any surface, on both rigid and flexible substrates. The conducting polymer can be deposited through printing (Ballarin et al 2004; Nelson 2007), coating (Denneulin, 2008; Jönsson et al 2002; Kemerink et al 2004), casting (Heeger et al 1997) or layer-by-layer deposition (Agarwal et al 2006; Lvov et al 2006; Qian et al 2006; Wistrand et al 2007; Gerhardt et al 2008) on any types of substrate. The conducting polymer can be deposited in macro-, micro-, and nano-scales. Plastics, by far, are the most widely utilized flexible substrate.

Another alternative flexible material is paper. Its renewability and low cost of production are two of its most attractive properties. Since its first discovery in the Nile river valley of Egypt in 600 BC where it was originally called papyrus , the use of paper has been evolving through the years (Roberts 1996). Paper has been used for the storage of information, in cooking, as a hygienic product, in packaging, in the arts and decorations, etc. It has also been used as insulating material in electrical cables owing to its intrinsically high resistivity. Lately, research on converting paper into a conductive material is increasing (Agarwal et al 2006; Lvov et al 2006; Qian et al 2006; Wistrand et al 2007). Although paper was modified into a conductive material already in the 1960s, the types of material being deposited and their uses are now different. As a porous material, paper has a great potential in large-area electronics. There is a need for flexible and renewable substrates in electronics. Since the field of flexible electronics is young, the utilization of paper as flexible substrate at its early stage would reduce the usage of fossil-based materials such as plastics.

The aim of this study is to prepare and characterize conductive paper as substrate or component wherein an electronic device can be constructed by the deposition of conducting materials. It also attempts to demonstrate that conductive paper can be a compact and robust electronic device in itself. This thesis consists of five papers. Paper I deals with the understanding of the interaction between cellulosic materials and the conducting polymer poly(3,4-ethylenedioxythiophene) doped with poly(4-styrene sulfonate) (PEDOT:PSS) using microcrystalline cellulose (MCC) as a model surface. A commercial paper is also dip-coated into the PEDOT:PSS dispersion and the coated paper is characterized. In paper II, a commercial base paper is coated with blends of PEDOT:PSS and various organic solvents. The effect of organic solvents on the conductivity of the coated paper is investigated. Paper III shows the effect of fiber beating on the conductivity of PEDOT:PSS-coated paper while Paper IV deals with the effect of adding multi-wall carbon nanotubes (MWCNT) to the PEDOT:PSS dispersion. In Paper V, an ionic liquid is deposited into the fiber network and the PEDOT:PSS is coated and screen printed to demonstrate that the conductive paper can facilitate ionic flow and hence facilitate electrochemical reactions.

2 Conducting Materials

Conducting materials are substances that allow the flow of ions and/or electrons along the structure. Generally, materials can be classified into insulators, semiconductors and conductors based on their conductivities, as shown in Figure 2.1. Conductivity is a measure of a material’s ability to conduct an electric current (electrical conductivity) or ionic charge carriers (ionic conductivity). It was in the 1970s when these two new classes of macromolecules started to develop (Armand et al 1997). The electronic conductive materials are doped compounds, in which a compound is inserted to induce conductivity, in a redox process, though the associated motion of ionic and electronic charges. Ionic conductivity, on the other hand, is obtained by dissolving a salt in a suitable polymer lattice.

Figure 2.1 Range of electrical conductivity from insulators to metals (Ajayaghosh 2004).

There are a number of materials that facilitate the flow of either ions or electrons. In this study, three conducting materials are considered: conducting polymer, multi-wall carbon nanotubes (MWCNT) and ionic liquid.

2.1 Conducting Polymers

2.1.1 Background

The accidental discovery of a conducting polymer in the 1970s by doping polyacetylene (Figure 2.2) with iodine opened up many possibilities in the field of electronics (Chiang et al 1977, Shirakawa et al 1974). A conducting polymer is a conjugated polymer with alternating single and double bonds along the backbone resulting in π-conjugated network. The π- conjugated polymer consists of a regular alternating system of single (C-C) and double (C=C) bonds which leads to a lower band gap energy, Eg in the delocalized π-system (Salaneck et al 1996).

H H H H H H

C C C C C C C C C C C C CC n n H H H H H H H H 2 Figure 2.2 The structure of trans-polyacetylene and schematic diagram of sp pz hybridized carbon.

Conjugated polymers are sp2pz hybridized polymers, where the sp2pz hybridized linear carbon chain is responsible for the conductivity (Heeger et al 2010). The three in-plane sigma-orbitals of the sp2 hybridized carbons create the “backbone”; two of them being bonded to the neighboring carbons and the third sigma-orbital being bonded to a hydrogen atom. The fourth electron resides in the pz orbital. This one-electron picture of the pz electron being decoupled from the backbone sigma-orbitals gives these polymers special electrical properties. Traditional polymers such as polyethylenes are electrical insulators because all of the valence electrons are bound in sp3 hybridized covalent bonds which means that there are no mobile electrons to participate in electronic transport.

Electronically, conducting polymers are extensively conjugated molecules that possess a spatially delocalized band-like electronic structure (Skotheim 1986). These bands stem from the splitting of interacting molecular orbitals of the constituent monomer units in a manner reminiscent of the band structure of solid-state semiconductors, as shown in Figure 2.3. The band gap is defined by the energy difference between the Lowest Unoccupied Molecular Orbital (LUMO) and the Highest Occupied Molecular Orbital (HOMO).

Conduction Band LUMO

Energy Band gap

HOMO Valence Band

Figure 2.3 Band structure in an electronically conducting polymer.

The main difference between conjugated and non-conjugated polymers is the presence of the extended π-electron systems in the former (Springborg et al 2001). The conductivity, σ,

4 of a conjugated polymer is proportional to the number of charge carriers, n and their mobility, µ. Since the band gap of a conjugated polymers is normally large, n is very small under ambient conditions. Consequently, conjugated polymer are insulators or semiconductors with too large energy gaps in their neutral state and no intrinsically conducting organic polymer is known at this time (Heeger et al 2010). Typical values of conductivity of polythiophene and polyacetylene are between 10 -10 and 10 -8 S/cm. A polymer can be made conductive by a process called doping.

2.1.2 Doping of Conducting Polymers

The doping of a conducting polymer can be performed via oxidation (p-doping) and/or, less frequently, by reduction (n-doping) of the polymer to generate mobile charge carriers. To compensate for the charge difference, due to the removal or addition of electrons in the semiconducting polymer, a counter-ion is inserted into the structure. Doping is reversible and does not interrupt the polymer chain interaction because of the strong covalent bonds within the chains of the polymer.

A number of doping methods are currently employed, including chemical, electrochemical, photo-induced, and interfacial doping (Heeger et al 2010). In electrochemical doping, the polymer is oxidized or reduced by the electrodes when a voltage is applied. The counter-ion moves from the electrolyte into the polymer, and it diffuses into the structure between the chains. Doping by acid-base chemistry as in the case of polyaniline occurs through a protonation-induced changed in the π-electron structure with no addition of electrons to or widrawal of electrons from the π-electron system. This type of protonation leads to an internal redox reaction resulting in the conversion from semiconductor to metal. In photoinduced doping, the photoexcitation of the polymer excites electron from the filled π- band to the empty π*-band. This means local oxidation of the conjugated polymer backbone and leads to mobile charges, photoinduced positive and negative polarons, and thereby to photoinduced conductivity.

(π-polymer) n + hν → [(π-polymer) +y + (π-polymer) -y]n where y is the number of electron-hole pairs. The interfacial doping relies on the charge injection at the metal-semiconducting polymer interface as in the case of polymer light- emitting diodes. Classic conductive polymers are typically derivatives of polyacetylene, polyaniline, polypyrrole, and polythiophene. One of the most widely used conducting polymers is poly(3,4-ethylenedioxythiophene) doped with polystyrene sulfonate or PEDOT:PSS (Figure 2.4).

Figure 2.4 The structure of PEDOT:PSS.

It consists of conducting PEDOT molecules surrounded by non-conducting PSS molecules. The films of PEDOT have been found to have a very high conductivity (>200 S/cm) (Aleshin et al 1997) and they are stable (Pei et al 1994). The presence of PSS in the PEDOT:PSS enables the complex to form a highly stable dispersion in water, and it has a high visible light transmissivity and robust film formation properties with substantial conductivity (~1-10 S/cm) (Wang et al 2005). This polymer exhibits many attractive properties in terms of ease of processing, high conductivity, and environmental stability (Groenendaal et al 2000; Xing et al 1997).

2.1.3 Applications of Conducting Polymers

The discovery of conducting polymers is quite revolutionary in the electronics industry and is beginning to have a large impact on other industries. Conducting polymers can be used in polymer light-emitting diodes, light-emitting electrochemical cells, as laser materials, in photovoltaic cells and photodetectors, and in polymer field effect transistors (Heeger et al 2010). The use of poly(fluorene) homopolymer and copolymer in detecting DNA, RNA, some proteins, glucose and other small molecules adds to the usefulness of conjugated polymer as biological and chemical sensors (Bazan and Wang 2008). Toxic gases can also be detected by using organic transistor-based phosphonate gas sensor (See et al 2008). PEDOT:PSS-based electrochemical transistors is useful in ion-to-electron transduction and in sensor signal amplification (Berggren et al 2008).

2.2 Multi-wall Carbon Nanotubes

2.2.1 Background

Carbon nanotubes (CNT) are allotropes of carbon having cylindrical nanostructures with remarkable electronic and mechanical properties. In 1991, Iijima first reported the observation of multi-wall carbon nanotubes (MWCNT) and, after less than two years, the single-wall carbon nanotubes (SWCNT) were discovered by him and his coworkers (Iijima 1991, Dresselhaus et al 2001). Since then, carbon nanotubes have been the subject of widespread theoretical and experimental investigations. Carbon nanotubes can be produced by laser vaporization (Guo et al 1995), carbon arch method (Iijima 1991, Ebbesen et al 1992), and chemical vapor deposition (José-Yacamán et al 1993). Naturally, nanotubes can also be present in the flames produced from burning methane (Yuan 1991) or ethylene (Yuan 1991), benzene (Duan et al 1994), and in the soot from both indoor and outdoor air (Murr et al 2004). However, irregular sizes and quality were observed in these naturally occurring nanotubes due to highly uncontrolled conditions. Figure 2.5 shows the structure of multi-wall carbon nanotubes. The MWCNTs are typically 2-25 nm in outer diameter, 1-5 nm inner diameter and a few micrometers in length.

Figure 2.5 Structure of multi-wall carbon nanotubes (MWCNT).

2.2.2 Physical Properties

Carbon nanotubes are nanomaterials with a very high aspect ratio that varies from 500 to 100 000 (Wang et al 2003). Carbon nanotubes are especially interesting because of their exotic electronic properties (Dresselhaus et al 2001). A MWCNT is composed of a set of coaxially arranged SWCNTs of different radii (Iijima 1991). SWCNTs can be cliassified as either semiconductors or metals. In an ideal case, 1/3 of SWCNTs would be metallic and 2/3 semiconducting. Assuming that neighboring sheets do not interact, the electronic properties of MWCNTs would be similar to a set of independent SWCNTs with radii in the

7 range of a few nanometers to ≈ 10 nm. In the case of MWCNTs with larger diameters (~20 nm), the bandgaps are only E g ≈ 44meV (Forró et al 2001). Carbon nanotubes have a magnetic property, as can be observed by Electron Spin Resonance (ESR) showing a mesoscopic nature in their spin relaxation. MWCNTs exhibit field and light emission when individual MWCNTs are attached to a conducting wire and the current is measured after applying a negative potential to the wire (Rinzler et al 1995, Bonard et al 1998).

In terms of tensile strength and elastic modulus, carbon nanotubes are the strongest and stiffest materials yet discovered owing to the covalent sp 2 bonds formed between the individual carbon atoms. The reported tensile strength of a MWCNT is about 63 GPa (Yu et al 2000) and Young’s modulus is up to 1000 GPa (Iijima et al 1991, Wei et al 2003). Carbon nanotubes exhibit good thermal conduction characteristics. It is also well known that MWCNTs tend to form aggregates in solution which can be reduced by physical or chemical alteration (Khosla et al 2010).

2.2.3 Applications of Carbon Nanotubes

There is a wide range of applications for these cylindrical carbon molecules because of their novel properties, as mentioned. These include nanotechnology, electronics, optics, and the architectural field. Attempts are also being made to use carbon nanotubes for ballistic resistance as a potential component in body armor (Zhang et al 2007). Since CNT have the right combination of properties – nanometer diameter, structural integrity, high electrical conductivity, and chemical stability, they are good electron emitters (Forró et al 2001). Prototype cathode-ray lighting elements, flat display panels and gas-discharge tubes (in telecom networks) are being made (Ajayan et al 2001). They are being considered for energy production and storage and in filled composites (Cui et al 2009, Ajyan et al 2001). The bulk nanotubes may never achieve the mechanical property similar to the individual tubes but the composites nevertheless have strengths sufficient for many applications. It can be added to epoxy for improved mechanical properties (Bonnamy et al 2004, Ajayan et al 1998). Other applications of CNT may include tips for atomic force microscopy and tissue engineering (Zanello et al 2010).

2.3 Ionic Liquids

2.3.1 Background

Ionic liquids (ILs) are generally salts in the liquid state based on a substituted heterocyclic cation and an organic or inorganic anion. One of the first ionic liquids synthesized was ethylammonium nitrate in 1914, although at that time it was called a fused salt (Walden 1914). The term “ionic liquid” was first used in 1943 (Barrer 1943). The melting point of ILs is below 100 ºC. The species of cation and anion and the length of the alkyl groups on the cation greatly influence their physical and thermal properties (Marsh et al 2002). Ionic liquids are made up of ions and short-lived ion pairs. Any salt that melts without decomposing or vaporizing usually yields an ionic liquid. Other terms for these substances include liquid electrolytes, ionic melts, ionic fluids, fused salts, and ionic glasses. Since the ionic bonds are stronger than the van der Waals forces between the molecules of an ordinary liquid, common salts tend to melt at a higher temperature than other solid molecules. There are ILs that are liquid at or below room temperature, and these are called room temperature ionic liquids (RTIL).

2.3.2 Room Temperature Ionic Liquids

RTILs have gained popularity in recent years. Figure 2.6 shows the structures of the commonly used cations in ILs, viz.: alkyimidizalium [R 1R2IM] +, alkylpyridium [RPy] +, tetraalkylammomium [NR 4]+ and tetraalkylphosphonium [PR 4]+. The commonly used anions are hexafluorophosphate, tetrafluoroborate, nitrate, methane sulfonate (mesylate), trifluoromethane sulfonate, and bis-(trifluoromethanesulfonyl) amide (Marsh et al 2009).

+ + + R1 R2 N N + N NR 4 PR 4 R1 R2 Imidazolium Pyridinium Ammonium Phosphonium

Figure 2.6 Commonly used cations for ionic liquids.

One interesting RTIL is 1-butyl-3-methylimidazolium tetrafluoroborate ([bmim]BF 4) with a melting point of about −80 °C, which is a colorless liquid with high viscosity at room temperature (Lee et al 2004). Although much research effort has been concentrated on dissolving cellulose in recent years (Marsh et al 2009), [bmim]BF 4 is an IL that does not dissolve cellulosic materials (Rogers et al 2002) and it is suitable as wood preservative (Pernak et al 2004). Figure 2.7 shows the chemical structure of [bmim]BF 4. Some of the attractive properties of RTILs that are useful in a sustainable process are biodegradability

(Gathergood et al 2003), low volatility (Earle et al 2006, Zhu et al 2006) and low toxicity (Jastorff et al 2003, Wasserscheid et al 2002).

+ - N N BF 4

Figure 2.7 Chemical Structure of [bmim]BF 4.

2.2.3 Applications of RTILs

RTILs have been used in biocatalysis and electrochemistry. They serve as electrolytes in solar cells, as double-layer capacitors and in the electrodeposition of metals (Matsunaga 2002, El Abedin et al 2006). They are considered as “green” solvents due to their low volatility (Zhu et al 2006). The applications of RTILs are increasing in various areas including chemical reaction (Ganske et al 2005, Mori et al 2005), electrochemical (Zhao et al 2004, Lu et al 2006), separation applications (Yao et al 2009, Ragonese et al 2009), inorganic nanomaterials (Cao et al 2005, Liu et al 2007), the food industry (Fort el al 2006) and cellulose processing (Marsh et al 2009).

3 Paper as a Renewable Material

One of the greatest discoveries of mankind is paper. For the last two thousand years, it has impacted humanity in a subtle way. One can always find paper at home, in the office, at the grocery stores, etc. It is one of the few inventions that have little impact on the environment because it is composed of cellulosic materials. Cellulose is the most abundant organic compound on the planet and the fact that cellulosic plant materials are continually being generated means that paper is a renewable material.

Since paper is an everyday material, the average person might think it to be a simple material but in reality it is a morphologically, physically, and chemically complex material. Its production process is also highly sophisticated in the sense that it involves a high-speed filtration process which yields a weak wet fibrous network. The fiber web then is pulled continuously, despite its weakness, through the pressing and drying sections of the paper machine to the reel at high speeds so that the web undergoes some extension (Roberts 1996).

3.1 The Cellulosic Fiber Network

3.1.1 Wood Fibers

The tubular wood fibers are composed of cellulosic elements produced from trees. They are primarily extracted from hardwood (deciduous) and softwood (coniferous) trees. Softwood fibers have prominent pits of characteristic shapes along their radial surfaces and they have relatively long average length, i.e. 3-4 mm (Clark 1985). The hardwood fibers are shorter (1- 1.5 mm length) and more slender than the softwood fibers. In general, wood fibers contain about 20-30% lignin, 25-35% hemicellulose, and 45-50% raw cellulose (pure cellulose, mannan, pentosan) (Clark 1985). The amount of these components varies depending on the type of wood fiber. Bundles of cellulose molecules of about 35 Å are called elementary fibrils or crystallites. Nanofibrils are bundles of elementary fibrils with diameter 35-300 Å whereas microfibrils have diameter 300 nm - 0.3 µm. Larger bundles (>0.3 µm) are called fibrils, which collectively include all elements that make up a fiber. It is also possible that fibrils and smaller elements are in the form of thin sheets. All the elements of a fiber are held together with hydrogen bonds.

Fiber properties are greatly influenced by the method of liberating the fibers from the wood matrix known as pulping. There are two general pulping methods, namely mechanical and chemical pulping. In mechanical pulping, wood chips are ground or fed into the center of two refining discs, the fibers are released and the mechanical pulp is obtained. Mechanical

11 pulp fibers are stiff and mostly uncollapsed, and the pulp yield is about 90 to 100% depending on the mechanical method used (Brännvall 2009, Clark 1985). Mechanical pulp has also a large portion of fines, which consist of fragments from the fiber wall and broken fibers. In chemical pulping, the fibers are liberated from the wood matrix by using caustic soda that removes most of the lignin. The most widely used chemical pulping method is kraft cooking (Brännvall 2009). The cooking chemicals used are sodium hydroxide, and sodium sulfide. Chemical pulp fibers are more flexible than mechanical pulp fibers.

The pulp undergoes bleaching so that the resulting paper has better print quality, cleanliness, and ability to resist ageing. Bleaching is usually performed in several stages, with the use of different chemicals at each stage. Bleaching chemicals that are being used are hydrogen peroxide, chlorine dioxide, ozone, and peracetic acid. Pulp beating/refining is another step performed to achieve pulp properties suitable for papermaking. Beating is a general term which means either batch treatment of a pulp slurry in a beater or continuous treatment of thick slurry through a refiner (Clark 1985). The major effects of beating include shortening or cutting of fibers, external fibrillation, abrasion, surface splitting, internal splitting, bruising, and crushing. Other effects of beating are production of debris, longitudinal compression, kinking/curling of fibers, redistribution of hemicelluloses from the interior of the fiber to the exterior, increase in the drainage resistance, and increase in tensile strength.

3.1.2 Paper and Papermaking

Paper is made up of a cellulosic fiber network as shown in Figure 3.1. It was around 105 AD in China that the art of papermaking was first invented by reducing fibrous matter to a pulp in water and forming it as a network of fibers (Pauluparo and Gullichsen 2000). The first continuous paper machine was invented in 1779 and the fourdrinier paper machine was invented in 1801 (Clark 1985). Nowadays, with the growing demand for paper, fibers from wood are increasingly being used (Niskanen et al 1998).

Paper can be handmade or machine-made from dilute pulp slurry with ~ 0.6% dry solids content (Bränvall 2009). The handmade paper is either a part of industrial history or often practiced papermaking procedure in laboratories. The commercial paper production is run at large paper mills using wide and automated paper machines for production efficiency and reliability. Handmade paper normally has randomly orientated fibers (isotropic) while machine-made papers displays varying degree of anisotropy due to the velocity difference between the nozzle of the headbox and the wire. At the points of contact between fibers, strong bonds are formed once the fibers have been dried, due to hydrogen bonds between the polysaccharides in the fiber networks. In general, the fibers lie predominantly in the

12 plane of the sheet and are broadly parallel to each other in the z-direction. The fiber orientation directly affects the in-plane mechanical properties and dimensional stability of paper (Niskanen et al 1998). The areal mass distribution is dependent on the distribution of the fibers in the x-y plane. This local grammage distribution has great influence on many sheet properties. Some properties, e.g. mechanical strength, also depend on the bonding between fibers and not merely on the nature of the fiber distribution. The bonding of fibers and the fiber strength are influenced by pulping and bleaching, and by the fiber preparation conditions employed prior to sheet formation.

Figure 3.1 Scanning electron micrograph of a paper surface.

3.2 Surface Treatment of Paper

The chemical and physical properties of paper can be tailored to suit its specific applications since paper is a versatile material. One way to modify the properties of paper is by modifying the fibers or by adding fillers before the headbox. Various additives such as sizing agents, optical brightening agents, polyelectrolyte, fillers, etc are added to the stock to improve the formation, enhance the paper properties, or add functionalities to the paper. Later in the papermaking process the surface of the paper maybe modified by surface sizing, coating, or calendering.

Paper coating is performed to alter the surface properties of the final paper product such as graphical and barrier properties. Coating methods include dip coating, rod coating, knife coating, blade coating, forward and reverse roll coating, slot and extrusion coating, slide coating and curtain coating (Cohen 1992). Normally, the coating formulations contain high

13 solids content, i.e. ~25-70% for barrier polymer dispersion coating (Kimpimäki and Savolainen 1997) and ~50-70% for pigment coating (Attrup and Hansen 1997). In laboratory rod coating, the coating material is placed in front of the wire-wound rod (Figure 3.2). As the rod moves, the coating material forms a wet film on the surface of the paper that corresponds to the thickness of the wire diameter. The laboratory rod coating is often used as a first step in a development process, both in industry and in academia. This is because the results correlate, within certain boundaries, to pilot and production scale and it is inexpensive.

Figure 3.2 Surface treatment of paper using rod coating.

The coating material may also penetrate deep into the fiber network, thereby modifying both the surface and bulk properties of paper as shown in Figure 3.3, although this is not often practiced. This is accomplished by using rod of thin wire diameter, slow movement of the rod, low solid content coating mixture, and using a highly porous base paper (e.g. made from unbeaten or slightly beaten fibers). In roll coating, the penetration of coating suspension is achieved by increasing the nip pressure. The viscosity of the coating mixture is also an important parameter that influences its penetration into the paper. The coated paper is then dried in an oven at a specified temperature and time.

wire -wound rod

coating material base paper

Figure 3.3 Schematic picture of rod coating showing the penetration of conducting material into the base paper.

Another surface treatment process is calendering, wherein the paper is compressed between two rolls. The rolls can be lined with different surface material; normally metal or polymer, to alter the nip length. Heating these rolls softens the paper material and increasing the line load increases the compression which can be used to govern the induced deformation. Calendering reduces the roughness, increases the gloss and improves the printability of paper and paperboard (Perry 1993, Holik 2006).

3.3 Some Paper Properties

3.3.1 Porosity and Surface Area

Paper is a porous material as shown in Figure 3.1. Porosity is the ratio of the pore volume to the total volume. The pores in papers are categorized into micro- and macro-pores. The micropore structures have diameters of ≤ 1.5 µm and consist of fiber pores, fine inter-fiber voids, and fine voids between fibers and fine materials. Macropores have diameters greater than 1.5 µm and consist of larger inter-fiber voids, and uncollapsed pulp fiber lumens (Yamauchi et al. 1979, Yamauchi and Kibblewhite 1988). The quantity of macropores is reduced by fiber beating because the fibers become more consolidated, with greater inter- fiber bonding. The specific internal surface area of paper is the internal surface area of pores per unit mass of sample. Porosity is one of the important factors to consider in designing a paper that may allow or hinder material penetration into the structure.

3.3.2 Surface Roughness

Since paper is a network of cellulosic fibers, the uneven surface of paper creates roughness. Roughness refers to the uneven surface of paper or board. It is difficult to distinguish between the end of the surface and the beginning of the internal pore structure of paper. Kajanto et al (1998) categorized roughness into three components according to the in-plane resolution: optical roughness (< 1 µm), micro-roughness (1 – 100 µm), and macro-roughness (0.1 – 1 mm). Roughness is significant in printing papers, graphical and packaging boards. It greatly influences the optical properties of paper such as gloss, ink absorption, and the amount of coating necessary in order to achieve good coverage (Kajanto et al 1998). To decrease the surface roughness and achieve good barrier properties against water, air, grease, etc., paper can be surface sized. Figure 3.4 shows the surface of a paper covered by surface sizing agent and this maybe compared to the base paper in Figure 3.1. Another way of decreasing the surface roughness is by calendering as described in Section 3.2. The paper web is pressed in one or more “rolling” nips formed by rollers (Feldmann 2006).

Figure 3.4 Scanning electron micrograph of the cross section of paper showing a portion of the surface covered by a surface sizing agent .

3.3.3 Tensile Strength

Tensile strength measures the load a paper specimen can withstand before breaking. It is the breaking force per unit area (in MPa), which has the same unit as the elastic modulus. Often, tensile strength is expressed in kN/m, i.e the breaking force divided by the width of the strip (Niskanen et al 1998). The tensile index is tensile strength divided by the basis weight (Nm/g). The strength of paper is influenced by the gradual failure of inter-fiber bond and fiber rapture. Figure 3.5 shows the typical load (stress) versus elongation (strain) curve of paper. Stress is being transferred to and from every fiber through many bonds. Paper stretching gradually opens the inter-fiber bonds. Bond failure occurs when the shear force on the bond exceeds its shear strength.

tensile strength

Load

strain at break Elongation

Figure 3.5 Schematic sketch of typical load-elongation curve of paper.

The tensile strength depends on the “weak” points of the specimen. A number of models have been presented to show how the fiber and bond failures affect the paper strength. One of the most celebrated empirical models is the Page equation (Page 1969),

ͥ ͭ ͧ2 ę (3.1) Ɣ ͬ ƍ cĕ'ę where T is the tensile strength, Z is the zero span strength of paper, RBA is the relative bonded area, w f is the fiber width, l f is the fiber length, and τ b is the breaking stress of bonds (breaking force over area). The Page model started from the experimental observation that the tensile strength is proportional to the fraction of broken fibers along the rupture line in beating trials (Niskanen et al 1998). This equation suggests that the inverse of tensile strength is linearly proportional to the inverse of fiber length, fiber strength and RBA. Fiber beating improves the tensile strength due to more flexible fibers, which facilitate the formation of fiber-fiber bond. It is found that machine-made papers have different cross machine direction (CD) and machine direction (MD) tensile strength. The tensile strength of paper is measured using a mechanical tester. A strip of paper is clamped at both ends and slowly pulled apart until it breaks. ISO 1924 describes the procedure for the test.

3.3.4 Wetting and Liquid Penetration

Wood fibers are hydrophilic in nature which means that their network is also hydrophilic. In principle, a liquid wets the surface only if its surface tension is lower than that of the surface (Dullien 1992). Wetting is a surface phenomenon which is influenced by chemical composition of the surface layers down to monomolecular thickness. Another factor that affects wetting is the surface morphology. It is reported that drops of non-wetting liquids tend to exhibit higher contact angles on rough surfaces and extend more readily along grooves or fibers than across them (Oliver and Mason 1976). On the other hand, wetting liquids exhibit a lower contact angle on a rougher surface of the same material. The reason is that all systems tend to move to the direction of reducing surface free energy. The liquid penetration into the fiber network can be interrupted by dislocation in local surfaces such as sharp edges of mineral fillers, e.g. clay and calcium carbonate or ridges, on fibers. Local non- uniformity and anisotropy in wetting behavior are due to variation and orientation in fiber and network morphology (Lyne 2002). Wetting can be altered by deposition of chemical agents either on pulp fibers or by surface sizing the paper (Nyström et al 2006). Beating the fiber to a considerable degree is another factor that influences wetting behavior of paper by creating a “smoother” paper surface (Clark 1985) due to a decrease in macroporosity.

3.3.5 Electrical Properties

The electrical properties of paper include conductivity (or resistivity), dielectric constants and dielectric loss, permittivity, and dielectric breakdown strength. This section focuses on the conductivity of paper as this study deposits conductive material to achieve highly conductive paper. The conductivity of any material, σ v is expressed as

1 Ɣ ∑$ ͢$͙$$ (3.2) where ni is the density charge of material, e i is the electronic charge on the carrier, and µ i is the mobility of the charge carrier in the solid. In paper, the n i is often expressed according to the Boltzmann equation (Baum 2002):

Ĝ $ $ (3.3) ͢ Ɣ ͈ ͙ͬͤ ʢƎ ͦ&ûʣ where Ni is the concentration of ionizable sites of species i, Ui is the energy required to dissociate the ion, k B is the Boltzmann constant, and T is the absolute temperature. The ionic or electronic conductivity of paper is influenced by moisture content, fiber orientation, temperature and the presence of salt or impurities (O’ Sullivan 1947, Baum 2002). Paper can be surface-conductive, bulk-conductive or both (Figure 3.6).

(a) (b) Figure 3.6 Schematic picture of (a) surface-conductive and (b) bulk-conductive paper.

The bulk or volume conductivity is the ratio of current density J (A/m 2) to the applied electric field strength E (N/C)

¦ Ɣ 1¡ (3.4)

This quantity is scalar for isotropic materials; however for anisotropic material such as paper, this varies in the three principal directions (machine direction, cross-machine direction, and Z direction). The bulk conductivity is the reciprocal of bulk resistivity ρ v. It is related to the bulk resistance R in the equation

ͥ ' (3.5) 1 Ɣ ʠʡ ʠʡ

18 where l is the distance between electrodes and a is the cross-sectional area of the specimen between electrodes. On the other hand, the surface conductivity (or resistivity) is taken as the surface conductance (or resistance) measured between the opposite sides of square on a surface.

The bulk conductivity of base papers is normally in the order of 10 -12 S/cm, which falls in the category of a resistive material (Josefowicz et al 1982). As a result, paper, in combination with other materials such as saturating oils and resins, can be used in transformers and in electronic circuit boards (Baum 2002). The electrical property of paper is important in reprographic processes because it affects both the image quality and runnability. The paper should have relatively high bulk and surface conductivities. Earlier studies on improving paper conductivities had reported a surface-conductive paper with a surface resistivity of 30 Ω/sq using 72% graphite (Casey 1968), and a variety of patents have described conductive papers using chemical impregnations such as poly(diallyldimethylammonium chloride) (Moser 1968) which is mainly used for antistatic applications.

3.4 Some Characterization Techniques

The following sections briefly describe the theory of some techniques used to characterize the adsorption of conducting polymer on the surface, and the properties of the base and coated paper.

3.4.1 Spectrophotometry

Spectrophotometry is a technique used to quantify the amounts of substance present in solution by measuring the intensity of transmitted light. Figure 3.7 shows a schematic diagram of how light passes through the sample to the detector. This method makes use of the Beer-Lambert law to measure the concentration of the absorbing species (Skoog et al 2007):

−=  I  = ε ⋅⋅ A log 10   Lc (3.6)  I0  where A is the measured absorbance, I 0 is the intensity of the incident light at a given wavelength, I is the intensity of the transmitted light, L is the path length through the sample, and c is the concentration of the absorbing species.

Adjustable aperture

I I O Light source photoresistor amplifier

monochromator sample output

Figure 3.7 Diagram of a single-beam spectrophotometer.

For each species and wavelength, ε is a constant known as the molar absorptivity or extinction coefficient. This constant is a fundamental molecular property in a given solvent, at a particular temperature and pressure, and has units of 1/(Mcm) or often AU/(Mcm). The absorbance and extinction coefficient ε are sometimes defined in terms of natural logarithms instead of base-10 logarithms. The Beer-Lambert Law is useful for characterizing many compounds that exhibit a direct relationship between the absorbance and concentration. It is important to note whether the absorbance and concentration have a linear relationship. Mostly, nonlinearity occurs at higher concentration of the compound where a calibration curve is required to show the relationship between the absorbance and the concentration.

3.4.2 Scanning Electron Microscopy - Energy Dispersive Spectroscopy

The Scanning Electron Microscopy (SEM) permits the observation and characterization of heterogeneous organic and inorganic materials on a nanometer or micrometer scale (Goldstein et al 2003). An electron beam, which has an energy ranging from 0.5 keV to 40 keV, is scanned across the surface of a sample. The schematic drawing of electron column in Figure 3.8 shows the electron beam emitted from electron gun fitted with a filament cathode. Normally, tungsten is used as a filament in thermionic electron gun because it has the highest melting point and lowest vapor pressure of all metals, thereby allowing it to be heated for electron emission. When the electrons strike the sample, a variety of signals are generated, and it is the detection of specific signals which produces an image or elemental composition. The three signals which provide the greatest amount of information are the secondary electrons, backscattered electrons, and X-rays.

Figure 3.8 Schematic diagram of electron column (Goldstein et al 2003).

Secondary electrons are loosely bound outer shell electrons from the specimen atoms which receive sufficient kinetic energy during inelastic scattering of the beam electrons. They are emitted from the atoms occupying the top surface and produce a readily interpretable image of the surface. The contrast in the image is determined by the sample morphology and a high image resolution can be obtained because of the small diameter of the primary electron beam. Backscattered electrons (BSE) are primary beam electrons whose trajectories have intercepted a surface, and which thus escape the specimen (Goldstein et al 2003). They have undergone numerous elastic scatterings to accumulate enough deviation from the incident beam path to return to the surface. The contrast in the image produced is determined by the atomic number of the elements in the sample. The distribution of different chemical phases in the sample can also be seen. The resolution in the image is however not as good as from secondary electrons because these electrons are emitted from a depth in the sample.

Energy-dispersive X-ray spectroscopy (EDS or EDX) is a technique used for the elemental analysis or chemical characterization of a sample. The analysis owes to the fundamental principle that each element has a unique atomic structure allowing X-rays that are characteristic of an element's atomic structure to be identified uniquely from one another. A high-energy beam of charged particles such as electrons or protons or a beam of X-rays is focused into the sample to stimulate emission of characteristic X-rays (Figure 3.9). The incident beam may excite an electron in an inner shell, ejecting it from the shell while creating an electron hole where the electron was. An electron from an outer, higher energy shell then fills the hole, and the difference in energy between the higher energy shell and the

21 lower energy shell may be released in the form of an X-ray. An energy-dispersive spectrometer detects the number and energy of the X-rays emitted from a sample. This allows the elemental composition of the specimen to be measured since the energy of the X- rays is characteristic of the difference in energy between the two shells and of the atomic structure of the element from which they were emitted.

Figure 3.9 Schematic picture of the atomic shell.

3.4.3 X-ray Photoelectron Spectroscopy

X-ray Photoelectron Spectroscopy (XPS) is used to determine the quantitative elemental composition and the chemistry of compounds on the surface. It is also known as Electron Spectroscopy for Chemical Analysis (ESCA). It is a surface analysis technique with a sampling volume that extends from the surface to a depth of approximately 10 nm. XPS is an elemental analysis technique that is unique in providing chemical state information of the detected elements. It is possible to discriminate between the same atoms bonded to other atoms, e.g. sulfur from thiophene and sulfur from sulfonic acid. The process starts by irradiating a sample with monochromatic X-rays, Al Kα (1486.6 eV) or Mg Kα (1253.6 eV), with energy hυ, and this leads to the emission of photoelectrons whose energies are characteristic of the elements within the sampled volume (Hüfner, 2003). The incident photon with energy hυ ionizes an electronic shell and an electron which was bound to the solid with energy Eb is ejected into the vacuum with kinetic energy Ek. By conservation of energy,

= υ − − φ k hE Eb (3.7)

22 where Ek is the kinetic energy of the electrons, h is Planck’s constant, υ is the light frequency, and φ is the work function of the spectrometer. These ejected electrons move towards the electron collection lens and proceed to electron energy analyzer where their kinetic energies are measured. These electrons are counted in the electron detector which then produces a characteristic spectrum of elements present in the sample. Figure 3.10 shows the schematic picture of the whole process. Most XPS analysis requires ultra-high vacuum (UHV) conditions which imply that volatile samples cannot be detected.

Energy + Analyzer

Electron Lens Crystal Detector System Source Slit Amplifier and Sample X-ray Ratemeter Monochromator

X-ray Computer Anode

Figure 3.10 Schematic picture of an XPS spectrometer employing hemispherical electron energy analyzer and monochromatic X-ray source (Leone and Signorelli 1996).

3.4.4 Raman Spectroscopy

Raman spectroscopy is a powerful technique for studying vibrational, rotational, and other low-frequency modes in a system (Gardiner 1989). It relies on inelastic scattering, or Raman scattering, of monochromatic light, usually from a laser in the visible, near infrared, or near ultraviolet range. The laser light interacts with phonons or other excitations in the system and the energy of the laser photons is then shifted up or down. The shift in energy gives information about the phonon modes in the system. This technique was discovered by C.V. Raman in 1928 using sunlight as the light source (Smith and Dent 2005). Figure 3.11 shows the schematic diagram of the energy states involved in a Raman signal. Typically, a sample is illuminated with a laser beam. Light from the illuminated spot is collected with a lens and passed through a monochromator. Wavelengths close to the laser line, due to elastic Rayleigh scattering, are filtered out while the rest of the collected light is dispersed onto a detector. The scattered radiation, observed at 90° to the incident beam is of three types: Rayleigh, Stokes, and anti-Stokes. The Rayleigh-scattered radiation is more intense than

23 either of the two types. In the quantum mechanical analysis of the Raman effect, radiation is considered as stream of photons scattered by collision between them and the molecules of the sample. Most of the collisions are elastic, i.e. no net transfer of energy occurs; a few collisions are inelastic.

The vibrational energy of a bond is added to or subtracted from the energy of the incident photon, changing its frequency. The energy of the molecule can assume any of the infinite number of virtual states between the ground state and excited electronic energy levels since the molecule is not generally excited into its first excited electronic energy level. For anti- Stokes behavior to occur, the molecule must already be in an excited vibrational energy level and the probability of this occurrence can be estimated by assuming a Boltzmann distribution. The polarizability of the molecule must change during the course of vibration in order for a particular vibrational mode to appear in the Raman spectrum (Mendham et al 2000). Raman spectroscopy has been used to study the properties of paper coatings (Tripp et al 2002) and conducting polymers (Duvail et al 2002, Ouyang et al 2004/2005, de Kok et al 2004, Furukawa et al 2005, Garreau et al 1999).

Virtual states

n m Vibrational states

Stokes Rayleigh anti -Stokes

Figure 3.11 Diagram of the Rayleigh and Raman scattering process (Smith and Dent 2005).

3.4.5 Mercury Porosimetry

Mercury porosimetry is an analytical technique used to determine the pore size distribution, the total pore volume or porosity, and bulk and absolute densities, and the specific surface area of a porous material (Giesche 2006). Liquid mercury is used because it does not wet most of the substances and it has to be forced to penetrate pores (Webb et al 1997). The pressure needed to force the mercury into the pores is inversely proportional to the opening size. Liquid mercury has high surface tension and exhibits a high contact angle. The surface tension of mercury acts along the circle of contact for a length equal to the perimeter of the circle when mercury is in contact with a pore opening of circular cross section. The force, F r with which mercury resists entering the pore is given by

̀- Ɣ Ǝ2ͦ cos (3.8) where D is the pore diameter (m) , γ is the surface tension (N/m), and θ is the contact angle. The negative sign appears because of θ > 90º, the term is intrinsically negative. On the other hand, the force, F a due to externally applied pressure acts over the area of the circle of contact and is given by,

ͦ ̀ Ɣ ͦ ͊ (3.9) where P is the applied pressure. At equilibrium,

_v (3.10) ͨ Ɣ Ǝ̾ cos or ͯͨR ΑΝΡ W ͯͦR ΑΝΡ W +*- or (3.11) ̾ Ɣ ͊ Ɣ -ģĢĥĘ

Equation 3.11 assumes cylindrical pores and is known as the Washburn equation (Giesche 2006). No pores are penetrated by mercury unless a force is applied (Webb et al 1997).

3.4.6 BET Area

Brunauer, Emmet and Teller (1938) published an article which explains the physical adsorption of gas molecules on a solid surface and this became the basis for the measurement of the specific surface area of a material. The BET theory is expressed as

ͥ ͯͥ ͥ (3.12) \ʞʚt⁄ ʛͯͥʟ Ɣ \Ġ ʠtʡ ƍ \Ġ where v is the amount of gas adsorbed, v m is the monolayer quantity of adsorbed gas, P and P0 are the equilibrium and saturation pressures of the gas molecule at the temperature of adsorption, and c is expressed as

ͯ u ą (3.13) ͗ Ɣ ͙ͬͤ ʠ ʡ where E 1 and E L are the heats of adsorption of layer 1 and higher layers, respectively, R is ͥ the universal gas constant , and T is the temperature. If is plotted vs , v m and \ʞʚt⁄ ʛͯͥʟ ʠtʡ c can be calculated from the slope and y-intercept. The linear relationship of this equation is

25 maintained only in the range of 0.05 < P / P0 < 0.35. The specific surface area, SBET can be determined using the equation

d . Ġ (3.14) ͍ Ɣ where N is the Avogadro number, s is the adsorption cross section of the adsorbing species, V is the molar volume of adsorbate gas, and a is the mass of the adsorbate. This theory is an extension of the Langmuir theory of monolayer molecular adsorption to multilayer adsorption with the following assumptions: (a) gas molecules are physically adsorbed on a solid in layers infinitely, (b) there is no interaction between each adsorption layer, and the Langmuir theory can be applied to each layer. The use of nitrogen adsorption isotherm at liquid nitrogen temperature with the BET calculation is a widely accepted method to measure the internal surface area of paper (Murakami and Yamauchi 2002).

3.4.7 Contact Angle Measurement

The contact angle is the angle at which a liquid/vapor interface of a liquid drop meets the solid surface. It is specific for any given system and is determined by the interactions across the three interfaces. Figure 3.12 shows the contact angle of a small liquid droplet resting on a flat horizontal solid surface.

vapor γLV θ liquid

γSV γSL solid

Figure 3.12 Contact angle of a liquid sample.

The Young equation describes the equilibrium of a liquid on a solid (Dullien 1992)

−= γγθγ LV cos SV SL (3.15) where γ is the interfacial free energy (surface tension), the subscripts SV, SL and LV refer to the solid-vapor, solid-liquid and liquid-vapor interfaces, respectively, and θ is the contact angle. When a drop of liquid is placed on a solid material, the drop spreads until the interfacial forces are balanced. If the forces of adhesion between the solid and the liquid are greater than the forces of cohesion of the liquid, the liquid will spread and will perfectly wet

26 the surface spontaneously. If the forces reach an intermediate balance determine by the surface tensions, then the liquid will form a definite contact angle (Roberts 1996). For rough surface, the Young equation becomes (Wenzel 1936; Good 1951)

( −= γγθγ ) LV cos r SV SL (3.16) where r is the roughness ratio which is a measure of how surface roughness affects a homogenous surface. The roughness ratio is the ratio of actual area of the interface to the apparent area of the geometrical surface. It is important to note that the Young equation assumes a perfectly flat surface, whereas in many cases surface roughness and impurities cause a deviation in the equilibrium contact angle from the contact angle predicted by this equation. Hydrophilicity can be assessed by placing a drop of water on the surface of a paper and observing whether it wets the surface. A systematic way of evaluating hydrophilicity is by measuring the contact angle of water on paper surface using the static/dynamic sessile drop method (Figure 3.13).

Figure 3.13 Contact angle measurement set up (FTÅ 200).

3.4.8 Impedance Spectroscopy

Impedance spectroscopy (IS) is a powerful technique in characterizing many of the electrical properties of materials and their interfaces with electronically conducting electrodes (MacDonald and Johnson 2005). The dynamics of the bound or mobile charge in the bulk of interfacial regions of any kind of solid and liquid materials: ionic, semiconducting, mixed electronic-ionic, and insulators can be investigated. The specimen is sandwiched between two identical electrodes in the form of a circular cylinder or a rectangular parallelepiped. An

27 electrical stimulus (a known voltage or current) is applied to the electrodes and the response (resulting voltage or current) is measured. The impedance is measured by applying a single frequency voltage or current. The phase shift and amplitude, or real and imaginary parts, of the resulting current at that frequency is also measured using either an analog circuit or fast Fourier transform (FFT) analysis of the response. The impedance is commonly reported as a function of frequency (1 mHz to 1 MHz) in commercial instruments. Any intrinsic property that influences the conductivity can be investigated by IS. Properties such as conductivity, dielectric constant, mobility of charges, equilibrium concentrations of the charged species, bulk generation-recombination rates, adsorption-reaction rate constants, capacitance of the interface region, and diffusion coefficient of neutral species in the electrodes itself can be investigated (MacDonald and Johnson 2005). The imaginary impedance can be plotted against the real impedance, and this plot is known as the Nyquist plot (Cole-Cole plot). Equivalent circuit element can then be determined based on the shape of this curve. Another form of data presentation is by plotting the phase angle and the logarithm of the magnitude of the impedance versus the logarithm of the frequency which is known as the Bode plot. This plot complements the Nyquist plot since it shows the electrical behavior of the sample within the range of frequency (Unwin 2003).

3.4.9 Four-Probe Technique

Paper has a rough surface which implies that there may not be a good contact between the surfaces of paper and the measuring electrodes resulting in a contact resistance. An adequate force should be applied in order to minimize the contact resistance without deforming the paper. For resistive material (paper), ASTM D 257 is widely used. The circular paper sample is sandwiched between two circular electrodes. This standard recommends an applied voltage of 500 ± 5 VDC, 60 s electrification time and 140-700 kPa pressure to minimize contact resistance. The four-probe technique (ASTM D4496-04) is used for the measurement of moderately conductive sheets (Figure 3.14). The technique measures both the surface and bulk conductivities, and reduces effect of contact resistance and current spreading. The technique allows the application of a force to compress the material but it has been reported that the force applied as well as the applied voltage affect the conductivity of the sheets (Borch et al 2002; Josefowicz et al 1981).

(b)

electrode

Power conductive sheet Source A plastic insulator

(a)

Figure 3.14 Schematic picture of four-probe assembly: (a) bottom view and (b) front view.

The paper samples are cut into 10 cm x 15 cm pieces and placed inside the measurement chamber. The two outer electrodes are the current electrodes connected to a multimeter while the inner two are potential electrodes connected to a multimeter. A bias was applied to the sample. The voltage and current across the sample were read after an exposure time of 30 s of exposure time. The direct current bulk conductivity, σ DC was calculated using the equation c  I  σ =   (3.17) DC t V  where c is the ratio of the distance between the potential electrodes to the width of the paper, t is the thickness of the paper (cm), I is the current that passes through the sample (A), and V is the voltage across the potential electrodes (V). The four-point probe technique was originally developed to measure the earth’s resistivity (Wenner 1916). Nowadays the technique has become popular in the semiconductor industry.

3.4.10 Optical Profilometry

Optical profilometry is a technique that measures surface roughness. There are three common techniques used in characterizing the paper surface namely, focus error detection, triangulation, and interferometry (Visscher and Struik 1994, Wågberg and Johansson 2002). The focus error detection is based on the pupil obscuration method wherein the signal difference received by the outer and inner diodes is detected by the so called focus error signal. In triangulation technique, a beam of light hits the surface of the sample, which is

29 inclined at a specific angle, and the specularly or diffused scattered light is collected and focused on a position-sensitive detector. In interferometry, a light beam is split inside an optical interference profiler that reflects half the beam from a test material which is passed through the focal plane of a microscope objective and the other half of the split beam is reflected from the reference mirror (Figure 3.15). The beams reflected from the test surface and the reference surface recombine to form interference fringes. These fringes are the alternating light and dark bands which are visible when the surface of the sample is in focus (in Wyko Surface Profilometer). The two most widely used techniques are phase shift interferometry (PSI) and vertical scanning interferometry (VSI) (Wågberg and Johansson 2002). The phase shift type uses a broadband light source with short coherence length to measure smooth surfaces whereas the vertical scanning type uses a narrowband light source with long coherence length suited to measure rough surfaces like paper. The resulting interference pattern at many different relative phase shift (for PSI mode) or degree of fringe modulation (for VSI mode) is converted to wavefront (phase) data by integrating the intensity data. The data are processed to remove ambiguities between adjacent pixels, and then the relative surface height can be calculated.

Figure 3.15 An interference microscope (Veeco Metrology Group 1999).

4 Preparation of Conductive Paper

This section presents the materials and methods used in order to prepare the substrate and deposit the conducting material or sizing agent. The materials are categorized into substrates, conducting materials and additives. There two types of substrate, namely microcrystalline cellulose and paper. The base paper can either be hand-made or machine-made. The methods below do not specify the type of paper used. In the next chapters however the type of paper is indicated and the article to which the results appeared.

4.1 Materials

4.1.1 Substrates

Microcrystalline cellulose (MCC) having a 20 µm average particle size was purchased from SigmaCell®. The MCC underwent pretreatment to remove residual hemicelluloses (Hamilton and Quimby 1957; van de Steeg 1992). The measured specific areas based on BET (Brunauer, Emmett, and Teller) analysis for the pretreated and untreated MCC samples were 3.04 m 2/g and 1.90 m 2/g, respectively.

Commercial base papers were obtained from Munksjo Paper AB (bleached yellow, 150 g/m 2), Nordic Paper AB (unbleached, 100 g/m 2), and Billerud AB (white, 100 g/m 2). The pulp substrate used was a bleached softwood kraft pulp flash-dried pulp (Billerud Kraft Pulp SW Gruvön) purchased from Billerud AB (Gruvön Mill, Sweden).

4.1.2 Conducting Materials

The conducting polymer PEDOT:PSS (Clevios or Baytron® P) from Clevios (formerly H.C. Starck Gmbh), Germany was used. This conducting polymer is known to have a particle size of about 80 nm, 1:2.5 PEDOT to PSS ratio, and a film conductivity of approximately 1- 10 S/cm (Kirchmer et al. 2005). Two PEDOT to PSS ratios were used: 1:2.5, and 1:6. The Orgacon® EL-P3040 from Agfa (Belgium) was used as screen-printing ink. This ink is a transparent conductive ink that contains PEDOT:PSS and a thermoplastic binder. The multi-wall carbon nanotubes (MWCNT) were purchased from Sigma-Aldrich (Germany) and had >90% carbon basis, O.D. × I.D. × L 10-15 nm × 2-6 nm × 0.1-10 µm. The analytical grade ionic liquid used in this study was 1-butyl-3-methylimidazolium tetrafluoroborate ([bmim]BF 4) from Sigma-Aldrich Gmbh (Germany).

4.1.3. Additives

Analytical grades of organic solvents such as N-methyl-2-pyrrolidone (NMP), dimethylsulfoxide (DMSO), sorbitol, and isopropanol were added to the conducting polymer dispersion as conductivity enhancers. Analytical grade of titanium dioxide TiO 2 powder (rutile and anatase forms) was purchased from Sigma-Aldrich (Germany). Other analytical grade chemicals used in the study include poly(dimethyldiallyl-ammonium chloride) (polyDADMAC), sodium polyethylene sulfate (PES-Na), NaOH, HCl, NaCl. Deionized water (18 MΩ) was used in all the experiments.

4.2 Adsorption of PEDOT:PSS onto MCC

The adsorption equilibrium was investigated using an indirect UV-Vis spectrophotometric method. The amount of PEDOT:PSS adsorbed onto the microcrystalline cellulose was inferred from the decrease in the absorbance of PEDOT:PSS solution after interaction with MCC. Filtration was performed to separate the MCC from the solution. Based on a preliminary investigations, there was no specifically defined value for the maximum wavelength for the PEDOT:PSS solution from 200 nm to 1100 nm in UV/Vis Spectrometer Lambda 14 (Perkin Elmer, USA). However, there were shoulder values from around 700 nm to 1100 nm of which a slight maximum value at 930 nm was observed. Thus, subsequent measurements were made at 930 nm. An equilibrium time of 27 hours was obtained for the adsorption of polymer onto MCC.

Different concentrations of PEDOT:PSS (50, 100, 150, 200, 250,… ppm) were prepared. The pH of each solution was adjusted to 3.0 by adding 0.1 M HCl. An amount of 0.05 g of pretreated MCC powder was added to each 15-ml tube containing PEDOT:PSS and agitation was maintained during the whole equilibration period. To determine the effect of pH, solutions of the same PEDOT:PSS concentration were prepared at various pH. To determine the effect of salt concentrations, solutions with different salt concentrations of PEDOT:PSS were prepared.

4.3 Fiber Beating and Papermaking

The pulp samples were disintegrated and beaten in a PFI mill (Lorentzen Wettre, Sweden) at 1,000 to 10,000 revolutions according to the standard method (ISO 5264-2 1979). The extent of beating of the different pulp samples was monitored by measuring the SR drainability (ISO 5267-1 1999) of the pulp suspensions after beating, as well as that of the unrefined samples. The SR number increases as the degree of fiber beating increases.

Paper sheets with a grammage of 100 g/m 2 were made using a standard British handsheet (HS) maker (Messmer Bücher, UK) and a dynamic sheet former (DSF) manufactured by Warner Tollo AB (Kristianstad, Sweden). A handsheet former filters a dilute fiber suspension producing a paper sheet containing mostly randomly oriented fibers (ISO 5269-1 1979), although “flocs” do occur. The DSF sprays a dilute fiber suspension onto a rotating wire, orienting fibers in the direction of rotation in a fashion similar to the machine direction orientation of fibers in a commercially produced paper. There is a least tendency that the fibers flocculate in this process because of the high shearing forces through the spray jet. The DSF operation parameters were: 0.2% fiber concentration, jet nozzle 300 kPa, and wire speed 1200 rpm, resulting in an anisotropy value of 2.0. Both sheet types were blotted, pressed sequentially at 300 kPa and 500 kPa, and dried. The handsheets were dried overnight (23 °C and 50% RH) under restraint according to ISO 5269-1. The DSF sheets were dried at 105 °C for 7 minutes under edge restraint (STFI Infra-Red dryer, Sweden) and stored at 23 °C and 50% RH overnight prior to coating.

4.4 Dip Coating of Base Paper into PEDOT:PSS Dispersion

Pieces of the commercial base papers with dimensions 5.0 cm x 6.0 cm were dipped into the PEDOT:PSS dispersion at room temperature for 10 minutes. These dip-coated papers were immersed in well-stirred milli-Q water at pH = 2.0 for 5 minutes and then placed on top of a blotter and dried at 50% RH and 23 ºC for 24 hours. Dried samples, including the uncoated base paper (control), were studied by XPS. Samples (about 1.0 mm 2) were studied using a monochromatic Al X-ray source and XPS spectra were recorded using a Kratos AXIS Ultra DLD X-ray photoelectron spectrometer (Kratos Anaytical, Manchester, UK). Most of the signal comes from an area of 700 µm x 300 µm. Wide spectra were run to detect elements present in the surface layer.

4.5 Deposition of PEDOT:PSS Blends into the Fiber Network

The base paper was placed on top of a blotter and coated using an RK Control Coater (UK) with different blends of PEDOT:PSS. A wire-wounded rod with 0.08 mm wire diameter was used to ensure deep deposition of PEDOT:PSS blends into the paper. Both sides of the paper were coated unless otherwise specified. The coated paper sample and the blotter were dried for 5 minutes using the STFI Infra-Red (IR) dryer (Sweden) at about 110 °C. Dried samples were stored at 23 °C and 50% RH.

4.6 Deposition of Ionic Liquid into the Fiber Network

The commercial base paper was placed on top of a blotter and coated using an RK Control Coater (UK) with [bmim]BF 4. The temperature of the ionic liquid was maintained at least 23 ºC. A wire-wound rod with 0.08 mm wire diameter was used to ensure deep deposition of the ionic liquid into the paper. The coated paper sample and the blotter were dried for 5 minutes using the STFI Infra-Red (IR) dryer (Sweden) at about 110 °C. Dried samples were stored at 23 °C and 50% RH.

4.7 Surface Sizing

A solution of 20 wt % cationic starch and 5 wt% Basoplast® was used as surface sizing agent. An amount of 66.5 g of the cationic starch (~90.2 % solid content) was poured into a flask containing the corresponding amount of water. The solution was boiled at 100 ºC in a hot water bath with constant stirring. After 30 minutes, the solution was placed in an ice bath and rapidly cooled to 40 ºC. An amount of 15.0 g of Basoplast® was added to the starch solution and the resulting mixture was stirred for 1 hour.

Surface sizing was performed using a bench coater (RK Coater) with rods having various wire-wound diameters. The ionic paper was placed on top of the bench coater and an amount of the surface sizing agent was placed in front of the rod. The surface-sized ionic paper was dried using the STFI- IR dryer for 90 s at 110 ºC. The dried surface-sized ionic papers were stored at 23 ºC and 50% RH.

4.8 Screen Printing and Coating of Electrochromic Polymer

This step was undertaken in order to study the electrochemical performance of the ionic paper. Screen-printing of the ionic paper/surface-sized ionic paper was performed using an A4 60-mesh screen print (Screentec AB, Sweden). The Orgacon EL-3040 screen printing ink was printed and the printed paper was dried under ambient conditions. A mixture of 5 wt% DMSO and PEDOT:PSS was rod-coated on one side of the ionic paper, and dried at 110 ºC for five minutes using the STFI- IR dryer.

5 Interaction Between PEDOT:PSS and Cellulosic Material

The interaction of microcrystalline cellulose and the conducting polymer PEDOT:PSS has been studied in order to obtain a perspective on the adsorption of PEDOT:PSS onto the cellulosic material under various conditions. The aim has been to obtain an idea of how the conducting polymer behaves on the surface of a cellulosic material by determining the adsorption isotherm. It is also important to know how the pH and salt concentration affect the adsorption in order to be able to control the deposition behavior.

5.1 Adsorption Isotherm

The adsorption isotherm (Figure 5.1) shows that the maximum adsorptive capacity of treated MCC is higher than that of the untreated MCC, due to an increase in the available surface area after treatment of the MCC, due to the removal of residual hemicelluloses attached to the crystalline structure upon exposure to NaOH and subsequently to HCl. At a low concentration, the amounts of PEDOT:PSS adsorbed onto the two samples were similar, but the difference in adsorbed amount increased as the concentration increased. This is a common observation in the case of adsorption onto substrates with different surface areas. As the bulk concentration of PEDOT:PSS increased, the curve reached a plateau at a concentration of PEDOT:PSS in solution of about 250 ppm for the untreated compared to 400 ppm for the treated sample. The difference in adsorbed amount is apparent at higher concentrations.

3000 untreated MCC pre-treated MCC 2500

2000

1500

1000 Amount Adsorbed,µg/gAmount MCC 500

0 0 100 200 300 400 500

Equilibrium Concentration, ppm

Figure 5.1 Adsorption isotherm of PEDOT:PSS onto treated and untreated MCC at pH 3 and 23 °C.

5.2 The Effect of pH and Salt Concentration

PEDOT:PSS, which has negatively charged PSS ions surrounding the PEDOT, is considered to be a polyanion. Polyelectrolyte adsorption is a complex phenomenon due to the electrostatics of the polymer-adsorbent interaction. Polyelectrolytes, in general, are adsorbed onto oppositely charged adsorbents. However, it is also possible for adsorption to occur onto substrates carrying the same charge as the polyelectrolyte, provided that the non- electrostatic contribution to the adsorption energy is higher than the repulsive electrostatic potential (Parfitt and Rochester 1983). Figure 5.2a shows that the amount of PEDOT:PSS adsorbed decreases as the pH increases. Increasing the pH of the solution favors the deprotonation of carboxylic acid in the MCC structure, and thus renders the surface charge of the MCC particles more negative. When the surface of the MCC becomes negatively charged, strong repulsive forces develop between the cellulosic surface and the PEDOT:PSS, hindering the adsorption process. Figure 5.2b shows the specific surface charge quantity of both PEDOT:PSS dispersion and MCC suspension as a function of pH determined using polyelectrolyte titration. As the pH increases, the surface of MCC becomes more negatively charged.

3000 0 0 MCC -500 2500 batch 1 -10 batch 2 PEDOT:PSS -1000 q/g) ... -20 2000 -1500 -30 1500 -2000

-40 ...( µ eq/g) -2500 1000 -50 -3000 sp. charge charge PEDOT:PSSsp. Amount (µg/g)Adsorbed Amount

500 ( µ e charge MCC sp -60 -3500

0 -70 -4000 0 2 4 6 8 10 12 0 2 4 6 8 10 12 pH pH (a) (b)

Figure 5.2 Effect of pH on (a) the PEDOT:PSS adsorption onto treated MCC, and (b) the surface charge of PEDOT:PSS and MCC at 23 °C.

The deprotonation of PSSH to PSS- ions surrounding the PEDOT may also contribute to the negative charge of PEDOT:PSS as the pH increases. The charge variation of PEDOT:PSS between pH 4 and pH 10 is however very small because sulfonic acid is a strong acid having a pKa of about 2. Thus, the observed decrease in the amount adsorbed with increasing pH may be attributed to the increasing negative charges of both PEDOT:PSS and MCC, resulting in an increase in the repulsive forces. Batches 1 and 2 denote two different PEDOT:PSS storage times.

Adsorption is governed primarily by two mechanisms: electrostatic and non-electrostatic attraction. Previous studies reported the screening-reduced and screening-enhanced adsorption of polymers (van de Steeg 1992; Wågberg and Ödberg 1989). Figure 5.3a shows the amount adsorbed as a function of salt concentration for two batches of PEDOT:PSS with different storage times. Both batches showed a curve passing through a maximum value. The amount of PEDOT:PSS adsorbed increased with increasing salt concentration. This is a case of screening-enhanced adsorption, where the salt ions in the solution screen the repulsive forces and thus favor adsorption. The salt concentration also affects the size of the double layer of the suspended particles which acts as a resistance to the movement of PEDOT:PSS across the interface. Increasing the salt concentration decreases the size of the double layer and the adsorption barrier dramatically increases as the salt concentration decreases (Cohen et al 1997).

2500 100 batch 1 MCC 2000 batch 2 80 PEDOT:PSS

1500 60

1000 40 zeta potentialzeta (-mV)

Amount Amount Adsorbed, µg/g 500 20

0 0 -5 -4 -3 -2 -1 0 -5 -4 -3 -2 -1 0 log[NaCl] log[NaCl] (a) (b)

Figure 5.3 Effect of salt concentration on the (a) adsorption of PEDOT:PSS onto treated MCC and (b) zeta potential of PEDOT:PSS dispersion and MCC suspension at 23 °C.

Another possible explanation is a shrinkage of the polyelectrolyte as the salt concentration increases. Shrunken polymers can penetrate into the pores of the MCC, and this increases the adsorption. However, when the salt concentration increases, a maximum value is encountered and the adsorbed amount decreases. At relatively high salt concentrations, screening-reduced adsorption occurs where these ions build up a layer on the surface and thus reduce the diffusion of polyelectrolyte into the surface. Competitive adsorption of these ions onto the surface of MCC may also lead to “knocking off” of the polyelectrolyte and decreasing the adsorption site for the polyelectrolyte. The character of the curve shows the existence of a short-range non-electrostatic attraction between polyelectrolyte and surface with both screening–enhanced and screening-reduced adsorption regimes (van de Steeg 1992). Varying the salt concentration affects the specific charge of the suspension, as shown in the variation in the zeta potential in Figure 5.3b.

6 Electroconductive Paper

The bulk and surface of a electroconductive paper has been characterized. After the deposition of the conducting polymer, the paper has been changed from an insulating material to a semiconducting material. Two types of base papers have been used: laboratory- made sheets and commercially produced base papers. The laboratory-made sheets (100 g/m 2) were handmade sheets (isotropic), and dynamically (machine) formed sheets (anisotropic). In the case of the commercial base papers, two grammages were used: 100 g/m 2 (PM White from Billerud AB) and 150 g/m 2 (from Munksjö Paper). These commercial base papers exhibited different optical, mechanical, and other properties. However, their performance has not been compared since they have been used in different sub-studies. The following sections discuss the behavior of the conducting polymer in the fiber network, the factors affecting bulk conductivity, surface properties of the cellulosic material, and tensile property of the paper.

6.1. The Conducting Polymer in the Fiber Network

6.1.1 Scanning Electron Micrographs

SEM and EDS were used to study the deposition of PEDOT:PSS on the surface (Figure 6.1) and in the thickness direction (Figure 6.2) of the paper. SEM images were obtained from a field emission scanning electron microscope operated at an accelerating voltage of 15 kV. Samples were examined with a backscattered electron detector, providing atomic number contrast. The same field of view was then scanned using an energy dispersive X-ray spectrometer to acquire a set of X-ray maps for S, O, and C using 1 ms point acquisition until the maps showed sufficient X-ray counts. Since the base paper contained no traceable amount of sulfur, the sulfur detected is attributed to the sulfur from PEDOT:PSS.

The polymer had not covered the paper surface; instead, the sulfur content was higher at the fiber locations (Figure 6.1). In Figure 6.2b, the blue color is due to the sulfur signal, which can be attributed to PEDOT:PSS. The conducting polymer is distributed throughout the thickness of the paper and not just on the surface. This image was taken at the middle of the cross section.

Figure 6.1 SEM image (left) and corresponding EDS sulfur densit y map, with background subtraction (right), of an unrefined bleached kraft pulp handsheet coated with PEDOT:PSS (handsheet 100 g/m 2, in Paper III).

Figure 6.2 EDS map of the cross section of paper sample coated on both sides with PEDOT:PSS (dark red = epoxy resin, light orange = fiber, white = TiO 2, blue = PEDOT:PSS): (a) at 350 X magnification, and (b) at 750X magnification taken at the middle of the cross section (Munksjö Paper 150 g/m 2, in Paper II) .

6.1.2 PEDOT to PSS ratio

XPS analysis has been use d to determine the chemical composition of the PEDOT:PSS adsorbed on the cellulosic substrate. PEDOT:PSS is composed of water -insoluble positively - charged PEDOT surrounded by negatively -charged PSS. Two types of PEDOT:PSS were used with PEDOT:PSS weight ra tios of 1:2.5 and 1:6. The pH of the washing water was maintained at 2.0 because the observed removal of PEDOT:PSS at higher pH values could affect the XPS analysis. Samples 1, 3, and 4 had a PEDOT:PSS ratio of 1:2.5, sample 2 a PEDOT:PSS ratio of 1:6, and sample 5 was the reference base paper with no coating. Sample 1 with a bulk ratio of 1:2.5 PEDOT to PSS had a higher amount adsorbed on the

40 surface than sample 2 (1:6 ratio), which suggests that higher PSS doping leads to less adsorption on the cellulosic surface. A higher PSS level means more PSS - ions surrounding the PEDOT, and this makes the surface charge more negative and leads to higher repulsive forces. The functional groups identified in the C(1s) spectra give no further information on the distribution of PEDOT and PSS on the surface because both the cellulosic fiber and the polymer contain these functional carbon groups. Table 6.1 shows the relative surface composition of different atoms. It is clear that the amount of sulfur on the surface of sample 1 is greater than on sample 2. Interestingly, the fiber contains no traceable amount of sulfur and the S(2p) spectrum can be used to discriminate between the sulfur from PEDOT and the sulfur from PSS.

Table 6.1 Relative surface composition of different atoms in atomic % in five paper samples (Munksjö Paper 150 g/m 2, in Paper I).

Sample C O N S Na Ca 1 65.2 28.6 1.0 4.2 0.9 0.2 2 61.9 35.7 0.7 1.6 0.2 - 3 64.4 31.6 1.1 2.8 - (0.1) 4 65.3 30.1 1.0 3.3 0.2 (0.1) 5 61.0 39.0 - - - (0.1) Note: (1) no washing (1:2.5 PEDOT:PSS ratio), (2) no washing (1:6 ratio), (3) three washings (1:2.5 ratio), (4) one washing (1:2.5 ratio) and (5) reference (base paper).

Figure 6.3 shows the S(2p) core-level XPS spectra of all the samples where the interpretations are based on previously reported studies (Greczynski et al 2001; de Kok et al 2004; Xing et al 1997). Coated samples exhibit two peaks in the S(2p) spectra at 168-169 eV and 164-165 eV which correspond to sulfur signals of the oxidized form of sulfonate (in PSS) and thiophene (in PEDOT), respectively. The calculated atomic sulfur ratios of PEDOT:PSS for samples 1, 3 and 4 with PEDOT:PSS = 1:2.5 are 1:1.67, 1:1.65, and 1:1.55, respectively. This corresponds to a 13%, 14%, and 19% decrease in PSS on the surface. Sample 2 with PEDOT:PSS =1:6 has a measured atomic ratio of 1:3 on the surface which corresponds to a 35% decrease in PSS, i.e. there is a PEDOT enrichment on the surface of the dip-coated paper. The atomic sulfur ratios of samples 1, 3 and 4 having different degrees of washing showed that the PEDOT:PSS ratio on the surface was not affected by the washing. Investigations on thin spin-coated films of PEDOT:PSS using XPS analysis reported that PSS was enriched on the surface of the film (Greczynski et al 2001; de Kok et al 2004; Xing et al 1997) whereas the XPS results on dip-coated samples with a cellulosic substrate suggest PEDOT enrichment.

Figure 6.3 S(2p) core level spectra of five samples: (1) no washing (1:2.5 PEDOT:PSS ratio), (2) no washing (1:6 ratio), (3) three washings (1:2.5 ratio), (4) one washing (1:2.5 ratio) and (5) reference (base paper) (Munksjö Paper 150 g/m 2, in Paper I).

6.2 Some Factors Affecting the Bulk Conductivity

After the deposition of the conducting polymer blends into the fiber network, both the electroconductive and base papers have been characterized. Coating the paper with a conducting polymer blend changed the electrical properties of paper as well as some other properties, such as the optical properties, etc. In Figure 6.4, the difference between one- sided and two-sided coating can be seen. Some portions are not totally covered at the back of the one-sided coating. Three methods have been used to characterize the bulk conductivity of the base and electroconductive papers. These methods include two-probe (ASTM D 257) and four-probe techniques (ASTMD4496-04), and impedance spectroscopy. The ASTM D257 method was used to measure the conductivity of the base paper while the four-probe technique (ASTM D4496-04) was used to characterize the coated papers.

Impedance spectroscopy was used to measure the impedance of both the base and the coated papers, the paper sample being sandwiched between two circular gold electrodes.

Front Back Front Back a) (b) Figure 6.4 Paper samples coated with pure PEDOT:PSS dispersion: (a) one-sided coating and (b) two- sided coating (Munksjö Paper 150 g/m2, in Paper II).

6.2.1 The Influence of Fiber Beating and Sheet Forming

The bulk conductivity of the base papers and some other properties are listed in Table 6.2. The measured conductivities of the base paper were of the order of 10-12 S/cm, which is a typical value for a non-conducting paper (Josefowicz et al. 1982). The unbeaten handsheets (HS0) had higher bulk conductivity than the unbeaten sheets made using the DSF (DS0).

Table 6.2. Some properties of various laboratory-made base papers (dynamic /handsheets 100 g/m2, in Paper III).

Paper Thickness Density Drainage Bulk Conductivity Samples* (µm) (g/m3) Resistance (x 10 -12 S/cm) t sd ρ sd (ºSR) σ sd HS0 190.6 17.7 0.615 0.073 13.8 1.91 0.15 HS5000 116.9 3.1 0.928 0.038 45.5 4.61 0.33 DS0 188.0 4.1 0.602 0.015 13.2 0.73 0.17 DS1000 137.9 3.0 0.801 0.025 17.6 1.64 0.08 DS3000 123.5 2.6 0.903 0.039 25.7 1.73 0.05 DS5000 119.9 3.2 0.924 0.039 47.0 1.57 0.12 DS7000 113.8 1.7 0.961 0.027 67.3 1.08 0.38 DS10000 107.4 3.0 0.997 0.045 81.7 1.39 0.14 * HS: handsheet formed, DS: dynamic sheet formed, 0-10000: number of revolutions in the PFI mill; t – ave thickness;ρ- ave density; σ- ave bulk conductivity; sd- standard deviation

Pulp beating appeared to increase the conductivity of the paper compared with the equivalent unbeaten samples. Similar order of magnitude was observed for the conductivity of the commercial base papers using Impedance Spectroscopy, as reported in Papers II and IV. After two-sided coating with pristine PEDOT:PSS, the bulk conductivity of the base paper significantly increased from 10-12 S/cm to 10-2 S/cm (Figure 6.5), i.e. surface treatment of base paper with a PEDOT:PSS dispersion leads to a much higher conductivity than treatment of the pulp fibers prior to forming, as reported by some studies (Agarwal et al 2006; Lvov et al 2006; Wistrand et al 2007; Gerhardt et al 2008).

The performance of coated paper with beaten fiber is generally poor. Fiber beating creates a “hairy” fiber with a large surface area for hydrogen bonding with neighboring fibers (Clark 1985). The improved fiber packing reduces the interfiber voids, and this reduces the absorbency for the polymer dispersion, which is evident in the reduction in sulfur content with increasing fiber beating. From DS3000 to DS10000, the trend of conductivity reduction with decreasing PEDOT:PSS content is halted. The conductivity of these samples can be considered equal within the statistical uncertainties, despite a drop in sulfur content. This is due to the increased surface contact of the highly refined and well-packed fibers which decreases the distance between PEDOT:PSS molecules. This results in a conductivity increase, despite the drop in PEDOT:PSS content, (see sulfur content in Figure 6.5).

1,0E+00 8,0 bulk conductivity

sulfur content 1,0E-01 6,0

1,0E-02 4,0 Sulfur Content, paper dry Content, g/kg Sulfur

Bulk Conductivity, S/cm Conductivity, Bulk 1,0E-03 2,0

1,0E-04 0,0 HS0 HS0* HS5000 DS0 DS1000 DS3000 DS5000 DS7000 DS10000 Paper Samples

* - one-sided coating, the rest of the samples are coated on both sides

Figure 6.5. Bulk conductivity and sulfur content of coated paper samples. HS: handsheet formed, DS: dynamic sheet formed, 0-10000: number of revolutions in the PFI mill, *: substrate coated on one side only (dynamic /handsheets 100 g/m2, in Paper III).

6.2.2 The Influence of Calendering

Another factor that may affect the conductivity of a coated paper is calendering. In the papermaking process, the purpose of calendering is to adjust the thickness and modify the surface characteristics, i.e. gloss, smoothness, density, brightness and opacity (Holik 2006). Earlier investigations have claimed that calendering improves the conductivity of the paper (Casey 1961). In Figure 6.6, calendering the coated paper with a line load of 174 kN/m increased the conductivity, otherwise the conductivity remained comparable to that of the uncalendered samples. Applying a pressure in the rolling nip reduces the paper thickness considerably, and this decreases the distance between the PEDOT chains within the paper network. Since calendering the coated paper at 21 and 87 kN/m had no significant effect on the conductivity, it is possible that the coated paper returns to its original thickness after calendering or that the collapsed fibers hinder electron transfer. All the one-side coatings (no calendering or calendering before coating) exhibited similar conductivity behavior.

5 two-side coating no calendering one-side coating no calendering 21 kN/m before one side coating 4 21 kN/m 87 kN/m

3 174 kN/m S/cm -3 -3 , x10 , DC

σ 2

0 references before coating after coating

Figure 6.6. Bulk conductivity of paper calendered before and after coating. All samples were coated on both sides unless otherwise indicated (Munksjö Paper 150 g/m2, in Paper II).

6.2.3 The Influence of Organic Solvents

The effect on the conductivity of adding various organic solvents to the PEDOT:PSS dispersion is shown in Figure 6.7. Paper samples coated with blends containing NMP exhibited the highest conductivity followed by the blend containing DMSO. As the concentration of NMP in the bulk liquid increased, the conductivity of the coated paper also increased, but the DMSO blends exhibited a maximum value at 5 wt % in the bulk. On the

45 other hand, blends containing sorbitol and isopropanol showed no increase in the conductivity and increasing their content decreased the conductivity. 12,0 3 wt% 5 wt% 10,0 7 wt% one-side coating 8,0

S/cm two-side coating -3 6,0 , x10 , DC

σ 4,0

2,0

0,0 pristine Sorbitol DMSO NMP Isopropanol

Figure 6.7. Bulk conductivity of paper coated with various blends of PEDOT:PSS. All samples were coated on both sides unless otherwise indicated (Munksjö Paper 150 g/m2, in Paper II).

There are four reported mechanisms of conductivity enhancement: a plasticizing effect, a screening effect, a possible washing out of the PSS- ions, and a conformational change. The plasticizing effect occurs when secondary dopants such as high boiling polyalcohols facilitate the reorientation of the PEDOT:PSS chains at high temperature forming better connections of the conducting PEDOT chains (Ghosh et al 2001; Pettersson et al 2002). The screening effect is due to a reduction in the Coulombic interaction between positively charged PEDOT and negatively charged PSS dopant by residual polar solvents with a high dielectric constant (Lee et al 2003; Joo et al 2002). PSS- chains can be washed away from the surface of the PEDOT:PSS film during the film-forming process leaving a thin film with a high concentration of PEDOT on the surface of the film (Jönsson et al 2003). Lastly, PEDOT molecule may undergo a chemical structure transformation due to a conformational change from the benzoid to a quinoid resonant structure which has a higher charge-carrier mobility (Ouyang et al 2004; Ouyang et al 2005). Organic solvents with two or more polar groups may induce a conformational change. The driving force for this conformational change is the formation of a hydrogen bond of one polar group to sulfonate or sufonic acid groups while another polar group is very close to the PEDOT chain, leading to an interaction between the dipole of this polar group and the dipole moment or the positive charge on the PEDOT. The increase in the charge-carrier mobility due to the interaction of the secondary dopant and PEDOT:PSS is probably the reason for enhanced conductivity regardless of the mechanism of the conducting polymer-organic solvent interaction (Ouyang et al 2004; Ouyang et al 2005).

Among these four mechanisms, conductivity increase is probably due to a better connectivity of PEDOT grains along the fiber networks accomplished by the plasticizing effect of organic solvents or a possible washing out of PSS- chains from the PEDOT. The coated sample containing NMP exhibits the greatest increase in conductivity, as shown in Figure 6.7. NMP is known for its high solvent power which, when present in a coating, gives greater homogeneity and a higher coverage of coating on the surface after curing at high temperature. Conformational changes of the PEDOT molecules may also occur as a result of the interaction between PEDOT:PSS and the organic solvents, i.e. NMP and DMSO, transforming the benzoid to a quinoid structure. Coil and linear or expanded coil conformations are possible for PEDOT (Ouyang et al 2004). There was an observed shift of Raman peak in the presence of NMP in the coated paper (Figure 6.8). Also, a shoulder peak is observed in the case of the pure PEDOT:PSS, which implies that both benzoid and quinoid resonant structures are present according to a previous interpretation (Ouyang et al 2004).

Figure 6.8 Raman spectra of paper samples coated with NMP/PEDOT:PSS, DMSO/PEDOT:PSS and pure PEDOT:PSS at an excitation wavelength of 785 nm. Both NMP and DMSO have 7% (w/w) concentration. (Inset: benzoid and quinoid resonant structures) (Munksjö Paper 150 g/m2, in Paper II).

The presence of NMP may transform the benzoid structure to its quinoid structure, which has a higher charge carrier mobility. This transformation leads to a shift of the peak in the Raman spectra. This phenomenon has also been observed in PEDOT:PSS film with ethylene glycol as a conductivity enhancer (Ouyang et al 2005).

The screening effect is the least likely mechanism in this system because the dielectric constants of the solvents do not correlate with the observed conductivity. NMP has a lower dielectric constant than DMSO and sorbitol, but the paper with NMP added exhibited the highest conductivity. Charge screening by organic solvents on the electrostatic interaction between PEDOT and PSS may not be significant due to the presence of charged functional groups within the non-homogeneous cellulosic network. Washing out of PSS - chains may happen with or without the solvent because of the presence of space charges and permanent polar groups in the network that may interact with PSS. Thus, it is probable that a plasticizing effect and a conformational change of the PEDOT chains are the mechanisms for conductivity enhancement by NMP and DMSO on coated paper, which sorbitol and isopropanol do not exhibit.

6.2.4 The Influence of TiO 2 Pigments

Pigments such as TiO 2 are used in paper production as fillers that enhance brightness. Titanium dioxide is a semiconductor metal oxide with a band gap of about 3.1-3.2 eV (Nagliati et al 2006). The most important function of TiO2 in coating, however, is to provide whiteness and opacity to the finished paper due to its high refractive index which gives high scattering and effective coverage. Materials with TiO 2 have a high resistance to discoloration under UV radiation. Two types of TiO 2 pigments were used in this study: rutile and anatase. Both forms of TiO 2 films are used for various optical, electrical, photocatalytic, photovoltaic, and biosensor applications (Nagliati et al 2006, Es-Souni et al 2008). Addition of TiO 2 into a PEDOT:PSS dispersion decreased the conductivity of the coated paper (Figure 6.9), because the pigment particles may disrupt the path of the charge carriers in the bulk of the coated paper (Figure 6.10).

3,0

2,5

2,0 S/cm

-3 1,5 , x10 ,

DC 1,0 σ

0,5

0,0 pristine 1 pristine 2 anatase rutile

pristine 1 - PEDOT:PSS one sided-coating pristine 2 - PEDOT:PSS two-sided coating anatase - PEDOT:PSS + 1wt% TiO2 anatase two-sided coating rutile - PEDOT:PSS + 1wt% TiO2 rutile two sided-coating

Figure 6.9 Bulk conductivity of paper coated with PEDOT:PSS blends containing TiO 2 pigments (Munksjö Paper 150 g/m 2).

Figure 6.10 SEM/EDS image of surface of coated paper containing TiO 2 pigments (350X). [The blue color is attributed to the sulfur signal from PEDOT:PSS molecules while the whitish color is attributed to the 2 TiO 2 pigments] (Munksjö Paper 150 g/m ).

The coated paper with rutile TiO 2 exhibited a lower conductivity than that with anatase because the rutile form is bulky and consequently less PEDOT:PSS was deposited on the surface, as is evident in the sulfur content analysis (see Table 6.3). Comparing the one-sided coating and the sample with anatase, both having the same amount of PEDOT:PSS suggests that the presence of anatase enhances the conductivity.

Table 6.3 Total sulfur content of various coated paper samples (Munksjö Paper 150 g/m 2).

Component Sulfur Content (g/kg)* Pristine: one-sided coating 2.1 Pristine: two-sided coating 2.5 1 wt % TiO 2 anatase 2.1 1 wt % TiO 2 rutile 1.8 * results are given in dry substance and all samples have scatter of less than 6% from the mean value

6.2.5 The Influence of Multi-Wall Carbon Nanotubes

In Table 6.4, the addition of 5% DMSO and carbon nanotubes enhanced the conductivity of the coated paper. The coated paper containing 0.1 wt % MWCNT had a lower conductivity than the coated paper containing only PEDOT:PSS + 5 wt% DMSO. Increasing the dosage of MWCNT had a positive effect on the conductivity of the coated sheets.

Table 6.4. Bulk conductivities of paper samples coated with different blends of PEDOT:PSS (Nordic Paper 100 g/m 2, in Paper IV).

Coated Paper Samples Conductivity Standard x10 -3 S/cm Deviation PEDOT:PSS 0.475 0.133 PEDOT:PSS + 5% DMSO 0.638 0.099 PEDOT:PSS + 5% DMSO + 0.1% MWCNT 0.500 0.021 PEDOT:PSS + 5% DMSO + 0.3% MWCNT 0.727 0.024 PEDOT:PSS + 5% DMSO + 0.5% MWCNT 1.133 0.081

Another technique that is often employed in characterizing conductive sheets is Impedance Spectroscopy. It is a powerful technique for studying electrochemical systems and processes based on the small perturbation of electrochemical cell, e.g. applied voltage, with alternating signal of small magnitude that allows measurements at equilibrium or steady state (Unwin et al 2003). From the impedance data, the imaginary part can be plotted versus the real part of the impedance. This plot is known as a Nyquist plot (Cole-Cole plot) or the complex plane and is shown in Figure 6.11. The inset graph shows the effect of the amplitude perturbation on the shape of the Nyquist plot. In this case, the higher amplitude perturbations (1.0 and 2.0 V) showed unsteady behavior. Thus, the perturbation applied in the subsequent experiments was 0.1 V. From the Nyquist plot, the network of electrical circuit elements known as equivalent circuits can be determined based on the shape of the plot.

5000 PEDOT:PSS 3000 0.1 V PEDOT:PSS+ 5% DMSO 2500 1.0 V 2.0 V PEDOT:PSS + 5% DMSO + 0.1% MWCNT 2000 PEDOT:PSS + 5% DMSO + 0.3% MWCNT 4000 PEDOT:PSS + 5% DMSO + 0.5% MWCNT 1500

-Z"(Ohm) 1000

Cdl 500

0 0 2000 4000 6000 8000 10000 3000 R E Z' (Ohm) Rct

-Z" (Ohms) -Z" 2000

1000

0 0 2000 4000 6000 8000 10000 Z' (Ohms)

Figure 6.11 Nyquist representation of the impedance data of various coated papers at perturbation voltage of 0.1 V[Insets: equivalent circuit drawing, and Nyquist plot of PEDOT:PSS -coated paper as at different perturbation voltages] (Nordic Paper 100 g/m 2, in Paper IV).

10000 0

-10 8000

-20 6000

-30

4000 phase Angle phase -40

Complex Impedance (Ohms) Impedance Complex 2000 -50

0 -60

1,00E-01 1,00E+00 1,00E+01 1,00E+02 1,00E+03 1,00E+04 1,00E+05 1,00E+06 1,00E+07 1,0E-01 1,0E+00 1,0E+01 1,0E+02 1,0E+03 1,0E+04 1,0E+05 1,0E+06 1,0E+07

log frequenzy log frequency

PEDOT:PSS PEDOT:PSS + 5%DMSO PEDOT:PSS PEDOT:PSS+5%DMSO PEDOT:PSS+5%DMSO+0.1%MWCNT PEDOT:PSS+5%DMSO+0.3%MWCNT PEDOT:PSS+5%DMSO+0.1%MWCNT PEDOT:PSS+5%DMSO+0.3%MWCNT PEDOT:PSS+5%DMSO+0.5%MWCNT PEDOT:PSS+5%DMSO+0.5%MWCNT

Figure 6.12 Bode plot showing the magnitude of the impedance and phase angle of all coated paper samples (Nordic Paper 100 g/m 2, in Paper IV).

All the coated paper samples had equivalent circuits consisting of a contact resistance R E in series with a capacitance C dl connected in parallel to a resistor R CT (see inset drawing). The values of R E were between 110 and 150 Ω. The contact resistance values may be due to the roughness of the paper surface which affects the contact between the measuring gold electrodes and the paper. This is minimized by applying an adequate compressive force that compresses the paper (Baum 2002).

Among the coated paper samples, the sample containing 0.5 wt% MWCNT had the lowest bulk resistance, as can be seen from the distance of the line connecting the two x-intercept points in Figure 6.11 while the sample with pristine PEDOT:PSS had the highest bulk resistance. Figure 6.12 is the Bode plot showing the plot of the magnitude of the impedance and phase angle versus the logarithm of the frequency. This plot shows the behavior of the impedance throughout the frequency range which normally complements the Nyquist plot. It is apparent that the resistive behavior dominates for a wider range of frequency of the samples containing the MWCNT while the capacitive behavior of samples containing pristine PEDOT:PSS starts to dominate at a frequency of 1000 Hz. Increasing the amount of MWCNT in the dispersion led to an increase in the conductivity of the coated paper. MWCNT has a remarkable electronic property that makes it attractive for many applications (Dresselhaus et al 2001).

6.3 Some Changes in Surface Properties

6.3.1 Contact Angle

The wetting and hydrophobicity of the paper surface after different treatments has been investigated by measuring the contact angle. In principle, a liquid wets the surface of the paper only if its surface tension is lower than that of the paper. Both sides of the base paper were hydrophilic which means that water is easily absorbed. Samples coated with pure PEDOT:PSS on both sides were hydrophobic on both sides, whereas single-side coated samples were hydrophobic on the coated side (Figure 6.13). This hydrophobicity was not stable over time, since the contact angle decreased. Coating the base paper with pure PEDOT:PSS thus converted the hydrophilic to a hydrophobic surface. Since paper is a rough and porous material, contact angle measurements are sometimes difficult to interpret. Both surface roughness and hydrophobicity affect the contact angle. Contact angle decreases due to spreading on the paper surface and or absorption of water into the fiber network. The observed results may be because an increase in the PEDOT to PSS ratio on the surface makes the surface more hydrophobic since the PEDOT molecule is immiscible with water (see Section 6.1.2).

100

contact angle(θº) angle(θº) contact 20

0 0 5 10 15 Time (s) pedot:pss front basepaper front pedot:pss single coat front pedot:pss calendered before single coat front pedot:pss back basepaper back pedot:pss single coat back pedot:pss calendered before single coat back

Figure 6.13 Apparent contact angle of water on base paper and samples coated with pure PEDOT:PSS as a function of time (Munksjö Paper 150 g/m2, in Paper II).

A study of the adsorption of a multilayer of weak polyelectrolytes reported that hydrophobic surfaces are created by converting the outermost layer of polyelectrolyte into uncharged molecules (Rubner et al 1998). In Figure 6.14, the coated paper samples containing PEDOT:PSS had negligible spreading while the contact angle decrease of the sample containing NMP was due to spreading and absorption. Neither calendering the base paper before coating nor the presence of organic solvents had any significant effect on the contact angle.

Figure 6.14 Apparent contact angles as a function of time for water on paper samples coated with (a) pure PEDOT:PSS, and (b) PEDOT:PSS and 7 wt% NMP (t = time (s), θ = contact angle (º), w = base diameter (mm)) (Munksjö Paper 150 g/m2, in Paper II).

The effect of fiber beating on the apparent contact angle of water on various paper samples is shown in Figure 6.15. In general, the contact angle is influenced by the chemical nature and roughness of the surface (Yirdilim 1997). On a porous hydrophilic surface such as paper which consists mainly of cellulosic fiber networks, the contact angle decreases with time due to absorption of water into the paper. Fiber beating leads to a smoother paper surface (Clark 1985) with lower macroporosity, which contributes to a larger contact angle. The contact angle was more stable on the coated paper because the absorptivity was lower. The HS5000 sample had the highest contact angle of both the base and coated papers.

Figure 6.15 Apparent contact angle of water on various (a) base (uncoated) paper and (b) coated paper samples. HS: handsheet formed, DS: dynamic sheet formed, 0-10000: number of revolutions in the PFI mill (dynamic/handsheet 100 g/m2, in Paper III).

6.3.2 Porosity and Surface Area

Fiber beating increases the surface area, but in the case of chemical pulp, the “hairy” fibers come into close contact so that the surface area is reduced (Figure 6.16). The increased packing of the fibers does not affect the surface area, as the highly aligned and therefore highly packed DS0 samples have surface area values similar to that of the less well-packed HS0 samples. In these results, the HS5000 samples (base and coated papers) show an anomalous behavior. The effect of PEDOT:PSS coating on the surface area of the paper samples was inconsistent, which suggests that the polymer has not covered the entire surface of the fiber, as this would fill the fiber pores and void spaces, reducing the surface area of the substrate. The PEDOT:PSS deposition is potentially higher in fiber contact areas, where it does not affect the fiber structure.

The pore size distributions of various paper samples are shown in Figures 6.17. A dominant feature of these curves is the lower porosity of the beaten paper samples in the > 1.5-2 µm region. The pores in kraft papers can be divided into micro- and macro-pores, where micropores have diameters of ≤ 1.5 µm and consist of fiber pores, fine inter-fiber voids, and fine voids between fibers and fines materials, macropores have diameters greater than 1.5 µm and consist of larger inter-fiber voids, and uncollapsed pulp fiber lumens (Yamauchi et al. 1979, Yamauchi and Kibblewhite 1988). According to this division, fiber beating reduces the quantity of macropores because the fiber network becomes more consolidated, with more inter-fiber bonding. The highly beaten samples have the largest quantity of micropores within the range of 0.15 to 1.5 µm. The pore structure of the kraft pulp fiber walls opens up during refining, with a significant increase in volume in the region of the “larger” micropores (Laine et al. 2004). The increase in fiber pore volume with beating is evident in these samples despite the pore closure that occurs during drying due to hydrogen bonding (Wang 2006). There is a little difference in the mercury porosimetry data between the base paper samples and the PEDOT:PSS-coated samples. This follows the conclusions from the BET data that the polymer does not fill the substrate pores at either the macro- or micro-pore level. This reinforces the hypothesis that the PEDOT:PSS deposits on the fiber contact areas as previously suggested.

1,00 base paper coated paper 0,80 /g) 2

0,60

0,40

BET surface0,20 area (m

0,00 HS0 HS0* HS5000 DS0 DS1000 DS3000 DS5000 DS7000 DS10000 Paper Samples

Figure 6.16 BET surface area of base (uncoated) and coated paper samples. HS: handsheet formed, DS: dynamic sheet formed, 0-10000: number of revolutions in the PFI mill, *: substrate coated on one side only (dynamic/handsheet 100 g/m 2, in Paper III).

80 80 0Hb 1000Db 3000Db 0Hc 1000Dc 3000Dc 5000Hb 5000Db 7000Db 5000Hc 5000Dc 7000Dc 10000Db 10000Dc 60 60 /µm) 2 /m 3 /µm) 2

40 /m 40 3 dV/dr dV/dr (cm 20 20 dV/dr dV/dr (cm

0 0 0,01 0,1 1 10 0,01 0,1 1 10 Pore radius (µm) Pore radius (µm) (a) (b) Figure 6.17 Hg-porosimetry curves of (a) base (uncoated) and (b) coated paper samples. H: handsheet formed, D: dynamic sheet formed, 0-10000: number of revolutions in the PFI mill, c: coated paper (dynamic/handsheet 100 g/m 2, in Paper III).

6.4 Tensile Strength

The tensile strength of paper normally decreases when it is wetted and redried again because of disruption of the inter-fiber bonding. Both fiber beating and additive deposition appear to affect the paper strength. The mechanical strength of paper increases with fiber beating

(Clark 1985). Tensile strength is influenced by fiber strength, fiber length, and relative bonded area (RBA) according to the Page equation (Equation 3.1). The tensile index in the machine direction (MD) of various beaten and unbeaten paper samples is shown in Figure 6.18. The DSF samples have higher tensile index than the handsheets because of the effect of fiber orientation on the paper strength. Tensile index reached a maximum at 7000 revolutions beating. Thereafter, the tensile strength decreased due to the combined effect of fiber shortening and higher fibrillation, which weakens both the fiber and inter-fiber bonding (Uesaka et al. 2002, Ruel et al. 2006).

The unbeaten samples showed a similar strength for both the uncoated and coated samples, possibly due to PEDOT:PSS. The tensile index of beaten samples decreased slightly with coating, which implies that the disruption of inter-fiber bonding by the coating procedure dominates. At higher beating (DS10000), the lack of absorbency of the polymer by the substrate may reduce this inter-fiber disruption and the polymer support network compensates for the fiber beating strength loss, leading to an overall increase in tensile strength in comparison to the uncoated paper.

200

base paper coated paper

150

100 Tensile Index, Nm/g 50

0 HS0 HS0* HS5000 DS0 DS1000 DS3000 DS5000 DS7000 DS10000 Paper Samples

Figure 6.18 Tensile index of base (uncoated) and coated paper samples. HS: handsheet formed, DS: dynamic sheet formed (machine direction), 0-10000: number of revolutions in the PFI mill, *: substrate coated on one side only (dynamic/handsheet 100 g/m2, in Paper III).

In the case of the commercial base papers, a slight increase in tensile strength was generally observed after coating the base paper with a conducting polymer blend (Table 6.5 and Table 6.6) as compared to the laboratory-made sheet. This is particularly interesting since the commercial base paper has experienced web tension in the paper machine. The presence of

57 the conducting polymer in the fiber network may influence the fiber strength, and form a supported network that leads to a strength increase. The increasing dosage of MWCNT improved the tensile index, which could be due to the more effective load transfer from fiber to the adsorbed MWCNTs, known as the reinforcement effect (Hsieh et al 2010). This is a good indication that the deposition of conducting polymer blends increases paper strength. The addition of organic solvents and pigments to PEDOT:PSS does not affect the tensile strength of the coated paper.

Table 6.5 MD tensile strength of all paper samples (Munksjö Paper 150 g/m 2, in Paper II). Paper Samples a Tensile Strength (kN/m) c Organic solvent blends 3 wt% 5 wt% 7 wt% PEDOT:PSS - sorbitol 9.0 ± 0.2 8.5 ± 0.2 8.9 ± 0.1 PEDOT:PSS - DMSO 8.9 ± 0.2 8.6 ± 0.3 8.4 ± 0.3 PEDOT:PSS - NMP 8.7 ± 0.2 8.4 ± 0.2 8.2 ± 0.2 PEDOT:PSS - isopropanol 9.0 ± 0.2 9.0 ± 0.3 8.3 ± 0.2 Calendering Line Load b 21 kN/m 87 kN/m 174 kN/m Before coating 9.0 ± 0.2 9.3 ± 0.2 9.1 ± 0.2 After Coating 9.2 ± 0.2 9.2 ± 0.2 9.2 ± 0.2 One-side coating Two-side coating PEDOT:PSS 9.2 ± 0.2 9.0 ± 0.2 Base paper 8.0 ± 0.1 a All samples are coated on both sides except those described b Pure PEDOT:PSS coating formulation c Errors are calculated from standard error based on the number of samples prescribed by ISO 1924-02

Table 6.6 Tensile Index of various paper samples (Nordic Paper 100 g/m 2, in Paper IV). Paper Sample Tensile Index kNm/kg base paper 62.88 ± 0.85 PEDOT:PSS 71.92 ± 0.23 PEDOT:PSS+5%DMSO 69.47 ± 0.51 PEDOT:PSS+5%DMSO+0.1%MWCNT 66.44 ± 0.24 PEDOT:PSS+5%DMSO+0.3%MWCNT 69.87 ± 0.32 PEDOT:PSS+5%DMSO+0.5%MWCNT 73.12 ± 0.23

7 Ion-Conductive Paper

Ionic paper is a porous and flexible ion conductor that facilitates ionic movement useful in various electronic devices. Porous solid electrolytes have attracted interest because of their potential applications in sensors, electrochemical transistors, high energy batteries, and generally in large area electronics. The ionic liquid [bmim]BF4 was deposited onto the paper, and the paper became ion-conductive. The following sections discuss the factors affecting the conductivity of the ionic paper, surface sizing, and the demonstration of its electrochemical performance.

7.1 Ionic Paper

The ionic conductivity of the paper samples was measured using the four-probe technique at various voltages (Figure 7.1). This technique is suitable for conductive sheets and minimizes the effect of surface roughness. The ionic conductivity did not vary with the voltage applied between 10±1 and 100 ± 1 V. The subsequent measurements were made at 50 ± 1 V.

0,8

0,6

0,4 σ (S/cm) σ

0,2

0,0 0 25 50 75 100 Voltage (V)

Figure 7.1 Ionic conductivity of ionic paper at various voltages (PM White Billerud 100 g/m2, in Paper V).

7.1.1 The Effect of Relative Humidity

The electrical properties of paper are known to be greatly influenced by the moisture content in the atmosphere (Baum 2002). The influence of relative humidity on the ionic conductivity showed a steep increase in ionic conductivity from 25±5% RH to 50±5% RH but only a small difference from 50±5 to 85±5% RH (Figure 7.2). In these measurements, the

59 temperature was maintained at 23±1 ºC inside a climate chamber. It was important that that the ionic paper still possesses considerable ionic conductivity at both low RH (25%) and high RH (85%). Paper as such is very sensitive to moisture content in the atmosphere.

0,8

0,6

0,4 σ (S/cm) σ

0,2

0,0 0 10 20 30 40 50 60 70 80 90 Relative Humidity (%)

Figure 7.2 Ionic conductivity of ionic paper as a function of relative humidity (RH) (PM White Billerud 100 g/m2, in Paper V).

7.1.2 The Effect of Temperature

Ionic conductivity is also influenced by temperature variation. The temperature dependence of the ionic conductivity was fitted to the William-Landel-Ferry (WLF) relationship and to a simple Arrhenius equation for the temperature range between 296 and 323 K. The mechanism of ionic conduction in a solid polymer electrolyte, as expressed by the WLF, involves a segmental motion of polymer chains in the amorphous region (Wakihara et al 2002, Williams et al 1955). The temperature behavior of the ionic conductivity does not fit with the WLF. The Arrhenius equation is expressed as

þ ͯ Ĕ Ğ č ʚ͎ʛ Ɣ ͙ͤ û (7.1) where σ is the conductivity (S/cm), Ea is the activation energy (eV), kB is the Boltzmann constant, and T is the absolute temperature (K). The temperature dependence fits the Arrhenius equation well. This behavior follows the hopping model of carrier ions which is typical of inorganic ion conductors (Wagner and Kiukukola 1957). This indicates that the mechanism of ionic conduction is due to the inorganic anion BF4-. From the plot of ln σ(T) vs 1000/T in Figure 7.3, the activation energy, Ea is calculated to be 0.112 eV. The activation

60 energy is the minimum energy required to start the electrochemical process or charge transport.

0,0

-0,2 Ea = 0.112

-0,4

-0,6 ln σ(T) (S/cm) σ(T) ln -0,8

-1,0 3,0 3,1 3,2 3,3 3,4 3,5 1000/T (K-1)

Figure 7.3 Temperature dependence of the ionic conductivity of paper containing [bmim]BF4 (PM White Billerud 100 g/m2, in Paper V).

7.2 Surface Sizing

The ionic paper was surface-sized with a mixture of cationic starch and latex. Surface sizing of the ionic paper covers both the micro- and macro-pores of the paper and improves the printability of the surface. This is an important step to create a smoother surface and thus better printability of an electronic structure on top of the ionic paper. It is also important that the surface sizing is thin enough to allow fast diffusion of ions to and from the ionic paper. The presence of small openings in the form of pinholes or small uncovered areas on the surface may be desirable provided that they do not compromise the surface smoothness and printability. In order to see whether this had been accomplished, a standard test to evaluate the presence of pinholes on the surface of the paper was carried out using a standard mixture of an oil-soluble red dye, turpentine and anhydrous calcium carbonate. The mixture penetrates into the base and unsized paper, as evident by the wetting of the rear side of the paper. The surface-sized ionic papers, on the other hand, did not allow complete penetration of the dye mixture. A comparison of the front and reverse side images of the surface-sized paper showed red spots on the reverse side of the paper which are attributed to uncovered areas on the front of the paper (Figure 7.4). These spots are not only due to pinholes on the front side but rather to small uncovered areas.

(a) (b) Figure 7.4 Micrograph images (magnification: 20X) of surface-sized ionic paper (SIYSSG) (a) front, and (b) back after application of oil-based dye (Sudan Red) on the front (PM White Billerud 100 g/m 2, in Paper V).

Figure 7.5 Surface topography of (a) base paper, (b) ionic paper (SIY), (c) surface-sized base paper (SSG), (d) surface-sized ionic paper (SIYSSG), (e) surface-sized ionic paper (SIYSSB), and surface-sized ionic paper (SIYSSO), SSG = green, SSB = black, SSO = orange rod, scan size: 736 x 480 µm (PM White Billerud 100 g/m 2, in Paper V).

Three different wire rods (green = 0.31mm, black = 0.51 mm, orange = 0.76 mm wire diameters) were used in surface sizing to give three different thickness of the surface sizing film. A difference in surface topography of the various images can be seen in Figure 7.5. The unsized samples (Figures 7.5a and 7.5b) have higher number of deep portions (blue color) than the sized surfaces. This implies that many of the pores in the paper are covered by the surface sizing agent. The color of the fiber lines is more intense and defined in the unsized samples than in the sized samples. The quantitative evaluation of the roughness average (Ra), and the root mean square roughness (Rq) for at least twelve samples did not however show any significant variation among the samples.

The Nyquist plots for various paper samples show the effect of the thickness of the surface sizing on the electrical behavior (Figure 7.6). The general behavior of ionic paper, whether unsized or surface sized, corresponds to a faradaic impedance as a serial combination of charge-transfer resistance and the Warburg impedance. The Warburg impedance indicates a purely diffusion-controlled reaction at the low-frequency limit (Unwin 2003). As surface sizing increased, the bulk resistivity (RB) also increased. The sizing film in itself has a very low ionic conductivity. The values of the series resistance (RE) ranged from 116-124 Ω of the ionic papers (unsized and sized).

Cdl RE

RCT ZW

Figure 7.6 Nyquist plots of and the Randles equivalent circuit for the unsized and surface-sized ionic papers (PM White Billerud 100 g/m2, in Paper V).

7.3 Demonstration of Electrochemical Performance

In order to evaluate the ionic property of the ionic paper, an electrochromic ink PEDOT:PSS was screen-printed on the surface of the surface-sized paper. A mixture of 5 wt% DMSO and 95 wt% PEDOT:PSS dispersion was also coated on one side of an unsized ionic paper. Figure 7.7 shows possible electronic structures that can be deposited on the surface of ionic paper. It is possible to print an electrochromic device, an organic electrochemical field effect transistor or some other electronic structure on the surface-sized ionic paper.

In a demonstration of the electrochromic device, the application of a voltage causes the printed PEDOT:PSS to change color from transparent to blue depending on the redox state of the molecules. PEDOT:PSS is dark blue in its reduced state and light blue or nearly

64 transparent in its oxidized state. It is possible to change the contrast of the printing by applying a voltage (Figure 7.8a). In a demonstration of the electrochemical FET, a positive voltage was applied to the gate with respect to the ground source, and the printed gate turned dark blue which indicated a change in the conductivity of the printed channel (Berggren et al 2009). It was then possible to regulate the current that passed through the active polymer channel between the transistor’s drain and source terminals by electrochemically doping or de-doping the PEDOT:PSS according to the reaction (Nilsson et al 2005):

PEDOT + :PSS - + M + + e - PEDOT 0 + M:PSS

The color change was reversible since when the polarity of the voltage was reverse, the color changed back. That the electrochemical FET works on a surface-sized ionic paper was shown by the change in color of the printed gate channel to dark blue after application of a positive voltage between the gate and source ground as compared to the unchanged color of the gate printed on the base paper (Figure 7.8b). Ions diffused to and from the conducting polymer due to some open structures present on the surface of the surface-sized ionic paper, as discussed previously. The presence of uncovered areas explains the color switching of printed electrochromic polymer when the voltage is applied. This phenomenon was not observed when an electrochromic polymer was printed on the surface-sized base paper. Figure 7.9 shows images of the PEDOT:PSS-coated side of the unsized ionic paper. This phenomenon was observed when an electrode was in contact with the uncoated side. When a voltage was applied, the coated side exhibited color change at the spot where the electrode was positioned.

Figure 7.7 Schematic representations of possible electronic structures that can be constructed on top of the ionic paper surface.

(a) (b) Figure 7.8 Images of the possible potential devices demonstrator: (a) electrochromic demonstration, (b) organic electrochemical field effect transistor (PM White Billerud 100 g/m 2, in Paper V).

(a) (b) Figure 7.9 Images of coated PEDOT:PSS blends on one side unsized ionic paper: a) before , b) after application of voltage on the electrode on the uncoated side (PM White Billerud 100 g/m 2, in Paper V)

8 Conclusion

The deposition of a conducting material into the fiber network turned the paper into a conductive material. Paper is a versatile yet complex material. In the investigation of the interaction between conducting polymer and MCC, the adsorption isotherms for both pretreated and untreated MCC samples were found to be of the high affinity type with a broad molecular distribution of the conducting polymer. The adsorbed amount was dependent on both pH and salt concentration. As the pH of the solution increased, the adsorbed amount of PEDOT:PSS decreased due to the higher negative charges on both polymer and MCC leading to higher repulsive forces. The adsorption at different salt concentrations revealed the presence of both screening-enhanced and screening-reduced adsorption.

Paper was developed into an electroconducitve material by coating it with PEDOT:PSS. The conducting polymer was deposited on the fiber and fiber-fiber contact areas as seen in the SEM/EDS images. There was an enrichment of PEDOT on the surface of the paper, leading to higher contact angles of water on the coated paper than on the base paper. The degree of washing of PEDOT:PSS after dip coating did not significantly affect the PEDOT enrichment on the surface. Fiber beating and paper formation affect the conductivity of the PEDOT:PSS-coated paper. Beating the fiber improved the fiber packing, reduced the interfiber voids, and decreased the absorption of the polymer dispersion, which was shown by the reduction in sulfur content of the coated surface with increased fiber beating. Calendering the paper, did not generally affect the conductivity except with a line load of 174 kN/m. The addition of multi-wall carbon nanotubes enhanced the conductivity, but TiO 2 had little influence on the conductivity. The addition of NMP and DMSO to the dispersion enhanced the conductivity of the coated sheets, but the presence of sorbitol or isopropanol did not increase the conductivity. Four mechanisms of conduction enhancement by organic solvents have been discussed, including a plasticizing effect, a screening effect, a washing-out of PSS - ions, and a conformational change of the PEDOT molecule in the fiber network. The amount of PEDOT:PSS deposited in the fiber network, according to a total sulfur analysis of paper samples, suggested that the enhanced conductivity with the NMP blend is due to a better connectivity of PEDOT molecules along the bulk of the paper. A conformational change was also considered to be a possible reason for the conductivity enhancement, as supported by Raman spectroscopy.

When the ionic liquid [bmim]BF 4 was deposited on the paper, the paper became an ion- conductive material. The voltage used in the measurement of conductivity did not affect the calculated ion conductivity. The ionic paper was exposed to atmospheres with a relative

67 humidity varying from 25% to 85% at a temperature of 23 ºC, and it was found that the conductivity of the ionic paper was fairly high even at low relative humidity. The temperature dependence of the conductivity fitted well with a simple Arrhenius equation which followed the hopping model of carrier ions which is typical behavior of inorganic ion conductors. The calculated activation energy E a was 0.112 eV. Surface sizing was adopted to reduce the surface roughness and improve printability. The bulk resistance increased with increasing amount of applied surface sizing. The Nyquist plot of the ionic papers (both unsized and surface-sized) revealed a Randles circuit equivalent. The Faradaic impedance was a serial combination of a charge-transfer resistance and the Warburg impedance. The electrochemical performance of the ionic paper was demonstrated by printing PEDOT:PSS ink onto the surface and applying a voltage. The color change indicated a flow of ions from or to the ionic paper. It was thus demonstrated that the ionic paper is a good ionic conductor that can be used as a component for a compact electronic device.

This study presented a method of modifying paper into both an ion- and electron- conducting material. Various factors had been considered that may affect the conductivity and powerful analytical techniques have been used in order to understand the behavior of conductive paper. It was demonstrated that conductive paper has a potential to be used as a flexible and renewable component material for electronic devices.

9 Future Aspects and Challenges

“Paper as a component for electronic devices” is something which is a challenge in itself. Paper is a complex heterogeneous material consisting of a network of fibers. However, paper has been transformed into various types over the years. This is possible because paper is a versatile material which means that its properties can easily be modified to suit necessary functions. In this respect, researchers and papermakers face the challenge to venture more and take radical steps to utilize the knowledge acquired in this thesis. There are some fundamental issues that need to be addressed such as charge transport mechanism in a heterogeneous system, temperature and relative humidity variation of electroconductivity, and stability and diffusion dynamics of conducting materials in the fiber network. Other challenges include proper tuning of the surface property, faster time response, and cost reduction. Although the deposition method employed here is a conventional method used on a laboratory scale (batch process), the continuous production of conductive paper could be a challenge.

This research may possibly offer different possibilities. To the electronic engineer, it will bring a more compact architecture of flexible electronic structures. To the papermaker, it will make/convert paper into a new type of material that suits the challenge of the 21 st century. To the businessman, it will offer a reduction in the cost of producing electronic devices. To the environmentalist, it will reduce the utilization of fossil-based materials. To the customer, it will bring another type of product to buy and enjoy. To the researcher, it will bring another challenge for future investigations.

There are two possible fates of this research. It may either become one of those many researches piled up in the libraries which only a few people will dare to read or a radical innovation in the pulp and paper sector to be integrated into the human lifestyle like many other present paper products such as tissue papers, etc. Possible applications would be wall paper that may change color, interactive magazines, paper sensors, batteries, and other printed electronic devices on paper. The author hopes, of course, that the latter will take place.

Acknowledgements

I am indebted to the following for their valuable contributions to the completion of this work:

The Graduate School in Material Science (Sweden) and the Karlstad University Faculty of Technology and Science for the financial support;

My main supervisor, Assoc. Prof. Magnus Lestelius, for being supportive and very approachable, for your wise inputs on this project, and for believing in my capacity;

My assistant supervisor, Prof. Lars Järnström, for giving me advises on how to start with the project, and for your valuable comments and suggestions on the project;

Our project partners at Chalmers Tekniska Högskolan, Prof. Mats Andersson and Dr. Stefan Hellström, and the people from ACREO for the initial collaborations. Karen Love of Scion (New Zealand) for being the coauthor of Paper III; Siv Skoglund, Peter Lilja, Oscar Larsson, Patrik Karlsson, Marie Ernstsson, Birgitta Lindén, Olle Henningsson, Mehmed Mulabdic, Lloyd Donaldson, Filip Jedra, Christer Burman, Moheb Nayeri and Ulf Sohlin for help with some analyses; Hans-Olof Larsson, Henrik Kjellgren and Anna Jonhed for providing the base papers;

Assoc. Prof Anthony Bristow, for the language proof and for the valuable comments;

My colleagues and friends at the Department of Chemical Engineering, and at the Faculty of Technology and Science, for being so accommodating and for the laughter and fruitful discussions;

Filipino/Koinonia International Fellowship for making my stay here at Karlstad meaningful and fruitful. To Carl-Henrik Strömberg for helping me in some drawings and for the friendship. To the brethren at Lakas-Angkan Ministry, Laguna Philippines for the prayers;

To Wayne Ly and family for your prayers and friendship. Mark Harold Rivera for being a very supportive bestfriend, helping me in every way possible. To my bible study leader and mentor, Dr. Fred Sinohin and his wife Ate Vera, for your prayers and for teaching me true Christianity that defines my life now. To those not mentioned here who, in one way or another, have helped me;

My beloved family: Papa, Mama, Manoy, Ate, Kuya for loving me, believing in me, giving me support, and for inspiring me to strive more. Daghan kaayong salamat sa pagpadako sa ako nga puno sa disiplina ug paghigugma;

And above all, I give all the praise and glory to the only God and Savior Jesus Christ for giving me wisdom and strength, and the opportunity to meet those wonderful people mentioned above.

References

Agarwal M., Lvov Y., and Varahramyan K. (2006): Conductive Wood Microfibers for Smart Paper Through Layer-by-Layer Nanocoating , Nanotechnology, 17, pp. 5319-5325.

Ajayan P.M., Schadler L.S., and Giannaris S.C. (1998): Load Transfer in Carbon Nanotube Epoxy Composites, Appl. Phys. Lett., Vol. 73, No. 26, pp. 3842-3844.

Ajayan P.M., and Zhou O.Z. (2001): Applications of Carbon Nanotubes . In MS Dresselhaus, G. Dresselhaus, Avouris P.H., (eds), Carbon Nanotubes: Synthesis, Struture, Properties, and Applications (Springer-Verlag Berlin, Germany, pp. 391-425 (2001).

Ajayaghosh A. (2004): Organic Semiconductor Lecture Notes . Indian Institute of Technology Kanpur, Photoscience and Photonics Division, Regional Research Lab, CSIR.

Aleshin A., Kiebooms R., Menon R, Wudl F., and Heeger A. J. (1997): Metallic conductivity at low temperatures in poly(3,4-ethylenedioxythiophene) doped with PF6 , Physical Review B, vol 56, 7, pp.3659-3663.

Armanda M, Baril D., and Michot C. (1997): Electrochemistry of Liquids vs. Solids: polymer electrolytes , Solid State Ionics, 94, pp. 35-47(1997).

Attrup I and Hansen S. Pigments for On-line Surface Treatment (1997). Brander J and Thorn I (eds). In Surface Appplication of Paper Chemicals. London, Blackie Academic Professional, pp. 129-137.

Auhorn, W. (2006): Chemical Additives . H. Holik (ed). In handbook of paper and Board. Wiey- Verlag GmbH and Co. KGaA, Weinheim, pp. 148.

Ballarina B., Fraleoni-Morgerab A., Frascaroc D., Marazzitab S., Pianab C., and Settib L. (2004): Thermal Inkjet Microdeposition of PEDOT:PSS on ITO Coated Glass and Characterization of the Obtained Film , Synthetic Metals, 146, pp. 201–205.

Barrer R M. (1943): The Viscosity of Pure Liquids. II. Polymerised Ionic Melts , Trans. Faraday Soc. 39, pp. 59–67.

Basan G.C. and Wang S. (2008): Water Soluble Poly(fluorene) Homopolymers and Copolymers for Chemical and Biological Sensors. In Bernards D.A., Owens R.M., Malliaras G.G. (eds) Organic Semiconductors in Sensor Applications. Springer Verlag Berlin, Germany, pp. 1-38.

Baum GA.(2002): Electrical Properties: I. Theory . Jens Borch (ed) In Handbook of Physical Testing of Paper (chap 8).. Marcel Dekker Inc, New Yor, USA, pp. 361-388.

Berggren M., Bolin M.H., Svennersten K., Nilsson D., Sawatdee A., Jager E.W.H., and Richter-Dahlfors A. (2009): Active Control of Epithelial Cell-Density GradientsGrown Along the Channel of an Organic Electrochemical Transistor , Advanced Materials, 21, pp. 4379–4382.

Berggren M., Bobacka J., Svensson P-O, Nilsson D, Larsson O, and Ivaska A. (2008): PEDOT:PSS-Based Electrochemical Transistors for Ion-to-Electron Transduction and Sensor Signal Amplification. In Bernards D.A., Owens R.M., Malliaras G.G. (eds) Organic Semiconductors in Sensor Applications. Springer Verlag Berlin, Germany, pp. 263-280.

Bonard JM, Stöckli T, Maier F, de Heer WA, Châtelain A, Salvetat JP and Forró L. (1998): Field-Emission-Induced Luminescence from Carbon Nanotubes , Phys. Rev. Lett. 81, pp. 1441–1444.

Bonnamy S, Breton Y, Désarmot G, Salvetat JP, Delpeux S, Sinturel C, Béguin F. 2004): Mechanical Properties of Multiwall Carbon Nanotubes/Epoxy Composites: Influence of Network Morphology , Carbon, vol 42, 5-6, pp. 1027-1030.

Brännvall, E. (2009): Overview of pulp and paper process . Monica Ek, Göran Gellerstedt, Gunnar Henriksson (eds), In Pulp and Paper Chemistry and Technology, book 2, chap 1. Walter de Gruyter GmbH and Co, Berlin, Germany, pp. 1-11.

Brunauer S., Emmett P. H. and Teller E. (1938): Adsorption of Gases in Multimolecular Layers , J. Am. Chem. Soc., 60, pp. 309- 319.

Cao JM, Fang BQ, Wang J, Zhen MB, Gao SJ, Ma XJ. (2005): Ionic liquids for the Convenient Synthesis of Functional Inorganic Nanomaterials , Prog. Chem., 17 (6), pp. 1028-1033.

Casey J.P. (1961): Paper Testing and Converting , In Pulp and Paper Chemistry and Chemical Technology (2 edition), volume 3, Interscience publishers, Inc, New York, pp. 1468-1469.

Chiang C.K., Fincher C.R., Park Y.W., Heeger A.J., Shirakawa H., Loius E.J., Gau S.C., and MacDiarmid A.G. (1977): Electrical Conductivity in Doped Polyacetylene , Phys. Rev. Lett, 39, pp. 1098-1101.

Clark J.A. (1985): Pulp Technology and Treatment of Paper (2 nd ed), Miller Freeman Publications, Inc., USA, pp. 277-450.

Cohen, ED. (1992): Choosing the Coating Method . Cohen E. and Gutoff E. (eds), In Modern coating and drying technology.. Wiley-VCH, USA, pp. 1-21.

Cui Y., Hu L., Choi JW., Yang Y., Jeong S., La Mantia F., Cui L. (2009): Highly Conductive Paper for Energy Storage . PNAS, vol 106, no. 51, pp. 21490-21494. de Kok M.M. et al. (2004): Modification of PEDOT:PSS as a Hole Injecting Layer in Polymer LEDs , Phys. Stat Sol. A, 6, pp. 1342-1359.

Denneulin A., Blayo A., Bras U.J., and Neuman C. (2008): PEDOT:PSS Coating on Specialty Papers: Process Optimization and Effects of Surface Properties on Electrical Performances , Progress in Organic Coatings 63, pp. 87-91.

Dresselhaus MS, Dresselhaus G., and Avouris P. (eds). (2001): Carbon Nanotubes: Synthesis, Structure, Properties, and Applications. Springer-Verlag Berlin, Germany

Duan, H. M.; McKinnon, J. T. (1994): Nanoclusters Produced in Flames , Journal of Physical Chemistry, 98, pp. 12815–12818.

Dullien, FAL. (1992): Porous Media Fluid Transport and Pore Structure (2 nd ed) . Academic Press, San Diego California, USA.

Earle M.J., Seddon K.R., Esperanc J.M.S.S., Gilea M.A., Canongia Lopes J.N., Rebelo L.P.N., Magee J.W. and Widegren J.A. (2006): The Distillation and Volatility of Ionic Liquids , Nature , 439, pp. 831-834.

Ebbesen, T. W.; Ajayan, P. M. (1992): Large-scale Synthesis of Carbon Nanotubes , Nature, 358, pp. 220–222.

El Abedin ZS, and Endres F. (2006): Electrodeposition of Metals and Semiconductors in Air and Water Stable Ionic Liquids , Chem. Phys.Chem, 7, pp. 58–61.

Es-Souni M., Kartopu G.,Habouti S., Piorra A., Solterbeck C.H., Brandies H.F. (2008): Processing and Thin Film formation of TiO 2-Pt Nanocomposites , Phys. Sta. Sol. (a), 205(2), pp. 305- 308.

Feldmann R. (2006): Calendering . H. Holik (ed). In Handbook of Paper and Board. Wiley- Verlag GmbH and Co. KGaA, Weinheim, pp. 294-309.

Forró L, and Schönenberger, C. (2001): Physical Properties of Multi-wall Nanotubes . In MS Dresselhaus, G. Dresselhaus Avouris PH, (eds), Carbon Nanotubes: Synthesis, Struture, Properties, and Applications. Springer-Verlag Berlin, Germany, pp. 329-390.

Fort D.A., Swatloski R.P., Moyna P., Rogers R.D., Moyna G. (2006): Use of Ionic Liquids in the Study of Fruit Ripening by High-Resolution 13C NMR Spectroscopy: ‘Green’ Solvents Meet Green Bananas , Chem Commun. , pp. 714-716.

Ganske F. and Bornscheuer U.T. (2005): Lipase- Catalyzed Glucose Fatty Acid Ester Synthesis in Ionic Liquids , Org. Lett., 7 (14), pp. 3097-3098.

Gathergood N., Garcia M.T., Scammells P.J. (2004): Biodegradable Ionic Liquids Part I: Concept, Preliminary Targets and Evaluation , Green Chemistry, 6(3), pp. 166-175.

Gerhardt R.A., Peng C.Q., and Thio Y.S. (2008): Conductive Paper Fabricated by Layer-by-Layer Assembly of Polyelectrolytes and ITO Nanoparticles , Nanotechnology, 19, pp. 1-9.

Ghosh S., and Inganäs O. (2001): Nano-structured Conducting Polymer Network Based on PEDOT-PSS , Synthetic Metals, 121(1-3), pp. 1321-1322.

Giesche H. (2006): Mercury Porosimetry: A General (Practical) Overview , Part. Part. Syst. Charact. 23, pp. 9-19.

Goldstein J., Newbury D., Joy D., Lyman C., Echlin P., Lifshin E., Sawyer L., and Michael J. (2003): Scanning Electron Microscopy and X-ray Microanalysis (3 rd ed). Kluwer Academic/Plenum Publishers, New York, USA.

Greczynski G., Kugler T., Keil M., Osikowicz W., Fahlman M., Salaneck W. (2001): Photoelectron Spectroscopy of Thin Films of PEDOT:PSS Conjugated Polymer Blend: a Mini-Review and Some New Results , J Electron Spec. Rel. Phenom, 121(1), pp. 1-17.

Guo T., Nikolaev P., Thess A., Colbert D., and Smalley R. (1995): Catalytic Growth of Single- Walled Nanotubes by Laser Vaporization , Chem. Phys. Lett., 243, pp. 49–54.

Heeger A.J., Aleshin A.N., and Williams S.R. (1997): Transport Properties of Poly(3,4- ethylenedioxythiophene)/Poly(styrenesulfonate) , Synthetic Metals, 94, pp. 173-177.

Heeger A.J., Sariciftci N.S., and Namdas E.B. (2010): Semiconducting and Metallic Polymers. Oxford University Press Inc, New York, USA

Holik H. (2006): Handook of Paper and Board , Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany.

Hsieh YL, and Lu P. (2010): Multiwalled Carbon Nanotube (MWCNT) Reinforced Cellulose Fibers by Electrospinning , ACS Applied Materials and Interfaces, vol. 2, 8, pp. 2413–2420.

Iijima M. (1991): Helical Microtubules of Graphitic Carbon , Nature, 354, pp. 56-58.

Jastorff B., Störmann R., Ranke J., Mölter K., Stock F., Oberheitmann B., Hoffmann W., Hoffmann J., Nüchter M., Ondruschka B., and Filser J. (2003): How Hazardous are Ionic Liquids? Structure–activity Relationships and Biological Testing as Important Elements for Sustainability Evaluation , Green Chem., 5 (2), pp. 136-142.

Joo J., Kim J.Y., Jung J.H., and Lee D.E. (2002): Enhancement of Electrical Conductivity of Poly(3,4-ethylenedioxythiophene)/Poly(4-styrenesulfonate) by a Change of Solvents , Synthetic Metals 126, pp. 311-316.

Josefowicz J.Y. and Deslandes Y. (1982): Electrical Conductivity of Paper: Measurement Methods and Charge Transport Mechanisms , American Chemical Society, 25, pp. 493-530.

José-Yacamán, M.; Miki-Yoshida, M.; Rendón, L.; Santiesteban, J. G. (1993): Catalytic Growth of Carbon Microtubules with Fullerene Structure , Appl. Phys. Lett., 62, pp. 657-659.

Jönsson S.K.M., Birgerson J., Crispin X., Greczynski G., Osikowicz W., van der Gon A.W.D., Salaneck W.R. and Falhman M. (2003): The Effects of Solvents on the Morphology and Sheet Resistance in Poly(3,4-ethylenedioxythiophene)-Polystyrenesulfonic Acid (PEDOT:PSS) Films , Synthetic Metals, 139, pp. 1-10.

Kajanto I, Laamanen J, and Kainulainen, M. (1998): Paper Bulk and Surface (chap 3). In Niskanen, K. (ed) Paper Physics. Gummerus Printing, Finland, pp. 89-115

Kemerink M., Timpanaro S., Touwslager F.J., De Kok M.M., and Schrader S. (2004). Morphology and Conductivity of PEDOT/PSS Films Studied by Scanning Tunneling Microscopy , Chemical Physics Letters, 394, pp. 339–343.

Kimpimäki T and Savolainen AV. (1997): Barrier Dispersion Coating of Paper and Board . In Brander J and Thorn I (eds). Surface Appplication of Paper Chemicals. London, Blackie Academic Professional, pp. 208-238.

Khosla A., and Gray BL. (2010): Physical Properties of Polymers: Preparation, Micropatterning and Electrical Characterization of Functionalized Carbon Nanotube Polydimethylsiloxane Nanocomposite Polymer , Macromol Symp, 297, 210-218.

Laine, C., Wang, X. S., Tenkanen, M. and Varhimo, A. (2004): Changes in the Fiber Wall during Refining of Bleached Pine Kraft Pulp , Holzforschung, 58, pp. 233-240.

Lee C.S., Kim J.Y., Lee D.E., Koo Y.K., Joo J., Han S., Beag Y. W., and Koh S. K. (2003): Organic Based Flexible Speaker Through Enhanced Conductivity of PEDOT/PSS with Various Solvents , Synthetic Metals, 135-136, pp. 13-14.

Lee H, Kim KS. (2004): Physical and Electrochemical Properties of 1-Butyl-3-methylimidazolium Bromide, 1-Butyl-3-methylimidazolium Iodide, and 1-Butyl-3-methylimidazolium Tetrafluoroborate , Korean J. Chem. Eng., 21(5), pp. 1010-1014.

Leone E.A. and Signorelli A.J. (1996): Surface Analysis . In Sibilia J.P (ed). Materials Characterization and Chemical Analysis. USA. VCH Publishers, Inc, pp. 221-259.

Liu W.M, Hou XM, Zhou F, and Sun YB. (2007): Ultrasound-assisted Synthesis of Dentritic ZnO Nanostructure in Ionic Liquid , Mater. Lett., 61, pp. 1789-1792.

Lvov Y., Zheng Z., McDonald J., Khillan R., Su Y., Shutava T., and Grozditz G. (2006). Layer-by-Layer Nanocoating of Lignocellulose Fibers for Paper Properties , Journal of Nanoscience and Nanotechnology, 6, pp. 624-632.

Lu X.B., Hu J.Q., Yao X., Wang Z.P., and Li J.H. (2006): Composite System Based on Chitosan and Room-Temperature Ionic Liquid: Direct Electrochemistry and Electrocatalysis of Hemoglobin , Biomacromolecules,7, PP. 975-980.

Lyne MB. (2002): Wetting and the Penetration of Liquids into Paper . In Jens Borch (ed) Handbook of Physical testing of Paper (chap 9). Marcel Dekker Inc, New Yor, USA, pp 303-332

MacDonald J.R., and Johnson W.B. (2005): Fundamentals of Impedance Spectroscopy . In Barsoukov, Evgenij Macdonald, J. Ross (eds) Impedance Spectroscopy : Theory, Experiment, and Applications (2nd Edition ) Wiley, Hoboken, NJ, USA, pp. 1-27.

Marsh K.N., Deev A., Wu A.C.T., Tran E., and Klamt A. (2002): Room Temperature Ionic Liquids as Replacements for Conventional Solvents , Korean Journal of Chemical Engineering, 19(3), pp.357-362.

Marsh K.N., Pinkert A., Pang S., and Staiger M.P. (2009): Ionic liquids and their interaction with Cellulose , Chem. Rev, 109, pp. 6712-6728.

Matsunaga M. (2002): History and Properties of Ambient Temperature Chloride Melts , Electrochemistry, 70, pp. 126.

Mori M., Garcia R.G., Belleville M.P., Paolucci-Jeanjean D., Sanchez J., Lozano P., Vaultier M., Rios G. (2005): A New Way to Conduct Enzymatic Synthesis in an Active Membrane by Using Ionic Liquids as Catalyst Support , Catal. Today, 104, pp. 313-317.

Mosher R.H. and Davis D.S. (1968): Industrial and Specialty Papers , volume 1, Technology, Chemical Publishing Company, Inc, New York, pp. 170-172.

Murukami K, and Yamauchi T. (2002): Porosity and Gas Permeability . In Jens Borch (ed) Handbook of Physical testing of Paper (chap 6). Marcel Dekker Inc, New Yor, USA, pp 267-302.

Murr, L. E.; Bang, J.J.; Esquivel, E.V.; Guerrero, P.A.; Lopez, D.A. (2004): Carbon Nanotubes, Nanocrystal Forms, and Complex Nanoparticle Aggregates in Common Fuel-Gas Combustion Sources and the Ambient Air , Journal of Nanoparticle Research, 6, pp. 241–251.

Nagliati M; Carotta MC: Gherardi S; Lesci IG; and Martinelli G. (2006): TiO 2 Nanopowders for Sensing Applications; a Comparison Between Traditional and Hydrothermal Synthesis Way , Advances in Science and Technology, 5, 205-208.

Nelson F. (2007): Materials in Ink Jet Printing of Electronic Structures , REU Research Accomplishments, pp. 70-71.

Nilsson D., Robinson N., Berggren M., and Forchheimer R. (2005): Electrochemical Logic Circuits , Advanced Materials, 17, 3, pp. 353-358.

Niskanen, K. (1998): Paper Physics . Gummerus Printing, Finland.

Niskanen, K. , Kajanto I, and Pakarinen P. (1998): Paper Structure (chap 1). In Niskanen, K. (ed). Paper Physics. Gummerus Printing, Finland, pp. 14-53.

Nyström D., Lindqvist J., Ostmark E.; Hult A., Malmström E. (2006): Superhydrophobic Bio- fiber Surfaces via Tailored Grafting Architecture , Chem. Commun., 34, pp. 3594-3596.

Oliver J.F., and Mason S.G. (1976): Scanning Electron Microscope Studies of Spreading of Liquids on Paper. In F. Bolam (ed.) The Fundamental Properties of Paper related to its Uses. Br. Paper Board Ind. Fed., London, pp. 209.

Ouyang J., Xua Q., Chua C.W., Yang Y., Lib G., and Shinar J. (2004): On the Mechanism of Conductivity Enhancement in Poly(3,4-ethylenedioxythiophene):Poly(styrene sulfonate) Film through Solvent Treatment , Polymer, 45, pp. 8443–8450.

Ouyang J., Chu C.W., Chen F.C., Xu Q., and Yang Y. (2005): High Conductivity in Poly(3,4- ethylenedioxythiophene): Poly(4-styrenesulfonate) Film and Its Application in Polymer Optoelectronic Devices , Advanced Functional Materials, 15(2), pp. 203-208.

O’Sullivan J.B. (1947): The Conduction of Electricity through Cellulose. 1. The Conduction of Cellulose Impregnated with Salts , J. Textile Inst. 38, T271-T284.

Page D.H. (1969): A Theory for the Tensile Strength of Paper . TAPPI, vol 52, no 4, pp. 679-681.

Parfitt G.D. and Rochester C.H. (1983): Adsorption from Solution at the Solid/Liquid Interface , Academic Press: London: pp. 377-413.

Pettersson L.A.A., Ghosh S., and Inganäs O. (2002): Optical Anisotropy in Thin Films of Poly(3,4-ethylenedioxythiophene)–Poly(4-styrenesulfonate), Organic Electronics, 3(3-4), pp. 143-148.

Paulapuro H., and Gullichsen J. (2000): Papermaking Part 1, Stock Preparation and Wet End , Fapet Oy, Finland.

Pei Q, Zuccarello G, Ahlskog M, and Inganäs O. (1994): Electrochromic and highly stable Poly(3,4-ethylenedioxythiophene) Switches between Opaque Blue-black and Transparent Sky Blue, Polymer, 35, 1347-1351.

Pernak J, Zabielska-Matejuk J,Kropacz A, Foksowics-Flaczyk J. (2004): Ionic Liquids in Wood Preservation , Holzforschung, 58, 286– 291.

Perry JA. (1993): Soft-nip Calendering of Coated Paper and Paperboard . In Walter JC (ed) The Coating Process. Tappi Press, Atlanta Georgia, pp. 225-230.

Qian X., Song H., and Wang L. (2006): Conductive Paper Manufactured with the Composite of Polyaniline/Pulp Fiber by In Situ Polymerization , proceedings from the 3 rd International Symposium on Emerging Technologies of Pulping and Papermaking, Guangzhou, China.

Ragonese C., Tranchida P.Q., Sciarrone D., Mondello L. (2009): Conventional and Fast Gas Chromatography Analysis of Biodiesel Blends using an Ionic Liquid Stationary Phase , J. Chromatogr. A, 1216, pp. 8992–8997.

Rinzler A.G., Hafner J.H., Nikolaev P., Lou L., Kim G., Tománek D., Norlander P., Colbert D.T., Smalley R.E. (1995): Unraveling Nanotubes: Field Emission from an Atomic Wire , Science, 269, pp. 1550-1553.

Roberts J.C. (1996): The Chemistry of Paper , The Royal Society of Chemistry, UK

Rogers R.D, Swatloski R.P., Spear S.K., and Holbrey J.D. (2002): Dissolution of Cellose with Ionic Liquids , J. Am. Chem. Soc., 124, pp. 4974-4975.

Rubner M.F., Yoo D., and Shiratori S.S. (1998): Controlling Bilayer Composition and Surface Wettability of Sequentially Adsorbed Multilayers of Weak Polyelectrolytes , Macromolecules, 31, pp. 4309-4318.

Ryu S.K., and Jang J. (2006): Physical Propery and Electrical Conductivity of Electroless Ag-plated Carbon Fiber-Reinforced Paper , Journal of Materials Processing Technology 180, pp. 66-73.

Salaneck WR, stafström S, and Bredas J-L. (1996): Conjugated Polymers and Interfaces . Cambridge University Press, UK.

See K, Huang J., Becknell A., and Katz, H. (2008): Performance Requirements and Mechanistic Analysis of Organic Transistor-Based Phosphonate Gas Sensors. In Bernards D.A., Owens R.M., Malliaras G.G. (eds) Organic Semiconductors in Sensor Applications. Springer Verlag Berlin, Germany, pp. 213-244.

Shirakawa H., Ito T., and Ikeda S. (1974): Simultaneous Polymerization and Formation of Polyacetylene Film on the Surface of Concentrated Soluble Ziegler-type Catalyst Solution , J. Polym. Sci. Chem. Ed., 12, pp. 11-20.

Simula S. (2002): Electrical Properties: II. Applications and Measurement Methods . In Jens Borch (ed) Handbook of Physical testing of Paper (chap 9).. Marcel Dekker Inc, New Yor, USA, pp 361-388.

Skoog D.A., Crouch S.R., and Holler F.J. (2007): Principles of Instrumental Analysis , 6th ed. Cengage Learning, Kentucky:USA.

Skotheim T. (ed) (1986): Handbook of Conducting Polymers , volume 1, 2 Marcel Dekker: New York.

Smith E. and Dent G. (2005): Modern Raman Spectroscopy- a Practical Approach. John Wiley & Sons, Ltd, West Sussex, England

Springborg M., Schmidt K., Meider H., and De Maria L. (2001): Theoretical Studies of Electronic Properties of Conjugated Polymers. In Farchioni R. and Grosso G. (eds) Organic Electronics Materials. Conjugated Polymers and Low Molecular Weight Organic Solids by, Springer- Verlag Berlin, Germany, pp. 39-87.

Struik K.G and Visscher M. (1994): Optical Profilometry and Application to Mechanically Inaccessible Surfaces Part I: Principles of Focus Error Detection . Butterworth-Heinemann, vol 16 (3), Eindhoven, The Netherlands, pp. 192-198.

Unwin P.R. (2003): Instrumentation and Electroanalytical Chemistry . In Bard A.J., and Stratman M. (EDS) Encyclopedia of Electrochemistry, vol. 3. Wiley-VCH GmbH & Co. KGaA, Weinheim, Germany, van de Steeg H.G.M. (1992): Cationic Starches on Cellulose Surfaces: a Study on Polyelectrolyte Adsorption, PhD Thesis. University of Wageningen.

Veeco Metrological Group (1999): Wyko Surface Profilers Technical Reference Manual, USA.

Wagner C., and Kiukkola K. (1957): Measurements on Galvanic Cells Involving Solid Electrolytes , J. Electrochemical Society, 104, pp. 379-387.

Wakihara M., Kato Y., Hasumi K., Yokoyama S., Yabe T., Ikuta H., Uchimoto Y. (2002): Influence of PEG-borate Ester on Thermal Property and Ionic Conductivity of the Polymer Electrolyte , Journal of Thermal Analysis and Calorimetr , 69, pp. 889-896.

Walden P. (1914): Molecular Weights and Electrical Conductivity of Several Fused Salts , Bull. Russian Acad. Sci.: St Petersburg, pp. 405–422.

Wang H.L., Tang Z., Donohoe S.T., Robinson J.M., and Chiarelli P.A. (2005): Film Formation, Surface Character, and Relative Density for Electrochromic PEI/(PSS:PEDOT) Multilayered Thin Films , Polymer 46, pp. 9043–9052.

Wang Y, Wu J, and Wei F. (2003): A Treatment Method to Give Separated Multi-wall Carbon Nanotubes with High Purity, High Crystallization and a Large Aspect Ratio , Carbon, 41, pp 2939- 2948.

Wasserscheid P., van Hal R., Bösmann A. (2002): 1-n-Butyl-3-methylimidazolium ([bmim]) octylsulfate—an Even ‘Greener’ ionic liquid , Green Chem.,4, pp. 400-404.

Webb PA, and Orr C. (1997): Analytical Methods in Fine Particle Technology . Micromeritics Instrument Corp, USA

Wei C, Cho K, and Srivastava D. (2003): Tensile Strength of Carbon Nanotubes Under Realistic Temperature and Strain Rate , Physical Review B, 67, pp. 115407.

Wenner F., (1916): A Method of Measuring Earth Resistivity , Bur. Stand. (US) Bull., 12, pp. 469- 478.

Williams M. L., Landel R. F., Ferry J. D. (1955): The Temperature Dependence of Relaxation Mechanisms in Amorphous Polymers and Other Glass-forming Liquids , Journal of American Chemical Society, 77, pp. 3701-3707.

Wistrand I., Lingström R., and Wågberg L. (2007): Preparation of Electrically Conducting Cellulose Fibers Utilizing Polyelectrolyte Multilayers of Poly(3,4-thylenedioxythiophene)/ Poly(styrene sulphonate) and Poly(allyl amine), European Polymer Journal, 43, pp. 4075-4091.

Wågberg L. (2000): Polyelectrolyte Adsorption onto Cellulose Fibers: A Review , Nordic Pulp Paper Research Journal, 15, 5, pp. 586-597.

Wågberg L., and Ödberg L. (1989): Polymer Adsorption on Cellulosic Fibers , Nordic Pulp and Paper Research Journal, 4(2), pp. 135-140.

Wågberg, P and Johansson, P. (2002): Characterization of Paper Surfaces using Optical Profilometer . In Jens Borch (ed) Handbook of Physical testing of Paper (chap 9).. Marcel Dekker Inc, New Yor, USA, pp 429-450.

Xing K.Z., Fahlman M., Chen X.W., Inganäs O., and Salaneck W.R. (1997): The Electronic Structure of Poly(3,4-ethylenedioxythiophene) Studied by XPS and UPS, Synthetic Metals, 89, pp. 161-165.

Yamauchi, T., Murakami, K. and Imamura, R. (1979): Development of the Pore Structure of Paper Web During Consolidation , Mokuzai Gakkaishi 25, pp. 414-421.

Yamauchi, T. and Kibblewhite, R. P. (1988): Pore Structure of Paper Webs from Radiata Pine Thermomechanical Pulp , Appita J. 41, pp. 37-42.

Yao C. and Anderson J.L. (2009): Retention Characteristics of Organic Compounds on Molten Salt and Ionic Liquid-based Gas Chromatography Stationary Phases, J. Chromatogr. A, 1216, pp. 1658– 1712.

Yirdilim Erbil H. (1997): Surface Tension of Polymers , in Handbook of Surface and Colloid Chemistry, CRC Press LLC, USA, pp. 265-312.

Yu MF, Lourie O, Dyer MJ, Moloni K, Kelly, TF., and Ruoff, RS. (2000): Strength and Breaking Mechanism of Multiwalled Carbon Nanotubes Under Tensile Load , Science, 287 (5453), pp. 637–640.

Yuan L. (2001): Nanotubes from Methane Flames , Chemical Physics Letters, 340, pp. 237–241.

Yuan L. (2001): Ethylene Flame Synthesis of Well-aligned Multi-wall Carbon Nanotubes , Chemical Physics Letters, 346, 23–28.

Zanello L.P., Zhao B., Hu H., and Haddon R.C. (2010): Bone Cell Proliferation on Carbon Nanotubes , Nano Letters 3, 6 (3), pp. 562–567.

Zhang L.C., and Mylvaganam K. (2007): Ballistic Resistance Capacity of Carbon Nanotubes , Nanotechnology, 18, 475701 (4pp).

Zhao F, Wu XE, Wang MK, Liu Y, Gao L.X., and Dong S.J. (2004): Electrochemical and Bioelectrochemistry Properties of Room-Temperature Ionic Liquids and Carbon Composite Materials , Anal. Chem., 76, pp. 4960-4967.

Zhu S, Wu Y, Chen Q, Yu Z, Wang C, Jin S, Ding Y and Wu G. (2006): Dissolution of Cellulose with Ionic Liquids and its Application: a Mini-review , Green Chemistry, 8, pp. 325-327.

List of Standards

ASTM D 257: DC Resistance or Conductance of Insulating Materials ASTM D 4496-04: DC Conductance of Moderately Conductive Materials ISO 1924-3: Paper and board — Determination of Tensile Properties, Part 3: Constant Rate of Elongation Method (100 mm/min) ISO 5263-1: Laboratory Wet Disintegration, Part I: Chemical Pulps ISO 5264-1: Pulps. Laboratory Beating, Part II: PFI Mill Method ISO 5267-1: Determination of Drainability, Part I: Schopper-Riegler ISO 5269-1: Preparation of Laboratory Sheets for Physical Testing, Part I: Conventional Sheet Former Method SCAN_CM 57:99: Total Sulfur Content

Modification of Paper into Conductive Substrate for Electronic Functions

The thesis investigates the modification of paper into an ion- and electron-conductive material, and as a renewable material for electronic device. The study stretches from investigating the interaction between the cellulosic materials and the conducting polymer to demonstrating the performance of the conductive paper by printing the electronic structure on the surface of the conductive paper. Conducting materials such as conducting polymer, ionic liquids, and multi- wall carbon nanotubes were deposited into the fiber networks.

In order to investigate the interaction between the conducting polymer and cellulosic material, the adsorption of the conducting polymer poly(3,4-ethylenedioxythiophene): poly (4-styrene sulfonate) (PEDOT:PSS) onto microcrystalline cellulose (MCC) was performed. Electroconductive papers were produced via dip coating and rod coating, and characterized. The Scanning Electron Microscopy (SEM) / Energy Dispersive Spectroscopy (EDS) images showed that the conducting polymer was deposited in the fiber and in fiber-fiber contact areas. The X-ray Photoelectron Spectroscopy (XPS) analysis of dip-coated paper samples showed PEDOT enrichment on the surface. The effects of fiber beating and paper formation, addition of organic solvents and pigments (TiO2, MWCNT), and calendering were investigated. Ionic paper was produced by depositing an ionic liquid into the commercial base paper. The dependence to temperature and relative humidity of the ionic conductivity was also investigated. In order to reduce the roughness and improve its printability, the ionic paper was surface-sized using different coating rods. The bulk resistance increased with increasing surface sizing. The electrochemical performance of the ionic paper was confirmed by printing PEDOT:PSS on the surface. There was change in color of the polymer when a voltage was applied. It was demonstrated that the ionic paper is a good ionic conductor that can be used as component for a more compact electronic device construction.

Conductive paper has a great potential to be a flexible substrate on which an electronic structure can be constructed. The conduction process in the modified paper is due to the density of charge carriers (ions and electrons), and their short range mobility in the material. The charge carrying is believed to be heterogeneous, involving many charged species as the paper material is chemically heterogeneous.

Karlstad University Studies ISSN 1403-8099 ISBN 978-91-7063-361-4