Electrochemical Switching in Conducting Polymers – Printing Paper Electronics

(1)

Electrochemical Switching in Conducting

Polymers – Printing Paper Electronics

Payman Tehrani

Norrköping 2008

(2)

Electrochemical switching in conducting polymers – printing paper electronics Payman Tehrani Linköping Studies in Science and Technology. Dissertations, no.1212 Copyright © 2008 Payman Tehrani, unless otherwise noted. Printed by LiU‐Tryck, Linköping, Sweden 2008 ISBN: 978‐91‐7393‐801‐3 ISSN: 0345‐7524

(3)

Abstract

During the last 30 years a new research and technology field of organic electronic materials has grown thanks to a groundbreaking discovery made during the late 70’s. This new field is today a worldwide research effort focusing on exploring a new class of materials that also enable many new areas of electronics applications. The reason behind the success of organic electronics is the flexibility to develop materials with new functionalities via clever chemical design and the possibility to use low‐cost production techniques to manufacture devices.

This thesis reports different aspects of electrochemical applications of organic electronics. We have shown that the color contrast in reflective and transmissive electrochromic displays can be almost doubled by adding an extra electrochromic polymer. The choice of electrochromic material was found to be limited by its electrochemical over‐oxidation (ECO) properties, which is one of the main degradation mechanisms found in displays. The irreversible and non‐conducting nature of over‐oxidized films encouraged us to use it in a novel patterning process in which polythiophene films can be patterned through local and controlled deactivation of the conductivity. ECO can be combined with various patterning tools such as screen printing for low‐cost roll‐to‐roll manufacturing or photolithography, which enables patterning of small features. Studies have shown that electronic

conductivity contrasts beyond 107 can be achieved, which is enough for various

simple electronic systems. To generate better understanding of the ECO phenomenon, the effect of pH on the over‐oxidation characteristics was studied. The

results suggest that a part of the mechanism for over‐oxidation depends on the OH–

concentration of the electrolyte used. Over‐oxidation has also been used in electrochemical loggers, where the temperature and time dependence of the propagation of an over‐oxidation front is used to monitor and record the temperature of a package.

(4)

Dagligen kommer vi i kontakt med olika plastmaterial. Dessa har vanligtvis mycket dålig elektrisk ledningsförmåga och används oftast som isolerande material. Det finns dock en klass av plaster som är halvledande eller ledande. Sedan upptäckten av dessa material för mer än 30 år sedan har nya material och användningsområden utvecklats och nu börjar de första produkterna baserad på organisk elektronik komma ut på marknaden. En stor fördel med de ledande plasterna är att egenskaperna kan anpassas genom att ändra den kemiska strukturen. Man kan dessutom lösa upp dem och skapa ledande bläck, som sedan kan användas i vanliga tryckmaskiner. Detta gör det möjligt att på ett enkelt och billigt sätt tillverka elektronik på liknande sätt som till exempel tidningar trycks idag.

Den här avhandlingen behandlar en del av det nya området som berör elektrokemiska komponenter och några av dess tillämpningar. Fokus ligger främst på billig, tryckt elektronik. Bland annat presenteras ett sätt att fördubbla kontrasten för tryckta pappersdisplayer, ett nytt sätt att mönstra ledande plaster och elektrokemisk temperaturloggningsetikett som kan övervaka temperaturen för förpackningar under transport. Den mekanism som förstör ledningsförmågan vid höga spänningar har varit ett återkommande inslag i de studier som har genomförts här. Denna mekanism förstör komponenterna under drift men kan också användas för att ta bort ledningsförmågan som mönstringsmetod eller för att lagra information, permanent, i temperaturloggningsetiketten.

(5)

Acknowledgements

I would like to express my sincere gratitude to the people who have helped me reach this point, especially …

… my supervisor Magnus Berggren, for all the help and support during these years and for creating this very inspiring working environment.

… Thomas Kugler, who gave me the opportunity to start working in this research field.

… Nate, Xavier and Isak, for ALL the valuable and interesting scientific discussions. … Sophie for knowing everything and helping with all the administrative duties. … David, Peter, Joakim and the others that have worked with me in the lab, for being patient with my very impatient experimental technique.

… the people that I have worked with, especially the co‐authors of the included papers, who have contributed with their expertise to bring science one tiny step forward.

A really big thank to all the present and past members of the Organic Electronics group at ITN (Daniel, David, Edwin, Elias, Elin, Emilien, Fredrik, Georgios, Hiam, Isak, Jessica, Jiang, Joakim, Klas, Kristin, Lars, Linda, Magnus, Maria, Max, Nate, Olga, Oscar, PeO, Peter, Sophie, Thomas, Xavier, Xiangjun and Yu) for creating an amazing working environment (and relaxing coffee breaks).

I want to thank my family and friends for believing in me and giving me all the support that has brought me here. And of course thanks for all the good times outside of work.

I also want to thank the two most important persons in my life: My loving wife Yashma for supporting me in my work. Without you this would have been more difficult. Last but not least I want to thank my little Kayvan that takes a lot of my time but gives back so much more.

(6)

Paper 1: Patterning polythiophene films using electrochemical over‐oxidation

Payman Tehrani, Nathaniel D. Robinson, Thomas Kugler, Tommi Remonen, Lars‐Olov Hennerdal, Jessica Häll, Anna Malmström, Luc Leenders and Magnus Berggren

Smart Materials and Structures 14 (2005) N21–N25

Contribution: All the experimental work. Wrote the first draft and was involved in the final editing of the paper. Paper 2: The effect of pH on the electrochemical over‐oxidation of PEDOT:PSS films Payman Tehrani, Anna Kanciurzewska, Xavier Crispin, Mats Fahlman, Nathaniel D. Robinson and Magnus Berggren Solid State Ionics 177 (2007) 3521–3527 Contribution: Most of the electrochemical measurements. Wrote the first draft and was involved in the final editing of the paper. Paper 3: Evaluation of active materials designed for use in printable electrochromic polymer displays

Payman Tehrani, Joakim Isaksson, Wendimagegn Mammo, Mats R. Andersson, Nathaniel D. Robinson and Magnus Berggren

Thin Solid Films 515 (2006) 2485–2492

Contribution: About half of the experimental work (not including the synthesis of the polymers). Wrote the first draft and was involved in the final editing of the paper. Paper 4: Improving the contrast of all‐printed electrochromic paper‐displays Payman Tehrani, Lars‐Olov Hennerdal, John R. Reynolds and Magnus Berggren Manuscript Contribution: All the experimental work except the synthesis of the polymer and the roll‐to‐roll patterning of PEDOT:PSS. Wrote the first draft and was involved in the final editing of the paper.

Paper 5: Printable organic temperature logger based on over‐oxidation front propagation in PEDOT:PSS

Payman Tehrani, Isak Engquist, Nathaniel D. Robinson, David Nilsson, Mats Robertsson and Magnus Berggren Manuscript

Contribution: All the experimental work except the DSC measurement. Wrote the first draft and was involved in the final editing of the paper.

(7)

Related work not included in this thesis:

Printable All‐Organic Electrochromic Active‐Matrix Displays Peter Andersson, Robert Forchheimer, Payman Tehrani, and Magnus Berggren Advanced Functional Materials 17 (2007) 3074–3082 Device, kit and method for monitoring a parameter history US2008013595, EP1862786 Patterning PEDOT:PSS layer by controlled electrochemical reaction EP1916670 A device for integrating and indicating a parameter over time Patent pending

(8)

(9)

new materials and many different applications have been demonstrated. The question is how organic electronics will affect our lives in 30 years? The properties of this class of materials promise novel and exciting applications beyond the potential of traditional, silicon‐based, electronics. Organic conducting polymers allow … …simple and fast manufacturing using printing techniques …use of flexible and lightweight carrier substrates such as paper or plastic films …novel applications and devices …unique ways of adapting the device function by tuning the material properties though chemical engineering, which enables endless opportunities. Figure 1. The first silicon based transistor demonstrated in Bell Laboratories.

1 Introduction

(12)

Thanks to these benefits, new products have been developed and are now emerging on the market. The most mature technology is organic light emitting diodes (OLED) where several companies have released products for the consumer market. The first product that integrated an OLED display was the Philips shaver (Figure 2a) announced in 2002. Recently Sony has started selling their first generation of OLED displays as an 11” TV with extraordinary high contrast, superb color accuracy and close to 180° viewing angle (Figure 2b). These products are only examples of what can be achieved with this new class of materials and many other technologies are approaching the market soon.

Figure 2. Two examples of products using organic based displays. (a) A Philips shaver from 2002. (b) The Sony OLED TV XEL‐1, only 3 mm thick was released in the market in 2007. Copyright © Philips and Sony Corp.

Above, the background and the motivation for using conducting polymers in electronics were briefly reviewed. In chapter 2 the material and the basic principles for charge conduction in conjugated polymers are presented. Chapter 3 describes electrochemistry and its application to conjugated polymers. Chapter 4 gives an overview of the field of electrochemical organic electronics devices and a more in depth description of the work presented in this thesis.

(13)

Traditionally, polymers have been considered as poor electronic conductors and have extensively been used as insulators in electronics and electric power applications. In 1977, A. Heeger, A. MacDiarmid and H. Shirakawa demonstrated that chemical doping of trans‐polyacetylene films, a π‐conjugated polymer, can enhance its electrical conductivity by many orders of magnitude to almost reach the typical

conductivity values of metals[2]. Their finding has resulted in a new class of materials

offering novel applications for electronics such as flexible solar cells, light emitting devices, printed field effect transistors and electrochemical components made on

paper. In 2000, these gentlemen were awarded the Nobel Prize in chemistry[3] for

their discovery. The wide range of conductivity accessible in polymers originates from the diverse nature of the carbon atom in different chemical structures. The conductivity of a material is determined by the number of charge carriers and their inherent mobility. These parameters are governed by the chemical structure and the morphology of the polymer film. In this chapter, the relationship between the chemical structure and the conductivity is discussed, especially the charge carriers and the charge transport processes in the polymers.

2.1 π

-conjugated polymers

Although all organic materials are based on the same building block, the carbon atom, materials with organic compounds exhibit large variety of properties. Let us compare diamond and graphite, for instance. Both consist only of carbon atoms, but they have completely different electrical, optical and mechanical properties: Diamond is non‐conducting, transparent and considered as one of the hardest material known to mankind, while graphite is conducting, black and brittle. The difference can be explained by the different hybridization of the carbon atoms, an

(14)

atom which has four valence electrons in one 2s‐ and three 2p‐orbitals. In diamond,

the carbon atom is sp3‐hybridized, meaning that the electrons occupy four hybrid

orbitals made of a mixture of one 2s‐ and three 2p‐orbitals. Because of the symmetry of the four equal hybridized orbitals, the tetrahedral structure that is characteristic

for all sp3‐hybridized materials is created (Figure 3a). The electrons in the sp3‐orbitals

have a small extension in space and form very strong bonds (σ‐bonds) with

neighboring atoms. Therefore materials with sp3‐hybridization are stable and

electrically insulating as there are no free electrons in the system.

Figure 3. The four valence electrons in the carbon form symmetrical σ‐bonds to four other atoms in case of sp3_{‐hybridized material (a). In case of sp}2_{‐hybridized materials three}

symmetrical are formed with the last valence electron left in the 2p orbital orthogonal to the plane of the σ‐bonds (b).

In sp2‐hybridized materials, such as graphite, three of the four valence electrons are

in sp2‐orbitals, a mixture of one 2s‐ and two 2p orbitals, while the last electron is in

the remaining 2p‐orbital. The sp2‐electrons form strong bonds (σ‐bonds) with

neighboring atoms and create the backbone of the material, while the 2p‐electron forms π‐bonds with neighboring 2p‐electrons creating a double bond (together with

the σ‐bond). Because there are three symmetrical sp2‐orbitals, the sp2‐hybridized

(15)

CH A P T E R 2 – CO N D U C T I N G P O L Y M E R S

orthogonal to the plane of the sp2‐orbitals. In contrast to the sp2‐orbitals they are

delocalized and can reach far beyond the nearest neighboring atom. The sp2‐

hybridization in graphite results in a layered structure that gives softer mechanical properties as compared to the rigid diamond structure. In the following sections the

specific electrical and optical properties of sp2‐hybridized material are discussed.

Figure 4. As the number of carbon atoms increases, the overlap between the π‐orbitals cause states to split, eventually giving rise to continuous bands as the polymer extends into a long chain.

In a molecule with two sp2‐hybridized carbon atoms, the interaction between the 2p‐

electrons results in two molecular orbitals, one being of bonding nature with lower

energy level (π) and one being antibonding (π*) located at a higher energy level

(Figure 4). As the number of carbon atoms increases in a molecule, the number of molecular orbitals increases. In large molecules with N carbon atoms, the N 2p‐ electrons yield N molecular orbitals all with different energy levels. Because N electrons can occupy N/2 states, half of the orbitals are filled while the rest are empty. For electronic and optical applications, the energy difference between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular level (LUMO) is important, since it predicts the fundamental properties of the material. As the number of carbon atoms, N, increases, the energy difference between the levels decreases until they merge in continuous bands (Figure 4).

(16)

Now consider an infinite chain of carbon atoms which are sp2‐hybridized. That would give an infinite amount of atomic orbitals that are brought together and overlapping. Because of the energy splitting, the molecular orbitals have to be packed densely with infinitesimal energy difference between them, creating bands of states in the energy diagram. This would result in a one dimensional metal without a band gap (the energy difference between the HOMO and LUMO). But in practice this is not a stable state and a band gap is formed between the bonding and the antibonding

states. This can be explained by Peierls theorem[4]: He stated that a one‐dimensional

metal is unstable with respect to geometrical modification that leads to lowering of the symmetry and removal of the degeneracy of the HOMO and LUMO levels, thus obtaining a semi‐conducting state. The Peierls theorem applied on trans‐ polyacetylene (Figure 6a) yields as a result that the metallic state with delocalized electrons in the polymer chain (Figure 5a) is unstable. The polymer is stabilized through a geometrical distortion that results in a dimerization of the unit cell with alternating single and double bonds (Figure 5b). In this state the energy levels of the HOMO and LUMO are separated creating a band gap that is characteristic for semi‐ conductors. Therefore all conjugated polymer are semi‐conductors in their undoped state.

Figure 5. The structure of trans‐polyacetylene (a) results in a degenerated HOMO and LUMO state, which gives the polymer metallic properties. But according to Peierls theorem this state is unstable and is instead transformed into the more stable form with alternated double and single bonds in (b) and a band gap between the HOMO and the LUMO state.

(17)

2.2 Examples of conjugated polymers

S S S S S O O O O O O O O O O N N N H N H a b c e f d S S S S S N H N H N H N H N H Figure 6. Examples of some conjugated polymers (a) trans‐polyacetylene (b) polythiophene (c) poly(para‐phenylene vinylene) (d) polypyrrole (e) polyaniline (f) poly(3,4‐ ethylenedioxythiophene)

The simplest conjugated polymer, from a structure point of view, is trans‐ polyacetylene (Figure 6a), being a straight conjugated chain of carbon atoms, each atom bonded to one hydrogen atom. Although trans‐polyacetylene is easy to model in theoretical studies, the films are very sensitive for exposure to air and water, making it a poor candidate in applications. Through the years, different classes of more stable conjugated polymers have been developed (Figure 6). One of the characteristics that make polymers interesting is that their fundamental electronic and optical properties can be tuned by changing their chemical structure. One polymer that is widely used because of its stability and high conductivity is poly(3,4‐ ethylenedioxythiophene), PEDOT, a polythiophene derivative with a low band gap (Figure 6f). PEDOT itself is insoluble, but when chemically polymerized with poly(4‐ styrenesulfonate), PSS, as counter ion, a water emulsion can be obtained (Figure 7). In the PEDOT:PSS couple, the PEDOT part is positively doped making the polymer

(18)

S O O S O O S O O S O O S O O S O O S O O SO3Na SO3 SO3Na SO3Na SO3Na SO3 SO3Na + + Figure 7. The positively doped PEDOT having PSS as the counter ion.

2.3 Charge carriers

The π‐electrons in a neutral conjugated polymer chain are bound in the π‐orbitals giving rise to an alternation of single and double bonds. In this state the conjugated polymer has typical semiconducting properties. To conduct electricity, charges have to be introduced into the polymer (removing or adding electrons). The charge in the polymer chain is stabilized by altering the conjugation over several monomers (Figure 8). The charge together with the distortion of the structure of the polymer is denoted a polaron and can be either negative or positive. When two polarons form a couple, a so called bipolaron is generated with a charge of +2 (–2). S S S S S S S S S S S S S S S S S S S S S + + e -S S S S S S S S S S S S S S S S S S S S S e -

Figure 8. The doping of the polymer and creation of a positive (left) and a negative (right) polaron.

The number of free charge carriers can be increased through doping, for instance via charge injection from electrodes or via chemical doping. Chemical doping involves

(19)

charging the polymer film through chemical oxidation or reduction by dopant species. The dopant, with opposite charge, stays in the film and balances the charge of the polymer. In many electrochemical systems, the polymer film can be doped and dedoped reversibly via oxidation and reduction.

Figure 9. Positive polaron and bipolaron in a conjugated polymer. Overlap between charge carriers at high doping levels results in the formation of bands.

Upon doping, the charge carriers alter the structure of the polymer by increasing the length of the double bonds and shortening the single bonds, thus giving a more quinoidic character. This results in a decreased energy splitting between the HOMO and the LUMO levels, moving those states towards each other within the band gap (Figure 9). As the doping level increases, the number of states within the band gap increases. At high doping levels, the states within the band gap start to overlap and create bands of bipolaronic states (Figure 9).

2.4 Charge transport

In the solid state, polymer films are usually disordered with chains forming random coils. The π‐system of the conjugated polymers makes the polymer straight and rigid but chemical defects or torsion break the conjugation along the chain. In the conjugated sections of the polymer (I in Figure 10), the charge is transported through rearrangement of the bonds which is not as fast as in inorganic crystalline

(20)

semiconductors. The rearrangement is limited by the movement of nuclei that are about 100 times slower than the electron rearrangement. But still, the limiting step for charge transport is the hopping of charges in between chains and around defects present in the bulk (II in Figure 10). Figure 10. Transport of a polaron in a lightly doped film can be divided in intra‐chain transport (I) and inter‐chain hopping (II) In heavily doped polymer films, the polarons interact to form bands inside the band gap promoting transport of the charge carriers in the bulk. Inhomogeneity in doping level and morphology creates regions with poor conductivity between the highly conducting regions. For example PEDOT:PSS films contain highly doped “metallic” grains surrounded by regions with low concentrations of PEDOT chains diluted in PSS (Figure 11). Even though the electronic conductivity is high in the grains, the charge transport is limited by the hopping from grain to grain.

Figure 11. Charge transport in highly doped grains with a low concentration of conjugated polymer chains in between the grains. Inside the grains, charges are delocalized and migrate easily (I), while charges needs to hop between the grains passing over the low‐conducting phases (II).

(21)

2.5 Optical properties

The electronic structure of a material governs its optical properties as a result of the interaction between electrons and photons. In order to absorb a photon, an electron has to be excited to a higher available energy level corresponding to an increase of its energy state equal to the photon energy (Figure 12a). The light emission process is similar but reversed; a photon is emitted as an electron at a high energy level state is de‐excited to an available state with lower energy (Figure 12b). Many conjugated polymers have a band gap that matches the range of visible light (1.7 – 3.2 eV). This means that conjugated polymers may exhibit absorption and light emission in the visible region. In an undoped polymer, the band gap equals the lowest energy that can be absorbed. Upon doping, the formation of polarons and bipolarons changes the electronic structure and introduces new states within the band gap. These new electronic states allow absorption at lower photon energies, which drastically changes the absorption spectrum. As the doping level increases, the original absorption peak diminishes at the expense of the new polaronic and bipolaronic peaks. An example of this is given in the spectroelectrochemical measurement of Figure 13. Here the original peak at 2.3 eV, corresponding to transition between HOMO and LUMO, decreases as the doping level increases. Instead a new peak appears at 1.3 eV corresponding to formation of polarons in the film. When the doping is increased even further, the original peak at 2.3 eV disappears completely while the main absorption occurs at further lower energies within the infrared region, which indicates high concentration of bipolarons.

a absorption

b emission

Figure 12. In an absorption (a)/emission (b) process an electron increases/decreases its energy with the same amount that has been absorbed/emitted in a photon.

(22)

If the change in absorption occurs within the visible range, the human eye will interpret this as a change of color. This means that the color of the film can reversibly be controlled via the doping and de‐doping which is controlled by an electrical addressing signal. This effect is known as electrochromism and utilized in displays (section 4.2) 1.0 1.5 2.0 2.5 3.0 3.5 0.0 0.2 0.4 0.6 0.0 V 0.7 V 0.9 V Ab s

Photon energy (eV)

Figure 13. The graph shows the absorption spectrum of an electrochromic polymer (PProDOT‐ Hx2) at different applied potential which changed the doping level in the polymer film.

(23)

The doping level of conjugated polymers can be altered through reduction or oxidation of the polymer film in an electrochemical cell. As discussed in the previous chapter, a change in doping level results in a modification of the charge carrier concentration, thus changing the fundamental electronic nature of the material. This effect is utilized in simple electrochemical devices such as transistors and electrochromic displays. In this chapter the basic principles of electrochemistry in organic electronic materials are discussed.

3.1 Electrochemical cell

Chemical processes involving an exchange of electrons are denoted electrochemical processes. Conventional chemical reactions are controlled by the concentration of the chemical species and their cross section for spontaneous chemical reactions. In an electrochemical reaction the potential applied to electrodes adds another degree of freedom to control the reaction. Figure 14. A simple electrochemical cell with two electrodes and an electrolyte in between. The potential is applied so that the left electrode is reduced and the right electrode is oxidized.

(24)

An electrochemical setup, also called an electrochemical cell, consists of two electrodes and an electrolyte that connects them (Figure 14). When a potential is applied over the cell a current is able to flow through the cell and is a measure of the reaction rate. Each electrode and the surrounding electrolyte represent a half‐cell and therefore the reaction at one of the electrodes is called a half reaction. In order to have a closed circuit in the electrochemical cell (required for the flow of current), complementary reactions have to occur at the two electrodes. Reduction, where electrons are consumed (at the left electrode in Figure 14) requires complementary oxidation, where electrons are generated (at the right electrode in Figure 14). Under these conditions the current will flow clockwise in the electrochemical cell according to the scheme given in Figure 14.

3.2 Electrodes of conjugated polymers

An example of a half reaction in a conjugated polymer is the formation of a positive polaron through oxidation: − − + − _↔ ₊ + X P X e P : Here, P denotes an undoped site in the polymer and X – an anion. When the polymer

is charged the anion moves inside the polymer film to stabilize the positive charge (PP + − − + − + − + + − + ). This is referred to as electrochemical p‐doping. The n‐doping process is similar but here instead, cations react with the polymer electrode material at the same time as electrons are delivered to the material at the electrode: ↔ + + M e P M P :

Films of PEDOT:PSS includes immobile anions that are taking part in the p‐doping, therefore the half reaction is slightly different, moving cations in and out of the film to compensate the negatively charged PSS. + + ↔ + PSS M PEDOT PSS M e PEDOT : :

When a potential is applied between the electrodes in an electrochemical cell, the ions migrate in the electrical field that is created within the cell. At equilibrium, the

(25)

CH A P T E R 3 – EL E C T R O C H E M I S T R Y O F C O N J U G A T E D P O L Y M E R S

high concentration of ions close to the electrode screens the electrical field from the bulk of the electrolyte. This results in that the major part of the potential drop between the electrodes occurs at the electrode/electrolyte interface (Figure 15a). The layer with most of the potential drop is called the Helmholtz double layer and consists of negative (positive) charge carriers in the electrode and cations (anions) in the electrolyte. The double layer is equivalent to a capacitance with a thickness of ∼3‐ 5 Å, corresponding to the size of the involved ions. When a current passes through the cell, i.e. at non‐equilibrium, only part of the potential drop inside the cell occurs at the electrodes (Figure 15a). This means that a potential drop also occurs within the bulk of the electrolyte (Figure 15b) resulting in a flow of ions within the electrolyte. Figure 15. The potential profile within the electrolyte when a potential is applied. (a) shows the equilibrium case and (b) shows when a current is passing through the cell. In both cases most of the potential drop is over the Helmholtz layer at the electrode interface.

The dynamics of the electrochemical reaction can be limited by the reaction rate itself along the electrodes, the electrical conductivity of the film or electrode, or the conductivity of the electrolyte. In non‐crystalline polymers the film is often porous so that the electrolyte can penetrate into the film making the interface between the electrode and electrolyte effectively enormous and less well‐defined. This implies that the Helmholtz double layer that is built at the interface of the electrolyte and the electrode can be very large and in principle include parts of the bulk electrode. This suggests that the reaction rate is limited either by the electronic conductivity of the

(26)

film or by the diffusion/drift of ions in the electrolyte. Such limitations have been observed in two extreme cases (Figure 16) where in the first situation poly(3‐

hexylthiophene) (P3HT) is p‐doped in acetonitrile electrolyte[6] and in the second,

PEDOT is undoped using solidified gelled electrolytes. P3HT, in its undoped state, has a very poor electronic conductivity. In this case, as a potential is applied, the end of the film closest to the anode will start to be oxidized (and hence become doped). This will enhance the conductivity of the film allowing oxidation to occur further along the film. Eventually the entire film will be oxidized, as the front moves across the film starting from the anode. In the PEDOT:PSS system PEDOT is doped in its pristine state and therefore highly conducting. Using a gelled electrolyte with low water content gives a low ionic conductivity. In a cell including PEDOT:PSS as one of the electrode materials with a gelled electrolyte including a low water content, the ion diffusion in the electrolyte entirely dictates the switch characteristics. Therefore the reduction of the PEDOT:PSS film will start at the end closest to the counter electrode (the source of ions) and then move towards the cathode as a moving front.

Figure 16. (a) doping process of P3HT in aqueous electrolyte with high ionic conductivity. (b) undoping of PEDOT:PSS in a gelled electrolyte of low ionic conductivity.

Because PEDOT:PSS is partly doped from its synthesis, it is possible to further oxidize or reduce the film. Therefore films of PEDOT:PSS can be used as the counter electrode for either oxidation or reduction. When the working electrode is reduced the counter electrode has to be oxidized and since a 1:1 charge transfer is required, one reduced species at the working electrode requires one oxidized species at the

counter electrode. The doping fraction of pristine PEDOT:PSS films is very high[7] (for

the films used in our lab it is estimated to be about 80% of the maximum doping level). This means that the volume of the counter electrode should be at least four times larger than that of the working electrode to enable full reduction to occur at

(27)

the working electrode[8]. This fact is one key criterion in designing electrochemical

devices.

3.3 Electrochemical measurements

Material analysis can be performed using electrochemical measurements. For example the energy levels (ionization potentials) of a polymer can be estimated using relatively simple measurement equipment. A simple model of conjugated polymers in electrochemical measurements is sketched in Figure 17 where the p‐doping of an undoped polymer is displayed. N‐doping is similar but a negative potential is applied to the electrode giving that electrons are injected into the polymer at the LUMO level instead. Figure 17. A sketch of an electrochemical process in conjugated polymers. (a) The Fermi level of the metal electrode is in the band gap of the polymer. (b) When a positive potential is applied to an electrode, the Fermi level is decreased, allowing charge transfer when the Fermi level matches the HOMO level of the polymer, resulting in p‐doping of the polymer film. (c) The charge in the polymer film is stabilized by forming a polaron resulting in moving the energy levels inside the band gap.

In an experiment, where the potential is linearly swept cyclically around the p‐doping potential of a polymer, the current response would in principle be as in Figure 18. As the potential is increased the polymer is oxidized (p‐doped) and a positive current is measured. The width of the current peak is caused by the disorder in the polymer that causes the chains in the film to have varying oxidation potential. When the potential is decreased, the p‐doped polymer is reduced to the neutral state and a

(28)

negative current is measured. Note that the reduction potential is lower than the oxidation potential because of the relaxation described in Figure 17c where the energy level of the p‐doped species is higher and therefore will be reduced at a lower potential compared to the oxidation.

Figure 18. Cyclic voltammogram expected when p‐doping (and dedoping) a polymer. The oxidation and reduction occur at different potentials since relaxation in the polymer occurs as a consequence of electrochemical reactions.

This technique, where the potential is repeatedly swept up and down in voltage is called cyclic voltammetry (CV) measurement and is widely used to measure material characteristics and for evaluation of the electrochemical properties of compounds. In paper 2, CV measurements have been used to look at the oxidation and over‐ oxidation potential of various polymers. In paper 3, a similar experiment has been used, but instead of cycling the potential, only one linear sweep is measured, called linear sweep voltammetry (LSV).

3.4 Electrolytes

The electrolyte is the ionic conduction medium between the electrodes and can be in a liquid, solid or gelled state. Two examples of common solvents used in liquid electrolytes are water and acetonitrile. Water is polar and can easily dissolve common salts, but its stable potential window is quite narrow preventing accurate electrochemical measurements at higher potentials. In contrast, acetonitrile is more stable and offers a much wider potential window for the electrochemical

(29)

measurements. Because acetonitrile is less polar, salt with larger ions are preferred, for example tetrabutylammonium perclorate. Liquid electrolytes are suitable for electrochemical measurements, but in devices, solid or gelled electrolytes are preferred due to their non‐volatile nature. Solid electrolytes are usually based on ions mixed in an ion‐conducting matrix. For example the electrolyte system in paper 5 is based on polyethylene glycol (PEG) as the ion‐conducting matrix and Lithium

trifluoromethanesulfonate (LiF3CSO3) as the salt. Polyelectrolytes define another class

of electrolytes which also contain ions – either polycations with smaller anions as the counter ion or polyanions with small cations. An example of the latter is poly(4‐

styrenesufonate) (PSS) that contains immobile sulfonate (‐SO3–) ions and positive

counter ions, such as sodium (Na+) or protons (H+). Gelled electrolytes consist mostly

of liquid (usually water) and a solid network that gives the gel its mechanical properties. Because of the high liquid content they have normally high ionic conductivity.

3.4.1 Ion transport

For high ionic mobility, easy and fast transport of ions through the material is required. The transport mechanism for the ions depends on the type of electrolyte. In liquids the ions are surrounded by solvent molecules that follow the ions, creating a large package that has to be transported through the electrolyte. The viscous drag of this large package within the liquid electrolyte will then be the rate limiting factor for the ion transport. In solid electrolytes based on a polymer matrix, the flexibility of the chains in the amorphous phase of the material allows for the ions to be transported. Like walking through a crowded dance floor, the polymer chains make room for the ion to travel across the material. But this is not possible in a crystalline phase where the material is densely packed and does not have enough space to allow fast transport of ions.

3.4.2 Ionic conductivity

The current in an electrochemical cell can be assigned to two mechanisms: diffusion caused by a concentration gradient or drift due to an electrical field. The total current

(30)

c D V F c z diffusion drift+ =− =J J μ∇ − ∇ J where z is the charge of the ions (for example –1 or +2), c is the concentration, F is

the Faradays constant (∼96500 C/mol), μ is the mobility of the ions and D is the

diffusion coefficient which is proportional to the mobility. The applied potential creates a force that drives the ions in the direction of the field. This drift part of the current is intuitive, but the applied potential can also build up a concentration gradient, through accumulation of electrochemical reaction products. The result of the random movement of the ions is a net flow opposite to the direction of the concentration gradient, being the diffusion part of the current. Which mechanism that is dominant depends on the ionic conductivity of the electrolyte and the geometrical conditions of the setup. The ionic conductivity (κ) of an electrolyte can be calculated from the potential drop (V) along the distance (l) and the current density which is the current (I) that passes through the cross section area (A): A V I l ⋅ = κ Figure 19. A simplified geometry for measuring the ionic conductivity. I is the current through the cell, V is the potential difference at a distance l and A is the area of the cross section of the ion path.

One method for measuring the ionic conductivity is to utilize a four‐point‐probe

(31)

measurement is easier to analyze. This has been employed in paper 5 to measure the ionic conductivity of PEG based electrolyte at different temperatures.

3.5 Over-oxidation

At low potentials electrochemical switching of conjugated polymers is reversible, meaning that a large number of reductions and oxidations can be done repeatedly. However, degradation of the electroactivity has been observed at high anodic

potentials in polythiophenes[11‐13]. The phenomenon is called over‐oxidation and

reduces the potential window available for the reversible reactions. Over‐oxidation is an irreversible reaction that results in a non‐conducting and electrochemically inactive film as it breaks the conjugation in the polymer chains. In a study by Barsch

et. al. on the over‐oxidation phenomenon in polythiophene films[14], electrolytes based on acetonitrile with different concentrations of water were used to electrochemically switch polythiophene films. They observed that both the degree of over‐oxidation and the potential at which it occurs is highly dependent on the amount of water in the electrolyte. They proposed a mechanism that step‐wise degrades the conjugation in the polymer (Figure 20). This mechanism suggests that

first oxygen binds to the sulfur atom in the thiophene ring and then SO2 is detached

from the chain. After losing the conjugation in the polymer chain, the backbone of the chain is ripped apart in the last step. S S O S O O O O O O H O O OH Figure 20. Proposed mechanism for over‐oxidation of polythiophene[14].

(32)

Even though a thorough study of the over‐oxidation of PEDOT has not been published yet, it is reasonable to suspect that the over‐oxidation of PEDOT follows a similar

mechanism. Coulometry measurements on PEDOT suggest SO2 formation on the

thiophene rings[15]. Also the Fourier transform infrared spectroscopy measurements

included in paper 3 support this hypothesis.

The PEDOT films can be over‐oxidized in other ways, besides electrochemical, which lead to a degradation of the conductivity. Ultraviolet light in combination with oxygen is known to induce photo‐oxidation of the conjugated PEDOT material, resulting in a reduction in electrical conductivity. This photo‐oxidation leads to shorter conjugation lengths, due to sulfone group formation and chain scission accompanied by the

addition of carbonyl or carboxylic groups[7, 16]. Similar chemistry occurs upon heating

in air, where oxidative degradation of conjugated polymers has also been

reported[17]. Under these conditions, the conductivity of PEDOT decreases

exponentially with time according to an Arrhenius law[18] type of behavior.

Over‐oxidation is one of the main sources of failure in many devices and is therefore avoided as much as possible. The mechanism is irreversible and results in a non‐ conducting film phase which can be locally defined. Controlled electrochemical over‐ oxidation can be used as a method for deactivation of the conductivity in a subtractive patterning process (see further in section 4.1 and paper 1). Because of its irreversibility, the over‐oxidation front propagation can be utilized in electrochemical timer and logger (see further in section 4.3 and paper 5).

(33)

For more than 30 years, the field of conducting polymers has been under development. Major progress has been made in increasing the stability and functionality of materials for various applications. One of the most simple applications where conducting polymers are used is antistatic coatings, used for

example on photographic foils[5]. More sophisticated applications include new solar

cells[19], light emitting devices[20], field effect transistors[21] and different

electrochemical devices[22‐26]. In some applications, organic materials offer better

performance or new functionalities as compared to their inorganic counterparts, exemplified by superior performance of OLED displays compared to LCD displays when it comes to energy consumption, weight, color accuracy and viewing angle. In other cases simplified manufacturing gives huge benefits in comparison with existing inorganic technologies. For example, using flexible substrates and low‐cost manufacturability makes the organic solar cell interesting despite their low efficiency. Another technology where low manufacturing costs enable new markets is RFID (radio frequency identification) added onto packages and low cost surfaces. Such device systems can be realized with organic circuitry made out of organic field effect transistors (OFET). Several companies are now targeting such RFID products for track and trace applications for logistics.

One class of devices of special interest for this thesis is the electrochemical devices,

including electrochromic displays[5, 26, 27], electrochemical transistors[24, 28] and

actuators[29, 30]. Electrochemical devices are robust and operating at low voltage and

well suited for manufacture using printing techniques. In these devices desired nano‐ features/dimension are spontaneously formed and defined by the Helmholtz layer that reduces the significance of controlling the film thickness that otherwise is crucial

(34)

for the devices driven by charge injection or field effects. This makes it possible to manufacture fully printed devices, which will give very low manufacturing costs on for instance paper and other non‐ideal surfaces for electronics. A common drawback of these devices is limited ionic conductivity, which results in rather slow devices.

4.1 Electrochemical over-oxidation patterning

The benefits of organic electronics include new functionality, materials with low environmental impact and the possibility to use printing tools in the manufacturing process. Printing is an ancient technology that has been developed over the past centuries to be a fast and high volume production technique for reproducing text and images. When now using printing for manufacturing organic electrics, a well developed technique with little need for major modifications is available for high volume manufacturing of electronics. Today, many different techniques have been

demonstrated for printing and patterning conducting polymers, such as inkjet[21, 31, 32],

screen[33], offset[34], gravure[34] and flexographic[34] printing the most common being the inkjets and screen‐printing technology. These additive techniques can be used to print polymers and to create patterns down to resolutions of around 100 µm, which

is enough for large area displays[32, 33] and solar cells[35]. Coated polymer films often

possess better and more homogenous solid‐state properties compared with those printed through additive processes. The coated film can then be patterned through subtractive patterning techniques such as chemical deactivation at the polymer surface in which a chemical agent converts the conjugated material to a non‐active

material[36, 37]. These techniques are typically very slow and often require hazardous

reagents, but most importantly the pattern resolution is limited by diffusion of the agent into and along the polymer film. Plasma‐etching techniques can be used in combination with photolithography; however, these techniques require a vacuum and are typically not suitable for low‐cost roll‐to‐roll processing. Another technique for subtractive patterning of conductive films of PEDOT:PSS is to use electrochemical over‐oxidation (ECO). As described in section 3.5, over‐oxidation is an electronically controlled, irreversible process that results in a non‐conducting film. ECO can be utilized in combination with various patterning tools and offers an electrically

(35)

CH A P T E R 4 – AP P L I C A T I O N O F C O N J U G A T E D P O L Y M E R S

controlled deactivation method. It can be used together with photolithography to achieve high resolution µ‐patterns or be used in a screen printing step in order to produce patterns at high volumes in a roll‐to‐roll process. The conductivity ratio

between the conducting and insulating parts of a patterned film can reach 107, which

is enough for various simple electronics with relatively low packing density.

Figure 21. The Nilpeter printing press, within the Printed Electronics Arena Manufacturing, is used to print organic electronics. © Niclas Kindahl, Fotofabriken.

In recent years Acreo AB in Norrköping has, in collaboration with Linköping University, developed a process for a Nilpeter printing press (Figure 21) to produce electrochemical devices in a roll‐to‐roll process. The starting material is Orgacon™, which is a pre‐coated PEDOT:PSS thin film, here deposited on paper or plastic substrates, supplied by Agfa‐Gevaert in Belgium. Manufacturing of electrochemical devices then basically requires only three printing steps. First, subtractive patterning of the Orgacon™ through ECO patterning in a screen printing step results in a pattern with a resolution down to 100 µm at around 5 – 50 meters per minute. In the next

(36)

step, the electrolyte is applied in another screen‐printing step. Finally, the devices are encapsulated, by a plastic film, for mechanical protection. This Nilpeter printing press is small and simple but at 5 meters per minute it can produce more than 50 million devices (1 dm²) each year.

4.2 Electrochromic display

The color of conjugated polymers can be changed reversibly by controlling their doping level while they are included as electrode in an electrochemical cell. This effect is utilized in the electrochromic displays (ECD) where an electrical signal is used to change the color state of the display. ECDs can be used in different types of

applications such as in smart windows to control the transmission of sunlight[38], as

simple indicators in smart package applications or as the individual display pixel in

matrix addressed e‐paper[22, 39].

Usually, the electrochromic (EC) polymer is cast or coated on a conducting electrode, such as onto indium tin oxide (ITO) or gold, to achieve transparent or reflective device configurations, respectively. But if the electrochromic polymer has high intrinsic conductivity in both the doped and de‐doped states, there is no need for such conducting substrate electrode; the electrochromic polymer can be its own electrode. This gives a simplified device architecture that also is simpler to manufacture. For this reason polymers like PEDOT, with high intrinsic conductivity,

have received major attention in ECDs[26]. The ease of processing PEDOT:PSS has

encouraged scientists for instance at Acreo AB in Norrköping to develop an all‐ organic display on paper substrates in collaboration with the Organic Electronics

group at Linköping University[40]. These displays are flexible, lightweight and can

easily be manufactured into any shape or design. In Figure 22 seven‐segment displays are shown that have been demonstrated by LiU and Acreo. The PEDOT:PSS display has clear color switching between blue color in the reduced state and nearly transparent in the oxidized state.

(37)

CH A P T E R 4 – AP P L I C A T I O N O F C O N J U G A T E D P O L Y M E R S

Figure 22. Two seven‐segment digits made with ECD in PEDOT:PSS. The display is made with printable techniques by Acreo AB and Organic Electronics at Linköpings University. © Niclas Kindahl, Fotofabriken.

Even though the color switch contrast of PEDOT:PSS displays are satisfied for indicators and simple displays, higher contrast is desirable for improved performance in more advanced displays. Several routes have been presented for improving the optical contrast of PEDOT:PSS displays. Admassie et. al. introduced a grating pattern (µm‐scale) in the PEDOT:PSS film that through diffraction highly increased the

absorption of the PEDOT:PSS film compared to an unpatterned film[41]. With the right

material choice, this method can be used for improvement of the optical contrast. The most common method, though, is to add an extra EC polymer that complements the color of PEDOT:PSS. For significant improvement of the optical contrast, the dark state should complement the blue color of PEDOT:PSS and together cover as much as possible of the visible spectrum. At the same time, the uncolored state should be highly transparent at all optical wavelengths in order to provide nearly a translucent state.

The extra EC polymer material can be added to the ECD structure in different ways depending on its optical and electrochemical properties (Figure 23). When the EC polymer has opposite coloring polarity as compared to PEDOT, i.e. colored while oxidized and transparent when it is reduced, the extra EC film is best utilized as the counter electrode in vertical structures (Figure 23a). Contrary, if the EC polymer has the same optical “polarity” as PEDOT it is better included as a film on top of the active

(38)

PEDOT:PSS electrode, which can be either in a vertical (Figure 23b) or lateral structure (Figure 23c). Usually the rate limiting step in the displays is the ionic conductivity of the electrolyte, which means that a vertical display will switch faster and more homogeneously. However, a lateral structure offers a more simplified manufacturing process.

PEDOT PEDOT PEDOT

Electrolyte Electrolyte

Substrate Substrate Substrate

Substrate Substrate PEDOT PEDOT a b c EC‐polymer EC‐polymer EC‐polymer Electrolyte Figure 23. Different ways to improve the contrast in an ECD. In paper 3 and paper 4, the improvement of optical contrast is discussed. The work in paper 3 is focused on the design rules for choosing materials in contrast improvement of PEDOT:PSS displays. Here transmissive measurements were conducted with a standard absorption spectrometer using transparent plastic substrates (PET‐foils).

But the observed switch contrast values of the transmissive displays presented in paper 3 cannot be directly translated to reflective display. Paper displays are interesting from a device point‐of‐view, therefore the work presented in paper 4 was angled towards devices and important parameters used in applications. ECDs with paper substrates manufactured in the printing press were modified by adding an extra layer of an EC‐polymer. Both contrast and switching speed measurements was based on reflective setups.

There are several practical issues that put constrains on the design of printable EC devices. The main problem is to choose an electrolyte that has high enough ionic conductivity, to achieve reasonably fast switching, is printable and also gives proper mechanical stability. The electrolyte should also maintain its high ionic conductivity even after long storage times in dry conditions. This turns out to be a great challenge as high ionic conductivity usually requires high water content, which eventually will evaporate in dry storage conditions. One route could be to use encapsulation that

(39)

CH A P T E R 4 – AP P L I C A T I O N O F C O N J U G A T E D P O L Y M E R S does not allow the drying out of the electrolyte, but water molecules are small and can go through most materials. Only metal films could give the desired protection but those are not transparent and cannot be used for displays. Different plastic films are not dense enough but could be noticeably improved by using additives such as clay nano‐particles. Scientists are working hard to improve the barrier function of plastic films but the materials are expensive and still not good enough as a barrier for water molecules. Another route could be to develop electrolyte materials with high ionic conductivity also at low humidity levels and only use the encapsulation as a mechanical protection.

Another issue that has to be considered as well is the design of the counter electrode. As mentioned earlier any electrochemical half reaction must be balanced in the counter electrode. When using PEDOT in the counter electrode, care must be taken to ensure that enough redox sites are available for active electrode to be fully switched. In a lateral design with the same film thickness, the area of the counter electrode should be at least four times larger than the active pixel (see section 3.2).

4.2.1 CIE Lab coordinates

The light that escapes from an illuminated or luminescent object is nothing more that a spectrum of photons of various wavelengths and intensities. When that light hits our eyes, it is absorbed by the rods and cones on the retina, transformed into nerve signals and translated into a perception of color in the brain. Because humans have a three‐dimensional color space (given by the three different cones that are sensitive to different colors), many of the existing color systems have three coordinates. In light emitting devices, three pixels with red, green and blue (RGB) color are used to span the color space. The RGB color system is practical for creating a display but is actually not a good description of the full color space accessible with human eyes. In 1931, the International Commission on Illumination (Commission Internationale de l'Eclairage, CIE) developed a model defining colors by three coordinates by X, Y and

Electrochemical Switching in Conducting Polymers – Printing Paper Electronics