Long-Term Stabilization of Organic Solar Cells Using Additives

Dissertation zur Erlangung des akademischen Grades

Doctor rerum naturalium

Vorgelegt der Fakultät für Mathematik und Naturwissenschaften der Technischen Universität Ilmenau

von

Dipl.-Ing. Vida Turkovic

Betreuender Hochschullehrer: Prof. Dr. Gerhard Gobsch 1. Erste Gutachter: Prof. Dr. Gerhard Gobsch 2. Zweite Gutachter: Prof. Dr. Jürgen Parisi

Tag der Einreichung: 15 April 2014 Tag der wissenschaftlichen Aussprache - nichtöffentlicher Teil (Rigorosum): 7 October 2014 - öffentlicher Teil (Disputation): 8 October 2014

urn:nbn:de:gbv:ilm1-2014000324

ii

If an alien arrived on Earth and saw all this sunlight, he’d be amazed to hear that we think we’ve got an energy problem.

-- Ian McEwan, Solar

The work described in this thesis was performed in the research group Experimental Physics I at Ilmenau University of Technology, Germany. The author was financially supported by the State of Thuringia and the Federal Ministry of Education and Research (BMBF).

Published by: Vida Turkovic Copyright © 2009-2014 by: Vida Turkovic

iii

Abstract

Organic solar cells have a high commercial potential, although certain issues still have to be resolved. Big efforts have already been invested into perfecting the constituent materials with regard to efficiency, and the current peak efficiencies lie at 12 %, as reported by Heliatek on vacuum deposited oligomer tandem solar cells. However, the stability of devices remains their weak point. Due to their organic nature, organic solar cells are extremely sensitive to light, heat, oxygen and humidity, all of which are commonly present in standard working conditions.

In my thesis I examine photooxidative stabilization by introducing additional stabilizing compounds into the active layer. A variety of compounds with different stabilizing mechanisms, including radical scavengers, hydrogen donors, UV blockers and hydroperoxide decomposers, were tested in the model active layer system consisting of poly(3-hexylthiophene-2,5-diyl):phenyl-C61-butyric acid methyl ester (P3HT:[60]PCBM). Such devices were aged in lifetime setup at 1 Sun illumination at moderate temperature and in presence of air. A tremendous increase in accumulated power generation, by over a factor of 3, was achieved with stabilizers of hydrogen donor and UV absorber classes, which was attributed, via spectroscopic and microscopic studies, to a reduction in formation of free radicals.

These results underscore the potential of compounds known to successfully stabilize insulating plastic for stabilization of organic solar cells. However, unlike in the case of insulating plastics, additional care has to be given that the energy levels of the additive are not overlapping with the effective band gap of the photoactive layer and that the resulting blend morphology does not disrupt the photoinduced charge transfer.

Using P3HT:[60]PCBM as the model system, it could be demonstrated that ternary blending photoactive layers of organic solar cells with antioxidants and UV absorbers can be applied to substantially prolong the lifetime of devices. This approach could easily be extended to other polymer:fullerene systems, under the condition that its presence does not influence the morphology. These findings hold a great importance for the commercialization of organic solar cells, as they offer a route to achieving highly reliable devices by adding small portions of cheap stabilizer compounds, regardless of the intrinsic photochemical stability of the polymer and the fullerene.

iv

Kurzfassung

Organische Solarzellen bestechen durch hohes wirtschaftliches Potenzial, obwohl bestimmte Fragen noch geklärt werden müssen. Große Anstrengungen wurden bereits auf die Perfektionierung der verwendeten Materialien mit Hinblick auf die Effizienz der hergestellten Solarzellen unternommen. Aktuelle Spitzen-wirkungsgrade liegen bei 12 %, wie zum Beispiel von Heliatek für Vakuum prozessierte Oligomer-Tandemsolarzellen berichtet, und sind vergleichbar mit anorganischen Dünnschichtsolarzellen. Allerdings bleibt die Stabilität organischer Solarzellen eine Schwachstelle auf dem Weg zur Marktreife. Aufgrund ihrer organischen Natur, sind organische Solarzellen sehr empfindlich gegenüber Licht, Wärme, Sauerstoff und Feuchtigkeit, wobei alle üblicherweise unter Standard-Arbeitsbedingungen gleichzeitig vorliegen.

In meiner Arbeit untersuche ich die Stabilisierung gegen photooxidative Degradation, durch die Verwendung zusätzlicher stabilisierender Verbindungen, welche direkt in die Aktivschicht eingebracht werden. Hierzu wurde eine Vielzahl an Verbindungen mit unterschiedlichen Stabilisierungsmechanismen, einschließlich Radikalfängern, Wasserstoffdonatoren, UV-Absorbern und Hydroperoxidzersetzern, im Modell-System Poly(3-Hexylthiophen-2,5-diyl):Phenyl-C61-Buttersäure Methylester (P3HT:[60]PCBM) getestet. Hierzu wurden Solarzellen unter simulierter Beleuchtung, entsprechend der Einstrahlung einer Sonne, bei mäßiger Temperatur und in Gegenwart von Luft gealtert und ihre Langzeitstabilität evaluiert. Eine enorme Zunahme der akkumulierten Leistungserzeugung um einen Faktor von mehr als 3, unter Verwendung von Stabilisatoren der Wasserstoffdonator- und UV-Absorber-Klassen, konnte mittels spektroskopischer und mikroskopischer Untersuchungen auf eine Verringerung der Bildung freier Radikale zurückgeführt werden.

Diese Ergebnisse unterstreichen das Potenzial von Additiven, welche bislang bekannt waren für eine erfolgreiche Stabilisierung isolierender Kunststoffe, zur Stabilisierung von organischen Solarzellen. Anders als im Fall von isolierenden Materialien, muss jedoch zusätzliche Beachtung geschenkt werden, dass die Energieniveaus des Additivs nicht in die effektive Bandlücke der photoaktiven Schicht fallen, und dass die resultierende ternäre Mischungsmorphologie nicht den photoinduzierten Ladungstransfer unterbricht.

Unter Verwendung von P3HT:[60]PCBM als Modellsystem, konnte gezeigt werden, dass photoaktive Schichten bestehend aus einer ternären Mischungen von Polymer, Fulleren und Antioxidant bzw. UV-Absorber angewendet werden können, um die Lebensdauer organischer Solarzellen wesentlich zu verlängern. Dieser Ansatz kann prinzipiell auf andere Polymer:Fulleren Systeme übertragen werden, sofern die Präsenz des Additivs in der Mischung die Schichtmorphologie und

v

photophysikalischen Prozesse innerhalb der photoaktiven Schicht nicht negativ beeinflussen. Diese Erkenntnisse haben eine große Bedeutung für die Vermarktung von organischen Solarzellen auf dem Weg langzeitstabiler OPV. Durch Zugabe kleinster Mengen kostengünstiger Stabilisatorverbindungen kann, unabhängig von der intrinsischen photochemischen Stabilität des Polymers und Fullerens, eine signifikante Verbesserung der Lebensdauer bei hoher Solarzelleneffizienz und somit bei gesteigertem Wert des Gesamtsystems organische Solarzelle erreicht werden.

vi

Table of Contents

Abstract ............................................................................. iii

Kurzfassung ....................................................................... iv

List of Abbreviations ........................................................... viii

Chapter 1 ............................................................................ 1

Chapter 2 ............................................................................ 5

2.1 Conjugated Polymers ....................................................... 5

2.2 Elementary Processes in Organic Solar Cells ........................ 6

2.3 Morphology ..................................................................... 9

Chapter 3 .......................................................................... 13

3.1 Photophysics ................................................................. 14

3.2 Photochemistry ............................................................. 16

3.3 Photodegradation of OPV Materials ................................... 22

3.3.1 Polymers ................................................................................... 22

3.3.2 Fullerenes .................................................................................. 27

Chapter 4 .......................................................................... 30

4.1 Material System and Device Architecture ........................... 30

4.2 J(V) Characteristics ........................................................ 31

4.2.1 Long-Term Stability Measurements ................................................ 33

4.3 Fourier Transform Infrared Spectroscopy .......................... 34

4.4 Cyclic Voltammetry ........................................................ 36

Chapter 5 .......................................................................... 39

5.1 Morphological Degradation .............................................. 40

5.1.1 Tracing the Scale of Phase Separation using Microscopic Techniques .. 40

5.1.2 Modeling the Extent of Phase Segregation using Optical Spectroscopy 43

5.1.3 Observation of Thermodynamically Driven Morphological Changes in OPV Devices ......................................................................................... 47

5.1.4 Short Summary .......................................................................... 49

5.2 Long-term Stabilization using Stabilizing Additives ............. 50

vii

5.2.1 Hydrogen Donors ........................................................................ 50

5.2.2 UV Absorbers ............................................................................. 68

5.2.3 Hydroperoxide Decomposers ........................................................ 76

Chapter 6 .......................................................................... 81

Appendix A ........................................................................ 83

a) Thermally Accelerated Morphological Degradation .................................. 85

b) Photolysis by UV Radiation .................................................................. 86

c) Thermo-Oxidative Degradation ............................................................ 89

d) Thermo-Oxidative Degradation in Presence of Humidity ........................... 91

e) Photo-Thermo-Oxidative & Humidity Degradation ................................... 93

Overview .............................................................................................. 95

Appendix B ........................................................................ 97

Hydrogen Donors ................................................................................... 97

UV Absorbers ........................................................................................ 98

Hydroperoxide Decomposers ................................................................... 99

List of Own Publications ..................................................... 102

Bibliography .................................................................... 103

Acknowledgments ............................................................. 136

viii

List of Abbreviations

AFM atomic force microscopy AM 1.5 air mass 1.5 AnE-PVcc anthracene containing PPV-type copolymer poly(p-phenylene-

ethynylene)-alt-poly(p-phenylene-vinylene) APG accumulated power generation BHJ bulk heterojunction CT charge transfer ESIPT excited state intramolecular proton transfer FF fill factor FTIR Fourier transform infrared spectroscopy HOMO highest occupied molecular orbital HOO• hydroperoxy IC internal conversion IMHB intramolecular hydrogen bond ISC intersystem crossing ISOS international summit on organic photovoltaic stability ITO indium tin oxide J(V) current voltage characteristics Jsc short circuit current LUMO lowest unoccupied molecular orbital MDMO-PPV poly[2-methoxy-5-(3′,7′-dimethyloctyloxy)-1,4-phenylenevinylene] MEH-PPV poly[2-methoxy-5-(2-ethylhexyloxy)-1,4-phenylenevinylene] MIM metal-insulator-metal O2•- superoxide anion OH• hydroxyl radical OPV organic photovoltaics P• macroalkyl radical P3HT poly(3-hexylthiophene-2,5-diyl) PCBM phenyl-C61-butyric acid methyl ester PCDTBT poly[N-9′-heptadecanyl-2,7-carbazole-alt-5,5-(4′,7′-di-2-thienyl-

2′,1′,3′-benzothiadiazole)] PCE power conversion efficiency PEDOT:PSS poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) PES photoelectron spectroscopy PET polyethylene terephthalate PH intact polymer PL photoluminescence PO• alkoxy radical POH alcohol POO• peroxy radical POOH hydroperoxide PVC polyvinyl chloride QM quinone methide

ix

SRH Shockley-Read-Hall TBAPF6 n-tetrabutyl-ammonium hexafluoro-phosphate toe tonne of oil equivalent UHV ultra high vacuum UPS ultraviolet photoelectron spectroscopy Voc open circuit voltage XPS X-ray photoelectron spectroscopy

1

Chapter 1

Introduction

In the last decades, the global awareness on being a part of a fragile ecosystem with limited resources has risen. In a restricted and overpopulated system in which we live, the energy opportunities that have been discovered and overexploited in the 20th century, based mainly on fossil fuels [1], are starting to turn into big problems for our generation and the ones to come [2, 3].

The strong increase in population in the last century, from 2 billion in 1927 to over 7 billion in 2013, and availability of large amounts of fossil fuels, resulted in a consumption of 1 trillion of barrels of oil in the last 140 years [4].

Current energy demand amounts to 1000 barrels of oil, 93000 m3 of natural gas and 221 tons of coal, in 1 second [5]. This global trend is unsustainable for a multitude of environmental, economic and social reasons [4]. Primary problem is that the fossil fuels are nonrenewable, and that they are already quite depleted [6]. Consequently, the extraction methods which are being developed to reach the remaining oil and gas are cost-extensive, making the energy return on investment sink drastically [7].

Another big concern is the severe damage to the human health and the environment caused by the CO2 emissions produced by burning the fossil fuels. As the annual emission reaches 30 Gt, scientists are warning of its grievous consequences [8]. The increase of the greenhouse effect is causing anthropogenic climate change [8], with a wide environmental impact, including mass-extinction of species [9], deforestations [10], food insecurity [11-13], lack of water availability [14], ocean acidification [15], permafrost melting [16].

A way to reduce CO2 emissions has to be devised, which will at the same time ensure sufficient supply of energy needed to allow a decent quality of life for the whole population. This complex problem can be resolved by a combination of energy saving, and development of new efficient ways of exploiting traditional energy sources and implementing renewable energy sources. A general consensus exists that energy of Sun, wind, water and Earth is abundant enough to supply the power that is going to be needed in 2050 by the 9 million people which are predicted to be inhabiting the planet at that time, each consuming over 7 toe per year [4, 17-21]. As depicted in Figure 1.1, the largest fraction of the

2

naturally available renewable energy constitutes the solar energy, offering a wide diapason of application, including photovoltaics, solar heating and artificial photosynthesis.

Figure 1.1: Natural supply of renewable energy sources (larger squares) and the fractions of each of the energy sources which are technically, economically and ecologically exploitable (smaller squares and the amount in the brackets), along with the world energy consumption. The numbers below each square represent the amount of energy as compared to the world consumption (Adapted from Kohl et al. [21])

Due to their flexibility, high transparency and the potential to be manufactured in a continuous printing process with a high throughput at a low price, organic photovoltaics became a popular research field in the last two decades.

The first milestone of organic photovoltaics was set in 1994, when Yu et al. from University of California at Santa Barbara [22] created the first bulk heterojunction polymer:fullerene solar cell. Since then, in search of high efficiencies, a variety of polymers, from MDMO-PPV and MEH-PPV, over P3HT, to the novel 3rd generation polymers such as PCPDTBT, PCDTBT, PSiF-DBT, PTB7 and PBDTTPD, have been synthesized and tested [23-27].

However, on the route to successful commercialization, several conditions have to be satisfied. This challenge could be summed up as finding a way to fabricate highly efficient solar cells with a long lifetime and at a low production cost, as depicted in Figure 1.2.

The recent record in power conversion efficiency of single organic solar cells is held by Heliatek with a high 12 % for vacuum deposited oligomer tandem solar cells [28]. Yet, organic solar cells are only considered exotic niche products, mostly due to a comparably poor durability and batch to batch variations of the used compounds. While batch to batch variations can easily be overcome once a large scale market is established via large scale purification of the materials, the

3

long-term stability of the OPV devices is still very modest. Their organic nature makes them especially sensitive to light, oxygen, heat and water [29-45], and overcoming this innate weak point is a serious issue that has to be resolved in order to make OPV commercially feasible [43, 46-48].

Figure 1.2: Schematic representation of the organic photovoltaic challenge (Adapted from Brabec et al. [49, 50])

In this study, I present a novel approach to increasing the lifetime of the OPV devices, by implementing a special third component into the active layer blend, which depending on the chemical structure can act as radical scavenger, hydrogen donor, hydroperoxide decomposer, UV absorbers, or a combination of those. This approach is complementary to the on-going scientific work on improvement of encapsulation properties and getter materials. Of course it is of utmost importance to create a strong barrier to the ingress of oxygen and humidity, however one has to account that a certain amount of reactive species will manage to diffuse into the active layer. Especially flexible, transparent substrates (like PET foils) are prone to permeating atmospheric gases and application of barrier layers limits their flexibility and transparency. Similar issues arise with the getter materials, where a tradeoff always exists between their functionality on one side, and flexibility and transparency on the other side. Implementation of stabilizing additives is known from insulating plastics, where the flexibility is not impaired, and the stability of the materials could be drastically improved. However, in case of organic solar cells, it is essential to make sure that, additionally, the electrical properties are preserved, which I extensively investigate in this study.

In Chapter 2, I introduce the basics of organic photovoltaics, including their working principle and thermodynamics behind the formation of vitally important favorable morphology of the active layers. Chapter 3 gives an overview of the processes taking place in organic materials during photodegradation, and reviews reports from the literature on the topic of degradation of organic solar cells. Chapter 4 provides explanation of the main experimental methods used in this

4

study. This includes J(V) characterization of solar cells, Fourier transform infrared spectroscopy and cyclic voltammetry.

The results of my experimental work are reported in Chapter 5. In Section 5.1, thermally accelerated morphological degradation is examined by means of spectroscopic and microscopic techniques. This is of particular interest, since optimized solar cells most often exhibit non-equilibrium active layer morphology. Although blend reorganization is kinetically hindered at room temperature, a slow but steady morphology evolution towards an energetically more favored arrangement can be observed with time. A coarsening in morphology typically leads to a strong reduction of the solar cell device performance, as it will be explained in the later sections. This section of the thesis is used to distinguish this effect apart from the photochemically caused decay in performance.

As the temperature in standard working conditions did not trigger significant thermodynamically driven phase separation, the most damaging influence presents the combination of oxygen and illumination. Thus the emphasis in prolonging the lifetime of solar cells lies in photochemical stabilization.

This issue is addressed in Section 5.2, where I present a new route for stabilization of organic solar cells, by implementing portions of stabilizing additives into the active layer. In the three subsections of Section 5.2, different classes of compounds were tested, and three main mechanisms of interfering with the degradation cycle are identified and discussed. In the first subsection compounds of hydrogen donor type, with the task to inhibit propagation by neutralizing radicals produced during photooxidation, are presented. Second subsection tests UV absorbers, with the task of slowing down the initiation step by absorbing UV radiation and non-radiative deactivation of excited chromophores. Third part investigates hydroperoxide decomposers, which prevent chain branching reactions by decomposing reactive hydroperoxide degradation intermediates to non-radical products. Their working mechanisms are investigated and their stabilizing effect quantified by the means of spectroscopic and microscopic measurements. Differences in performance and stability of devices are discussed in terms of morphological compatibility of the stabilizer and the active layer components and the possibility of trap formation due to the unfavorable energetic levels of the stabilizer. The most important findings are briefly summarized and possibilities for further investigations outlined in Chapter 6.

5

Chapter 2

Organic Photovoltaics

This chapter introduces conjugated polymers and fullerenes which are the two main photoactive components of organic solar cells. An overview of the photoinduced charge transfer as the working principle of the devices, and the thermodynamically driven formation of film morphology is described.

2.1 Conjugated Polymers In the beginning of the early 1960s, several groups reported on the discovery of polymers capable of conducting electricity, particularly the works of Donald E. Weiss on polypyrrole [51-56], R. Buvet and M. Jozefowicz on polyaniline [57-63]. The most famous work in this field followed in the 1970s, conducted by H. Shirakawa, A.G. McDiarmid and A.J. Heeger on polyacetylene [64-67], which was eventually crowned with the Nobel Prize in Chemistry in year 2000, thus opening the doors for a wide variety of new technologies, such as flat screen video displays, solar cells and sensors, medical implants and flexible electronic circuitry.

The electrical conductivity of this class of materials stems from the existence of alternating single/double bonds between the carbon atoms that make up the backbone, in contrast to the insulating polymers where the backbone consists of only single bonds. Single bonds are usually -bonds, consisting of localized electrons, while the double bonds consist of one - and one - bond. In conjugated polymers, each of the backbone carbons are bound to three adjacent atoms, which leaves one electron in a pz-orbital. It is the overlap between pz-orbitals of the carbon atoms that forms a -bond, enabling the electrons to delocalize over the whole conjugated backbone. Such filled -bonds represent the HOMO (highest occupied molecular orbital) state, and the empty () bonds the LUMO (lowest unoccupied molecular orbital) state. The transition from the bonding -orbital to the anti-bonding -orbital represents the band gap, as shown in Figure 2.1.

6

Figure 2.1: Schematic representation of and bonds in ethane molecule, and the corresponding energy diagram

2.2 Elementary Processes in Organic Solar Cells First report on the photovoltaic effect in organic materials dates back to 1959, in which H. Kallmann and M. Pope described the photovoltaic activity of anthracene crystal [68]. However, the photovoltage obtained then was only 0.2 V and the efficiency below 0.1 %. The reason for such a low efficiency of the devices made of a single organic material (homojunction), is the characteristically low dielectric constant of the organic materials (2-4). Because of that, upon illumination, mobile excited states (excitons), rather than free electron-hole pairs, are generated. As excitons are strongly bound (0.7 eV in case of P3HT [69]), the electric field in the device, which arises due to the difference in the work functions of the electrodes, is insufficient to dissociate the excitons. Since the diffusion lengths of the excitons (5 nm [70]) are much shorter than the thickness of the device, most of the excitons recombine before managing to diffuse to the electrodes.

A big breakthrough came in 1986, when C.W. Tang of Eastman Kodak introduced the concept of bilayer heterojunction devices, by superposing a perylene tetracarboxylic derivative, in the role of the acceptor component, onto the conventional copper phthalocyanine donor layer, thus increasing the efficiency to about 1 % [71]. The reason for this drastic improvement in the performance lies in the more efficient exciton dissociation facilitated by the organic D/A interface, as compared to the organic layer/metal contact interface on the homojunction devices. The interface between two materials with different electron affinities and ionization potentials favors the dissociation of the excitons, and, under the condition that the differences in the potential energies are larger than the exciton binding energy, promotes the electrons to be accepted by the material with the larger electron affinity, and the holes by the material with the lower ionization potential. However, since the exciton diffusion length of the organic materials is

7

much shorter than the absorption depth of the films, it reduces the thickness of the effective light-harvesting layer to the region in the immediate proximity to the interface.

In 1995 G. Yu introduced the groundbreaking concept of bulk heterojunction devices, where the donor and acceptor material are intimately mixed, thus increasing the interfacial area and reducing the distance that the excitons have to overcome to reach the interface [22]. As this allows excitons to be dissociated practically everywhere in the active layer, it dramatically reduces exciton decay. If, additionally, percolating paths from the interface to the electrodes exist, high photon-to-electron conversion efficiency can be achieved. This concept is even nowadays accepted as the most efficient one, and is thus prevailing in current organic photovoltaics.

In Figure 2.2 the operational principle of a bulk heterojunction solar cell device is stepwise portrayed [72]. In the first step, the photon with energy larger than the band gap of the polymer is absorbed, leading to the formation of a Coulombically strongly bound exciton. In the second step the exciton diffuses to the donor-acceptor interface, provided that it has not decayed. Next, the energy of the donor-acceptor interface overcomes the Coulomb binding energy, due to the difference of the LUMO levels of the donor and the acceptor. Upon dissociation of the exciton, positive and negative polarons are formed, still existing as Coulombically bound pairs. This polaron pair gets dissociated by the internal electric field existing due to the difference of the work functions of the electrodes, creating free charge carriers. Now the positive polarons drift to the anode and the negative ones to the cathode. Finally, to extract the polarons from the active layer through their respective phases to the electrodes, the work function of the anode has to match the HOMO level of the donor, and the work function of the cathode has to match the LUMO of the acceptor, which assures the ohmic contacts.

Although the bulk heterojunction concept offers higher conversion efficiencies as compared to the bilayer devices, due to the increased interface between the donor and acceptor components needed for the exciton dissociation, the transport of the dissociated charges will only be able to take place if percolation pathways from the interface through each of the phases to the respective electrodes exist. However, it is important to find a tradeoff between the larger domains size, as favored for the charge transport, and blending intimacy, as needed for the exciton dissociation. Another important issue is the vertical phase segregation which might negatively influence the electrode selectivity. There are many pathways that have been explored in order to optimize the morphology, by both the intrinsic structural parameters, and by extrinsic processing factors influencing the film drying [73-78].

8

Figure 2.2: Photoinduced charge transfer in a donor-acceptor solar cell

9

2.3 Morphology Active layer of organic solar cells is a mixture of two components, and it can spontaneously phase-separate to produce a structure rich in internal interfaces. Polymer mixtures are prone to demixing into separate phases, rather than forming a uniformly dispersed single phase. This is a fundamental property of large molecules, which can be explained using the Flory-Huggins model of polymer mixing thermodynamics [79, 80].

According to this theory the mixing behavior can be predicted based on parameter representing the interaction between the segments of the polymer chains. It represents the energy change in units of kBT upon placing a segment of polymer A in an environment of pure polymer B. A positive value of represents the energetic tendency to demix, which is opposed by the entropy increase arising by the mixing of the two components. A negative value of occurs only in cases of some specific interactions between the two polymers, such as hydrogen bonding or charge transfer interactions. In case of small molecules, the critical point corresponding to the value of that is just enough to make the components start demixing. In case of polymers, the contribution of entropy to the total free energy is much smaller, and the at the critical point is reduced by the degree of polymerization. Because of that, demixing can be triggered even by the slightest energetic interaction between two polymers.

However, Flory-Huggins theory holds true only under the assumption that the interactions between the polymers are not strong enough to cause perturbation of the chain conformation. This may not always be the case.

For example, when the equilibrium state of one of the polymers is crystalline, the tendency to demix will be stronger. As too-intimately mixed domains are an obstacle to successful charge transport, coarsening of domains if sometimes required. In cases of semicrystalline polymer blend systems, such as P3HT:[60]PCBM, this can be induced by short thermal annealing, which promotes crystallization of polymer phase and demixing of the two phases [81].

Also, the situation is slightly different as typical OPV active layers are processed from solution [82]. A good solvent dilutes the unfavorable interactions between two polymers, and above a certain concentration, it allows mixing of two, otherwise immiscible, polymers, as shown in Figure 2.3. Still, upon casting and evaporation of the solution, the interactions become strong again, which might initiate phase separation within the film.

10

Figure 2.3: (left) Phase diagram of a ternary polymer mixture in isothermic environment, with coexistence (binodal) curve; (right) Phase diagram with the coexistence and spinodal curve. Coexistence curve separates the single phase stable region from the region with coexisting phases. Spinodal curve separates the unstable and metastable regions, with respect to the small composition fluctuations. TC and C are the temperature and composition at the critical point

The properties of the solvent, such as boiling point and evaporation rate, viscosity and solubility of polymer and fullerene, play an important role. The strong impact that the solvent can exert on the morphology was first noticed on the MDMO-PPV:[60]PCBM system, where substitution of toluene with chlorobenzene resulted with a finer nanomophology, thus almost tripling the efficiency from 0.9 % to 2.5 % [83]. Mixed solvents, consisting of equal portions of low and high boiling point solvents (cosolvent approach) [84-86], or small additions of high boiling point processing additives with selective solubility of only the fullerene component, (such as diiodooctane, octanedithiol) [87-91] are often used. In both cases, the low evaporation rate of the high boiling point solvents prolongs the drying time of the films, improving thus the ordering of the polymer.

Solvents in gaseous phase can also be used to control the morphology of an already casted film, which is known as solvent annealing. It was shown that ortho-dichlorobenzene vapors can increase the order of P3HT [92]. In general it was found that under exposure to poor solvent vapors, the PCBM segregation is enhanced [93], and it migrates towards the surface [94], while P3HT forms a nanofibrillar crystalline network [95-97].

Since polymers are large, entangled molecules, they move slowly, and the kinetics of reaching the equilibrium is important. In practice, the motion of polymers gets frozen before the system can reach equilibrium, by going through glass transition or crystallization.

11

Motion of polymer chains in concentrated solutions occurs by reptation, in which the chains wiggle according to Brownian motion [98]. The diffusion of such polymers is defined by the mobility of chain segments and the degree of polymerization. The strong dependence of mobility on temperature and solvent concentration, results in the glass transition, a state in which the polymer chains are no longer able to move towards the equilibrium and they become practically immobile.

Based on the consideration of the shape of the curve of free energy versus concentration, the mixture of two polymers can be stable, metastable and unstable, as depicted in Figure 2.3. Convex shape of the free energy curve implies instability, in which any fluctuation in the concentration, such as the thermally induced oscillations in concentration, results in the lowering of the free energy of the system, causing phase separation. The fundamental quantity that drives it is the chemical potential, which is related to the first derivative of the free energy with respect to concentration. In regions above the spinodal curve, where the second derivative of the free energy is positive, the regions of high concentration have high chemical potential and the diffusion occurs in normal downhill direction. However, inside the spinodal curve, the chemical potential gradient has the opposite sign to the concentration gradient and the material diffuses uphill from regions of low to regions of high concentrations. This process is known as spinodal decomposition.

In spinodal decomposition, not all concentration fluctuations grow at the same rate, as the long-length scale fluctuations grow slower and the small scale fluctuations result in increased amount of interfaces and associated free energy cost. Thus the phase separation pattern is dominated by a single length scale (order of tens of nanometers), which is defined by the molecular size of the polymers and the temperature difference from the critical point. The resulting pattern can be defined as the superposition of waves, random in phase and direction, and with a well-defined wavelength [99].

The interactions of surfaces and interfaces on the composition of polymer blend films are very powerful, and they can lead to vertical phase separation. This phase separation occurs perpendicular to the surface, resulting in formation of layers of different component phases [100].

The two components which form a single phase, under even a small difference in surface energies, are likely to create a surface enrichment consisting of the lower surface energy component. The relaxation depth over which the enrichment reaches the bulk concentration is influenced by the dimensions of the polymer chain and by how close the mixture is to the limits of miscibility. The growth of this layer occurs by normal diffusion, so the time to reach the equilibrium is very

12

long. Accordingly, the non-equilibrium state of partial surface segregation can easily be frozen if one of the components has a tendency to crystallize or become glassy.

In case the polymers are immiscible, two different situations can take place, depending whether the system is in the metastable or unstable region of the phase diagram. If the thin film is in the metastable part of the phase diagram, the surface-segregated layer grows continuously, and as long as the system stays close to the critical point, the surface gets completely wetted by the lowest energy phase. Growth of this layer occurs by normal diffusion from the other phase, and if there is no competing nucleation processes in the bulk, it will result in complete self-stratification.

Complete layering in the plane of the film can also occur in the unstable part of the phase diagram, by a surface-driven spinodal decomposition mechanism. As the surface favors one of the two components, there is a distinct phase and direction of composition waves. This results in creation of alternating layers of the two components, the thickness of which is set by the same wavelength selection mechanism that determines the initial size of the domains in bulk phase separation by spinodal decomposition. Although the layers propagate from the surface and gradually lose coherence towards the bulk, in thin films the self-stratification is almost complete.

It was found that P3HT:[60]PCBM blends casted on silica substrates undergo spontaneous vertical phase segregation of P3HT towards the top, which is further enhanced by thermal annealing [73, 101, 102]. When a substrate with low surface energy is used, in order to decrease the interfacial free energy, the lower surface energy component (P3HT) migrates to the interface. When a high surface energy substrate (PEDOT:PSS, SiO2) is used, the higher surface energy component (PCBM) segregates to this surface [103-105].

All of the above described morphological changes that occur in the active layers with aging time, have an impact on the photochemical changes. All of the reactions that occur in the polymer within the photochemical degradation cycle are depending on the amount of oxygen available, which is determined by diffusion and solubility of oxygen in the polymer [106, 107]. As oxygen is generally soluble only in amorphous areas of semicrystalline polymers, there is a strong dependence on the morphology of the polymer and its stability [106-110].

13

Chapter 3

Photodegradation

In presence of oxygen and light, which are present throughout the whole life cycle of solar cells, organic materials undergo degradation reactions, resulting in different physical and chemical property changes. Physical changes include decrease in molecular weight, tensile strength, impact strength, elongation at break, loss of gloss, and surface erosion. Chemical changes include alteration of the chemical structure by formation of oxidation products, leading to electrical failure [111].

Stark-Einstein law states that every photon that is absorbed will cause a chemical or physical reaction. This section gives a brief overview of the main photophysical and photochemical processes relevant to conjugated polymers, as summarized in Figure 3.1.

Figure 3.1: An overview of main photo-processes in conjugated polymers (Adapted from Abdou et al. [112])

14

3.1 Photophysics Incident light is either reflected from the surface or scattered or absorbed in the bulk of the polymer. Grotthuß-Draper law states that only the light which is absorbed by a system can result in a photochemical reaction. Two types of chromophores can take part in such reaction, the polymer functional groups (intrinsic chromophores) and chromophores from impurities and additives (extrinsic chromophores). The principal law relating the absorption of light to the properties of the material through which it is travelling is the Beer-Lambert-Bouguer law, stating that the amount of transmitted light depends exponentially on the absorption coefficient of the material , and the thickness of the sample d:

dαT eII -

0= Equation 1

where IT is the intensity of the transmitted light and I0 the intensity of the incident light.

Absorption of light occurs very quickly, on the order of 10-15 s, in discrete amounts termed quanta and corresponds to excitation of the fluorophore to a higher energy state. The energy in a quantum is directly proportional to the frequency and inversely proportional to the wavelength, as stated in the Planck’s law:

λhc

νhE == Equation 2

where E is the energy, h Planck’s constant, frequency of the incoming photon, wavelength of the photon, and c the speed of light.

The absorption of photon occurs due to an interaction of the oscillating electric field vector of the light wave with electrons in the molecule. It can occur only with incident photon of specific wavelengths, known as absorption bands. If a photon contains excess energy than what is necessary for an electronic transition, it will be converted in vibrational and rotational energy. If a photon contains less energy than necessary, no absorption will be possible.

Upon absorption of UV or visible radiation, the chromophore is excited from ground (M) to an excited energy state, usually singlet state (1M*). Radiation of wide spectrum of wavelengths generates a wide range of such transitions that populate various vibrational energy levels of the excited states, which is termed absorption spectrum. The main photophysical deactivation routes are shown in Figure 3.2.

15

Figure 3.2: Jabłoński diagram of photophysical processes in conjugated polymers (Adapted from Abdou et al. [112])

Vibrational Relaxation and Internal Conversion Following absorption, several deactivation processes are possible. The most probable is the relaxation to the lowest vibrational energy level of the excited state, in which the energy is given away to other vibrational modes as kinetic energy. This process is very fast, in order between 10-14 and 10-11 seconds, so it is extremely likely to follow immediately upon absorption.

Usually this relaxation occurs between vibrational levels, however if the vibrational energy levels overlap strongly with electronic energy levels, it is possible for excited electrons to change from one electronic level to another. This case of vibrational relaxation is termed internal conversion (IC). As vibrational and electronic energies do not overlap and a large energy difference exists between the ground and first excited state, the transition of an electron to the ground state via this mechanism is very slow. This makes it possible for other processes to compete with internal conversion, and although both vibrational relaxation and internal conversion usually take place, they are rarely the final transition [113-116].

Fluorescence Fluorescence (Fl) is a process where the absorbed energy is released in form of a photon. It is a slow process, on order of 10-12 to 10-7 seconds, which makes it an unlikely process for electrons at states higher than first excited state. It is an allowed transition without the change in spin. Most often it is observed between the first and ground states, as at higher energies vibrational relaxation and internal conversion are more probable.

Intersystem Crossing In intersystem crossing (ISC) the electron changes the spin from an excited singlet to an excited triplet state. It is the slowest transition, several orders

16

slower than fluorescence (on order 10-11 to 10-6 seconds). Based strictly on electronic selection rules, it is a forbidden transition, however by coupling vibrational factors in selection rules the transition becomes weakly allowed, so it can compete with fluorescence. There are several routes from the triplet to the ground state. Phosphorescence (Ph) is a slow forbidden radiative transition (on order 10-6-102 seconds) from an excited triplet to a singlet ground state. In delayed fluorescence, first the transition back to the first excited singlet state occurs, followed by emitting transition to the ground electronic state.

In -conjugated polymers, the local density of chromophores is high, and the neighboring excited and ground state chromophores can interact. This makes bimolecular mechanisms, such as quenching (path 1 and 4 in Figure 3.2), migrative energy transfer (path 2 and 5 in Figure 3.2) and annihilation (path 3 in Figure 3.2), the dominant deactivation processes. An excited singlet state can also interact with a ground state molecule to create an excimer (1D*). Its ground state dimer (D) is unstable and it readily dissociates. In case the complex is formed between excited and ground states of different molecules, this complex is called an exciplex.

Although they generally deactivate excited states, non-radiative processes can affect photochemical reactions. For example, energy transfer process, where the excitation energy is passed from one molecule to another, can be used to channel the energy into a reactive center [117, 118]. Also, excited singlet and triplet states can be quenched by oxygen, thereby forming singlet oxygen (1O2), which is long lived and reacts easily with unsaturated groups within polymer [112].

3.2 Photochemistry The source of the activation energy for photochemical degradation is sunlight. Only a small portion of Sun’s electromagnetic energy reaches Earth’s surface, with wavelengths above 290 nm. Most photochemical reactions require activation energies between 60 and 279 kJ/mol, which correspond to light wavelengths of 440 to 2000 nm. This fits the energy required to break single covalent bonds, which ranges from 165 to 420 kJ/mol, i.e. radiation wavelengths of 280 to 720 nm. Therefore the near UV radiation (300-400 nm) is sufficient to break most single covalent bonds, with the exception of strong bonds such as C-H and O-H [119].

Autoxidation The first oxidation studies of polymers were conducted on natural rubber, and the influence of oxygen on aging was described already early by Hoffmann et al. [120]. Bolland and Gee [121-123] developed the scheme of autoxidation in more

17

detail, based on studies of oxidation of low molecular weight hydrocarbons, although they can be also applied to the oxidation of polyolefins, conjugated polymers and other plastics. The presence of -conjugated system next to an aliphatic substituent supports hydrogen abstraction from the substituent by stabilizing the resulting free radical, an effect that has been observed for polystyrene and P3HT [112].

Autoxidation is an autocatalytic free radical initiated chain reaction. Initial induction period starts slow, involving negligible oxygen uptake and hydroperoxide formation. The duration of this period is indirectly proportional to initiation rate. The second stage is autocatalytic, for which a quick increase of oxygen uptake and hydroperoxide content are characteristic. With a steep increase in concentration of hydroperoxides, there is a strong correlation between oxygen uptake and the amount of formed degradation products detectable by IR spectroscopy (-OOH, CO, -OH, and C=C unsaturation). Finally, in the last stage, the rate of oxygen uptake drops [106, 124, 125].

Figure 3.3: Possible pathways of photochemical degradation in polymers (Adapted from Chirinos-Padron et al. [126])

18

Oxidation reactions can be divided in initiation, propagation, branching and termination step, which are cyclically repeated, as shown in Figure 3.3. The number of realized cycles, the kinetic chain length, is dependent on the initiation and termination rates [124].

Initiation In the initiation step the alkyl radicals P•, which are the initiators for the chain reaction, are created. There are several possibilities for initiation to occur. One is by interaction of external molecular impurities (RR‘) which absorb UV-Vis radiation and produce low molecular weight radicals (R•, R’•), which then react with the polymer (PH) and produce an alkyl radical (P•) by hydrogen abstraction [127]:

)' (+)' (+'+→'

•••

••

HRorRHPRorRPHRRRR νh

→ Equation 3

Examples of external impurities include traces of catalysts and solvents used in polymerization, additives, and traces of metal particles from the polymer processing [127].

Another possibility is via internal impurity chromophores present as a part of the polymer structure, which absorb UV-Vis radiation and produce polymer alkyl radicals (P•) and low molecular weight radical fragments (R•) [127]:

)+ (+→ •••• RPorPPPolymer νh Equation 4

Internal impurities may be hydroperoxides, carbonyl and unsaturated bonds, anomalous structural units such as branching, and catalyst residues attached to the chain ends of the polymer [127].

Initiation can also occur by direct dissociation reactions of a chemical bond (via Norrish Type I and Type II reactions), which was excited to a singlet or triplet state [126, 128-131], or by dissociation of charge-transfer (CT) complexes formed between polymer and oxygen [126-130, 132-136]:

•2

•2 +→ HOPOPH νh

Equation 5

In reaction with oxygen, the above mentioned initiating impurities, dissociated bonds and charge-transfer complexes, form peroxy radicals POO• which further abstract hydrogen from the polymer, thus creating a macromolecular alkyl radical.

19

Propagation In a very fast reaction, molecular oxygen is fixated on the alkyl radical, forming a peroxy radical. The peroxy radical regenerates the alkyl radical by abstraction of a hydrogen atom from the polymer, thus forming a hydroperoxide:

••2

•22

•

++

+

PPOOHPHPO

POOP

→

→ Equation 6

Branching The hydroperoxides formed in the propagation step are homolyzed by breaking the O-O bond, photochemically (by radiation lower than 300 nm) [137-141], thermally [142] or catalyzed by various metal ions [126, 128-130, 143, 144]. This creates highly reactive alkoxy and hydroxyl radicals which react with molecular oxygen and initiate new oxidation chains [124]:

•2

•

••2

•2

•

••

++++

++2

+

POHPHOHPPOHPHPO

OHPOPOPOOH

OHPOPOOH

→

→

→

→

Equation 7

Hydroperoxide homolysis which follows the decay of the kinetic chain in which it was formed, is referred to as degenerate branching. As a result of hydroperoxide homolysis, carbonyl groups are formed on carbon atoms where the primary oxidation attack took place [124].

Chain Termination Eventually, the chain termination occurs by recombination of radical species, thereby forming non-radical products and oxygen in singlet state [145]:

OHORORROOHPPPP

POOPPOP

POPPOP

OPOOPPOPO

22

••

•2

•

••2

•2

•2

++2+

+

+

++

→

→

→

→

→

Equation 8

A process identical to the above described autoxidation, takes also place in the process of thermal oxidation, the difference merely being that, instead of light, the source of radical formation is heat.

Chain Scission Radicals formed in the above described initiation, propagation and chain branching steps are not limited to fixing oxygen and abstracting hydrogen atoms.

20

Often they take part in unimolecular decomposition reactions or chain scissions, and as a consequence decrease the molecular weight of the polymer.

Crosslinking Also, they can take part in intermolecular crosslinking, i.e. covalent bonding between different polymer chains, thus increasing the molecular weight and eventually combining all of the macromolecules into a three-dimensional insoluble network (gels) [126, 128-131, 146, 147].

Rearrangements Upon light absorption certain groups undergo reactions where the carbon skeleton of a molecule is rearranged to give a structural isomer of the original molecule. A common such reaction is the photo-Fries in which the phenyl ester is rearranged to a hydroxyl aryl ketone, as shown in Figure 3.4 [125, 148, 149].

Figure 3.4: Photo Fries rearrangement

Photoreactions of carbonyl groups Upon photoexcitation of ketone and aldehyde groups, formed in the oxidation cycle, the carbon bond of the -carbon cleaves, forming two radicals, as shown in Figure 3.5 on the left. This is Norrish type I reaction. Also possible is the Norrish Type II reaction, where the photoexcited carbonyl compound, by abstracting a -hydrogen (hydrogen atom three carbon positions away from the carbonyl group), produces a 1,4-biradical, as shown in Figure 3.5 on the right [112, 125].

Figure 3.5: Norrish reaction type 1 and type 2

21

Singlet oxygen and superoxide anion reactions As shown in Figure 3.6, formation of singlet oxygen is often formed in triplet-triplet annihilation processes. The energy gap between the ground and first excited singlet state is only 22.5 kcal/mol, so the formation is thermodynamically possible. Molecular oxygen can enhance intersystem crossing of singlet excited state, and increase the triplet concentration, which enhances triplet-triplet annihilation [112, 150].

Figure 3.6: Singlet oxygen photosensitization pathways (Adapted from Abdou et al. [112])

Figure 3.7: Singlet oxygen reactions with aromatic compounds (top) and olefins (bottom) (Adapted from Abdou et al. [112])

Singlet oxygen is unreactive towards saturated hydrocarbons, but it is strongly reactive with aromatic compounds [150] and olefins containing allylic hydrogen atoms [151], as shown in Figure 3.7 (top). In the latter case, “ene”-type reaction occurs in the -position to a C=C bond of the polymer with singlet oxygen 1O2,

22

thus producing allylic hydroperoxide groups with shifted double bonds, as shown in Figure 3.7 (bottom) [112, 126, 128, 129, 152-157].

Singlet oxygen is one of two possible reactive forms of molecular oxygen- the other form being the superoxide anion (O2•-)- that can be formed by photolytic dissociation of oxygen-polymer CT complexes (M•+…O2•-) [108, 112, 158-161]. Superoxide anion (O2•-), hydroxyl HO• and hydroperoxy HO2• radicals can be formed, by interconvertible reactions, one into another [108, 162].

3.3 Photodegradation of OPV Materials

3.3.1 Polymers

Oxidation Pioneering OPV materials stability investigations documented the formation of PPV (poly(phenylene vinylene)) triplet states. Those polymer triplet states would generate singlet oxygen, which was attacking the bridging double bonds of polymer to form endoperoxides that further decompose into aromatic aldehydes accompanied by chain scission, followed by oxidation of ether substituents to form ester [163-165]. It was long believed that this pathway is the main responsible for the photodegradation [163, 164, 166-168].

However, Ma et al. demonstrated that, besides singlet oxygen, there are several other transient species in the photooxidation of PPV in solution: superoxide radical anion, PPV triplet state and PPV radical cation and anion. Using different dye-sensitizers they showed that singlet oxygen does not play the main role in the photodegradation of PPV [169]. Additionally a study by Chambon et al. observed that the same mechanisms take place in both thermooxidation and photooxidation of the solid state PPV, invalidating thus the role of singlet oxygen as the main active intermediate [170]. In a later study, Chambon et al. identified the main intermediate specie to be the radical polymer cation (MDMO-PPV•+), formed by photoinduced electronic transfer, which leads to the formation of the superoxide radical anion (O2•-). The presence of oxygen and photo-aging favors the formation of the radical cation, suggesting that oxygen and resulting photooxidation products act as electron acceptors. These charged radicals evolve further by oxidation and scission, propagating thus the chain oxidation process by abstraction of labile hydrogen in the -position of the ether function, and by photo-Fries type rearrangement, resulting with addition on the double bonds, as depicted in Figure 3.8 [171].

23

Figure 3.8: Photooxidation initiation mechanism of MDMO-PPV polymer (Adapted from Chambon et al. [171])

Figure 3.9: Singlet oxygen degradation mechanism of P3HT polymer (Adapted from Abdou et al. [172, 173])

Similarly, in the case of photooxidation of P3HT, based on experiment involving P3HT in solution exposed to chemically-generated singlet oxygen [172, 174, 175], it was initially proposed that the photodegradation in solid state proceeds also via oxidation of thiophene rings, induced by the photosensitized singlet oxygen, as

24

shown in Figure 3.9 [173]. According to that mechanism, the singlet oxygen undergoes a 1,4 Diels-Alder addition to the thienyl units, ultimately forming an unstable endoperoxide, which further decomposes into carbonyl, olefinic and sulfine derivatives and breaks the macromolecular backbone, leading to bleaching without affecting the side chains [176].

However, although the quantum yield of singlet oxygen formation in P3HT solution is above 30 % [177], in film both triplet and singlet oxygen yields are strongly reduced [178, 179]. Thus it seems more plausible that the prevailing mechanism in solid state is radical based [173, 180-183]. Furthermore, a study by Manceau et al. showed that in the solid state, even at high concentrations of singlet oxygen, P3HT thin films remain unoxidized [181]. Additionally, it was found in a subsequent study by Manceau et al. that in both thermo- and photo- oxidation of P3HT the same products are formed, carbonyl and sulfur containing moieties, thus further refuting the dominating role of singlet oxygen in the degradation [180]. In that study, it is reported upon the radical mechanism that takes place, involving the oxidation of the side chain by ground state oxygen, which proceeds via a free radical chain reaction route. The process starts with the attack of the oxygen centered radicals on the -carbon atom of the alkyl side chain, preceded by hydrogen abstraction. This leads to the formation of a macroalkyl radical, followed by oxygen fixation and hydrogen abstraction, resulting with the formation of unsaturated hydrogen peroxide. Hydroperoxide is further decomposed thermally and photochemically, yielding alkoxy and hydroxyl radicals. Three main reaction pathways exist for the alkoxy radicals. First one involves hydrogen abstraction, leading to the formation of -unsaturated alcohols. Second one proceeds with a cage reaction of the alkoxy radical with the HO•, resulting with a photo-unstable but thermally stabile aromatic ketone, which get photolyzed and oxidated into aromatic and aliphatic carboxylic acids. And the third possibility is the scission which leads to the formation of an unstable and rapidly oxidizable aromatic aldehyde and also alkyl radicals which get transformed in aliphatic carboxylic acids and anhydride groups generated by condensation of carboxylic acids. These processes have been depicted schematically in Figure 3.10 [172, 173, 180, 181, 184].

Recently, it has been demonstrated by Hintz et al. that in the solid state, the degradation mechanism depends on the wavelength of the light source [183]. They have shown that under UV radiation, the conjugated system and the hexyl side chain are degraded simultaneously, involving Norrish type I reaction that leads to complete disintegration of the side chains, consistent with the radical mechanism starting at the -carbon of the alkyl side chain [180, 183, 185]. Under visible light, only the -conjugated system is affected, and the rate decreases with time, indicating that the active species is a direct product of photosensitization of the

25

polymer [183], possibly singlet oxygen [172, 173, 183, 186]. When exposed to sunlight, both mechanisms take place, but since the UV mechanism has a much higher effectiveness than the visible light, radical chain mechanism dominates even at low UV intensities [183].

Figure 3.10: Radical chain mechanism of (top) P3HT side chain and (bottom) thiophene group (Adapted from Manceau et al. [180])

Oxygen Doping Besides the irreversible degradation assigned to the formation of carbonyl and carboxylic groups, additional reversible processes impact the performance of organic solar cells, such as reversible oxygen doping of the polymer in form of a polymer-oxygen charge transfer complex [172, 187-192].

26

For polymers with low ionization potential the interaction with oxygen is especially strong, resulting in p-doping. The electronic structure of P3HT gets modified by the adsorption of the molecular oxygen, thereby not forming any irreversible chemical bonds. Abdou et al. first reported on the formation of charge transfer complex between P3HT and oxygen[193], with the energy E = IPD-EAA-W (where IPD is the ionization potential of the donor, EAA the electron affinity of the acceptor, and W the Coulombic attraction energy of the complex) [194] which can be empirically approximated by ED

1/2-EA1/2+0.15 eV (where ED

1/2 and EA

1/2 are the half-wave electrochemical potentials of the donor and the acceptor) [195, 196]. As it requires oxygen diffusing into the bulk, the morphology of the polymer is affecting the formation of such a charge transfer complex [197], making it more pronounced in amorphous polymers and polymers which are not closely -stacked [198]. This weakly bound state modulates the electrical properties of the polymer, and it is proportional to the oxygen pressure. Analysis of P3HT FET devices under increasing oxygen pressure showed that by the formation of charge transfer complex, the charge carrier concentration and the conductivity increase, and charge carrier mobility decreases [193]. This effect is fully reversible upon removal of oxygen. But while the oxygen adsorbed near the surface of the film can be easily desorbed even at room temperature under vacuum, the deeply trapped oxygen molecules can only be removed by a high temperature annealing near the glass transition temperature of the polymer, necessary to mobilize the polymer chains and release the oxygen [189].

Figure 3.11: Energy levels of oxygen doped P3HT (Adapted from Hintz et al. [188])

Lu et al. explained using ab initio calculations that adsorption of oxygen into the polymer changes its electronic structure from semiconductor to a metal through hybridization of the wave functions of oxygen and polymer. The oxygen band becomes degenerate with the polymer valence band and the Fermi level gets pinned by the oxygen band to lie inside the valence band. This allows the polymer to get oxygen doped even in the dark [189, 199, 200], although, as it is

27

limited by the slow diffusion through the polymer matrix, the process is much slower than under illumination, where, additionally, photons excite the electrons in the sulfur band and the valence band to the empty states in the oxygen band [199]. This was further confirmed using photoelectron spectroscopy by Hintz et al., who found that the oxygen molecule bound to the -system of the polymer in the charge transfer complex traps an electron which leaves behind a mobile hole on the electronic system of the polymer, leading to a shift of the Fermi level toward the HOMO level of the polymer [188], as depicted in Figure 3.11.

The same effect takes place in polymer:fullerene blends [187, 191, 201, 202]. The process here also consists of the transfer of an electron from P3HT to oxygen. Mobile holes on P3HT chains (free charge carrier) and immobile superoxide anions (O2•-, i.e. electron trapped on the oxygen) are created, leading to the formation of a space charge region in front of the electron extracting electrode (anode) [201], whose width depends on the doping level as well as on the applied bias. This space charge region shields the electric field inside the photoactive layer, thus hindering the extraction of the charge carriers, leading to a loss of short circuit current [187].

A study by Schafferhans et al. found that upon oxygen exposure in the dark the charge carrier concentration is increased which raises the bimolecular recombination probability. The additional charges lead to less band bending, and therefore a reduced electric field within the solar cell. Because of that, the extraction time for the charge carriers increases, leading to a higher recombination probability and a lower short circuit current [191]. Additional losses in open circuit voltage and fill factor, as observed under illumination, have been assigned to the increase of the density of deep traps, as they can act as charge carrier recombination centers [187, 191, 201].

Just as in the case of single components, the process is reversed upon annealing, by removing the trapped electron from the oxygen, after which the electron recombines with a hole and oxygen concentration adjusts to the equilibrium value at the given temperature [187].

3.3.2 Fullerenes

Oxygen Doping The fullerene component per se is also affected by oxygen, present in the devices either as the residual from manufacture, contamination by ITO, or by diffusion from the environment through the electrodes and the encapsulation [47].

Even at room temperature, the oxygen quickly diffuses into the bulk of C60, reducing the effective electron mobility dramatically. This happens due to the intercalation of oxygen, which causes the Fermi level to shift down to the middle

28

of the gap, making the oxygen act as an electron trap for the charge carriers [203-

205], and as non-radiative recombination center [206]. Just as in the case of polymers, upon annealing at elevated temperatures in vacuum, this effect can be reversed.

Oxidation When the fullerene is exposed to oxygen under illumination, the process cannot be fully reversed [202, 204]. A study by Eklund et al. found that UV-Vis radiation enhances the oxygen diffusion through interstitial voids in the fullerene lattice, which promotes formation of C-O and inter-fullerene bonds [204, 207]. Oxygen was found to form epoxidic species, where two different carbon atoms are linked to one oxygen atom by single bonds, via addition of oxygen to the fullerene [208]. After prolonged exposure to moderate temperature (80 °C), the fullerene cages linked with a single bond to atomic oxygen further decompose and polymerize [209], while at more elevated temperatures (200-300 °C), double bonds C=O get formed, leading to cage openings [210]. It was shown by Wohlers et al. that after once exposed to oxygen, fullerene samples oxidatively polymerize upon any thermal treatment, even when no oxygen is present [206, 210]. From the quantum calculations by Reese et al, it is known that PCBM oxides have deeper LUMO levels than the pristine PCBM, making them act as traps within the PCBM domains [44, 211]. However, it is to be noted that Könenkamp et al. found that only 1-10 % of the oxygen incorporated in C60 film effectively induce oxidation reactions [212].

Distler et al. reported on the stabilizing effect of PCBM on -conjugated polymers under different conditions [34, 40, 41, 180, 213-216], like photooxidation [34, 41, 44, 217], thermolysis [41], oxygen atmosphere in the dark [44] and photolysis [41]. This effect is assigned to three properties: exciton quenching, radical scavenging and UV light screening [218]. Since the photo-oxidation rate is linearly dependent on light intensity [38], the stabilization occurs also partly due to UV light screening of the polymer from the incident light by the PCBM molecules. However, this effect contributes to the stabilization only in small amounts [218]. The quenching of the polymer exciton by electron transfer to PCBM reduces the lifetime of the excited singlet state, thus reducing the population of all reactive states populated via singlet states. One of such reactive states is the triplet state, which is long lived and highly reactive, due to which it favors bimolecular reactions, such as the singlet oxygen sensitization [192, 218, 219]. However, Distler et al. have demonstrated that the extent of fluorescence quenching by fullerene does not correlate with the stabilization of the polymer, thus indicating that this is also not the main stabilizing mechanism of PCBM [218]. As the stabilizing effect of PCBM on aging of P3HT has been observed in absence of light, it is obvious that excited state quenching is not the only mechanism of stabilization by PCBM, but that also

29

a ground state property of PCBM is involved. Manceau et al. assign this to radical scavenging [41, 44]. This property of fullerenes is well documented in literature [220-

223], and it is mainly due to the stabilization of radicals by their addition to the extended -system of the fullerenes.

It has also been reported that the PCBM can accelerate the degradation. Although the inter-system crossing yield in solid films of PCBM [224] is not as high as in the solution [225-227], excited PCBM in the presence of oxygen can still generate singlet oxygen and cause significant degradation [218, 228]. It has also been shown by Yamakoshi et al. that electron transfer from fullerene anions to molecular oxygen can result in the formation of superoxide anion [229]. This has further been confirmed in the study by Hoke et al. where the increase of degradation rate of a series of polymers in blends with fullerenes of lower electron affinities [217, 230] has been explained as a result of enhanced formation of superoxide anions via the fullerene anion [230]. Other possible losses include the transfer of energy from the PCBM triplet to the polymer triplet state [231], and the recombination of the charge transfer state formed in the photoinduced charge transfer at the interface between the polymer and the PCBM which, if the charge transfer state is of higher energy than the triplet state [232-234], might populate the polymer triplet [218, 232, 234-237]. If the energy of the polymer triplet state is higher than the energy of single oxygen, it will be possible to transfer the energy to oxygen, thus creating a highly reactive species that has been known to degrade the conjugated polymers [238], e.g. in case of P3HT by 4,2-cycloaddition to the thiophene ring [172, 173].

30

Chapter 4

Characterization Techniques and Material System

This section gives a brief overview of the material system and the device architecture examined in this study, as well as the experimental techniques used for characterization.

4.1 Material System and Device Architecture Bulk heterojunction (BHJ) solar cells are typically fabricated in standard architecture, depicted in Figure 4.1. Such a device consists of the bulk heterojunction active layer, consisting of a mixture of polymer:fullerene components, sandwiched between two electrodes. The substrate glass coated with tin-doped indium oxide (ITO) is used as the transparent conductive anode with a high work function. Due to the roughness of the ITO layer, and also to additionally increase the work function of the anode, an additional layer of poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) is commonly deposited before the active layer. As cathode materials, low work function metals, such as aluminum, magnesium and calcium are employed.

Since March 1994, when Yu, Zhang and Heeger reported the first successful creation of polymer:fullerene bulk heterojunction solar cells [239], in search of high performances, a variety of materials has been synthesized and tested for use in active layers of organic solar cells. The recent record in power conversion efficiency is held by Heliatek with a high 12 % for vacuum deposited oligomer tandem solar cells [28]. However, as mentioned in the previous section, improving their stability remains a serious issue on the way to commercialization of organic solar cells. In this study I propose a novel way to resolve this issue, by introducing a third component into the active layer, which is designed to stabilize the performance of the devices. The model system for active layer in this study is poly(3-hexylthiophene-2,5-diyl):phenyl-C61-butyric acid methyl ester (P3HT:[60]PCBM), as it is a commercially available, commonly used system with well-understood processes documented in the literature.

31

Figure 4.1: Architecture of a standard organic solar cell device, along with the chemical structures of the constituent materials used in this study

4.2 J(V) Characteristics Photovoltaic cells are electronic devices that convert sunlight into electrical power. Their performance and basic electrical characteristics can be determined by current-voltage measurements. A typical characteristic of a bulk heterojunction solar cell in the dark and under illumination is shown in Figure 4.2. Under illumination the J(V) curve is shifted down by the amount of photocurrent generated (Jph).

The device performance is described by several characteristic parameters:

Open Circuit Voltage (Voc), the maximum possible voltage a cell can supply, corresponding to the point at which the current density under illumination (JL) is zero, limited by the energy gap between the HOMO of the polymer and the LUMO of the fullerene [240].

Short Circuit Current Density (Jsc), the current density under illumination that flows at zero applied voltage (Va), caused by the presence of an electric field due to the asymmetry in ionization energies/work functions of the anode and cathode.

Fill Factor (FF), by relating the maximum power (Pmpp = JmppVmpp) that could be drawn from the device to the open circuit voltage and short circuit current; it describes the quality of diode behavior of the solar cell:

32

ocsc

mppmpp

VJVJ

FF = Equation 9

Photovoltaic Power Conversion Efficiency (PCE) is the common figure of merit, connecting the three parameters as:

in

ocsc

PFFVJ

PCE = Equation 10

where Pin is the incident light under standard test conditions, which include the temperature of cell 25 °C, light intensity 1 kW/m2 and AM 1.5 spectral distribution of light.

Figure 4.2: Typical current-voltage characteristic of a bulk heterojunction solar cell, along with the characteristic points. The green point denotes the maximum power that the device can deliver

The solar cell device can be represented by a one diode model, as shown in Figure 4.3. In this equivalent circuit, the photocurrent generated by the illumination of the solar cell is represented by the current source, where the current flows in inverse direction as compared to the diode. The applied voltage is represented by the voltage source. The ohmic contributions of the electrodes, the semiconductor-metal contacts and the resistivity of the active layer materials are summarized as series resistance. Its value decreases with decreasing thickness, and increasing temperature and light intensity [241], and in case of ideal solar cell, this resistance tends to zero. The shunt resistance represents the leakage current across the device. Its value rises with decreasing thickness and lowering of the light intensity, and in case of ideal solar cell, its value tends to infinity.

There are three contributions to the current in this circuit, diode current as described by Shockley relation, current through the shunt resistance and the photogenerated current Jph. Due to the series resistance, the applied voltage is reduced. It can be described as:

33

phsh

sasa I

RIRVIRV

nkTqII

1exp0

Equation 11

where I0 is the dark or reverse saturation current, Iph the photocurrent, n the diode ideality factor, Va the applied voltage, Rs the series and Rsh the shunt resistance.

The photocurrent of bulk heterojunction solar cells is determined by the amount of absorbed photons, exciton dissociation, charge transport and collection [242,

243].

Figure 4.3: Equivalent circuit of a solar cell

4.2.1 Long-Term Stability Measurements As described in more detail in the previous section, upon exposure to UV-Vis radiation, temperature and atmospheric conditions, the performance of solar cell devices deteriorates. To track the decay development and predict their useful lifetime, the J(V) characteristics of the cells are measured continuously at set periods under defined conditions. The controlled conditions include temperature, radiation with defined intensity and spectrum and atmospheric conditions, as required by the ISOS international consensus on stability testing of organic solar cells [31, 244, 245].

In this study, a metal halide lamp with 1 kW/m2 and the emission spectrum as shown in Figure 4.4 was used; the temperature as well as the humidity during the experiment were simultaneously recorded and the system typically self-regulated to a temperature of about 45 °C and relative humidity below 15 % in the direct vicinity of the solar tester.

34

300 400 500 600 700 800 90010000.0

0.2

0.4

0.6

0.8

1.0

1.2

Inte

nsity

(a.

u.)

Wavelength (nm)

AM 1.5 Lamp

Figure 4.4: Emission spectrum of the metal halide lamp and the standard AM 1.5 spectrum

4.3 Fourier Transform Infrared Spectroscopy Infrared spectroscopy gives information on molecular structure through the frequencies of the normal modes of vibration of the molecule. Normal mode of vibration is one in which each atom executes a simple harmonic oscillation about its equilibrium position. All atoms move in phase with the same frequency while the center of gravity of the molecules does not move. A molecule can be modeled as balls representing atoms and springs representing bonds between them. Vibrations involve stretching and bending of the springs together with motion of balls. The frequency of vibration of two balls of mass m connected by a spring with force constant k can be calculated as:

mk

πν

21

=

Equation 12

The force constant is a measure of the resistance to stretching of the spring. The force needed to displace the mass m by distance x is F = -kx. This method, based on classical mechanics, can be applied to more complex structures, and it forms the basis for interpretation of vibrational spectra.

There are 3N-6 normal modes of vibration of a molecule, where N is the number of atoms. Each atom has three degrees of motional freedom, which represent motion in x, y and z directions. Thus N unconnected atoms have 3N independent motions. However, in case of molecules, where atoms are connected, the motions are not independent. Three motions become translations of the molecule, where all atoms move simultaneously in x, y and z directions. Another

35

three are rotations, where all atoms rotate in phase about the x, y and z axes. In the remaining 3N-6 motions the internuclear distances and bond angles change without moving the center of gravity of the molecule.

The IR spectrum is obtained upon absorption of radiation in IR frequency range, due to the molecular vibrations of the functional groups contained in the polymer chain. In infrared absorption, energy is transferred from incident radiation to the molecule, and a quantum mechanical transition occurs between two vibrational energy levels, E1 and E2. The difference in energies is directly related to the frequency of the electromagnetic radiation:

νhEEE ==∆ 12 - Equation 13

where h is Planck’s constant. The frequency of vibration of the molecule directly corresponds to the frequency of absorbed infrared radiation.

Infrared spectra can be obtained by either dispersive or interferometric methods. Dispersive instruments record the spectrum in the frequency domain, whereas interferometers record the spectrum in the time domain. The latter gives an interferogram which must be transformed to the frequency domain by means of a Fourier transformation to obtain the infrared spectrum [246-248].

This method is widely applied to polymer degradation studies. During thermal or photooxidative aging, the structure of the materials alters, and new chemical groups are formed. The spectral maxima of these new groups can be distinguished from the non-degraded polymer, by detecting the occurrence of characteristic spectral features [249-253]. There are five major regions of FTIR spectra that are significant for identifying thermal and photooxidation aging [249]:

1) hydroxyl stretching of hydroperoxides, in the range from 3225 to 3640 cm-1

2) carbonyl stretching of anhydrides, esters, ketones, aldehydes and carboxylic acids, in the range from 1670 to 1785 cm-1

3) stretching of ionized carboxylates and salts of carboxylic acids, in the range from 1540 to 1610 cm-1

4) stretching of unsaturations, near 1644 cm-1 with peak broadening below 1510 cm-1 because of increasing conjugation 5) unsaturation deformation of internal, vinyl ends and pendant vinyl, in the range from 870 to 1000 cm-1

36

4.4 Cyclic Voltammetry Cyclic voltammetry is a standard method for characterization of electrochemical processes, used to identify electrode potential values indicative of the onset, cessation etc. of the electrode processes such as oxidation, reduction of species from solution, metal dissolution, adsorption, desorption. Information about the thermodynamics of the redox systems, energetic positions of HOMOs and LUMOs, kinetics of chemical reactions coupled to the electron transfer step and the rate of this reaction can be obtained.

Cyclic voltammetry setup used in this study, as shown in Figure 4.5, contains an electrolysis cell, a potentiostat, a current-voltage converter and a data acquisition system. The electrolysis cell used in this study consists of three electrodes, working (WE), auxiliary (AE) and reference (RE) electrodes, in electrolytic solution which provides ions to the electrodes in the oxidation and reduction processes. The potential applied using the potentiostat on the working electrode is linearly varied with time, while the reference electrode maintains a known constant potential.

Figure 4.5: Cyclic voltammetry measurement setup with the 3-electrode cell (Adapted from Gosser et al. [254])

It is important that the electrodes do not get polarized away from the equilibrium and that the measurements are conducted in absence of current. However, in non-equilibrium conditions, the current does flow, which polarizes the electrodes and displaces them from the equilibrium potential. To avoid that, a third electrode, counter or auxiliary, is used as shown in Figure 4.5. The reference electrode is connected to the potentiostat at the point with very high input impedance, resulting with the current being conducted away from the signal source to the working electrode, which then develops the potential necessary to oppose the current originating at the working electrode. In this way the

37

reference electrode stays protected from the current flow, and it remains constant during the measurement [255].

Cyclic voltammetry measures the current that develops in an electrochemical cell under conditions where voltage is in excess of that predicted by the Nernst equation:

[ ][ ]0

00

RedOx

ln+'=nFRT

EE appl Equation 14

where Eappl is the applied potential, n is the number of electrons exchanged by Ox/Red couple, E0’ is the formal potential of the couple and [Ox]0 and [Red]0 are the respective concentrations at the electrode surface [255].

Nernst equation is a thermodynamic relation, and it is valid only in equilibrium conditions, in absence of net current flow. In potential controlled experiments involving redox couples (Ox/Red), with fast electron transfer, it is possible to adjust the surface concentrations of Ox and Red to the ratio predicted by Nernst equation. Thus, when the overall current flowing across the cell is not controlled by electrochemical kinetics, the concentrations of electroactive species at the surface of the electrode can be determined as a function of the applied potential, according to Nernst equation [255].

Figure 4.6: (left) Excitation potential, and (right) the voltammogram of a reduction-oxidation reaction (Adapted from Quiroga et al. [256])

The working electrode potential is measured against a reference electrode with a constant potential, and the resulting applied potential produces an excitation signal as shown in Figure 4.6 (right). In the forward scan, the potential scans positively, starting from a higher potential (a) and ending at a lower potential (d). At potential (d) the voltage has caused the oxidation or reduction of the analyte, and is referred to as the switching potential. The reverse scan takes place from (d) to (g), where the potential scans positively. This cycle, consisting

38

of reduction from (a) to (d) and oxidation from (d) to (g), can be repeated and the scan rate varied [256-258].

The plot of the actual electrode potential versus the reference electrode against the current at the working electrode yields a cyclic voltammogram, as shown in Figure 4.6 (left). The reduction process starts at the initial potential (a), where only the oxidized form of the reversible redox couple (Ox) and no reduced (Red) are present, and lasts till the switching potential (d) is reached. The potential is scanned negatively to cause reduction, and the resulting current is termed cathodic ipc. The ratio (Ox)/(Red) decreases according to the Nernst equation, and the corresponding cathodic peak potential Epc occurs at (c) at which all of the substance at the surface of the electrode has been reduced. From the switching potential (d) on, the potential scans positively from (d) to (g), resulting in anodic current ipa and oxidation. The anodic peak potential Epa (f) is reached when all of the substance at the surface of the electrode has been oxidized. These characteristic peaks of the voltammogram reflect the continuous change of the concentration gradient at the electrode surface with time, while the formal potential E0’ is the expression of relative stability of the two redox partners in the medium:

2+

='0 pcpa EEE Equation 15

The measured formal potentials of the first oxidation and reduction peaks are correlated with HOMO and LUMO levels. As the vacuum level potentials of the common reference electrodes are known, the HOMO and LUMO energies can be approximated [259-261]. HOMO represents the energy required to extract one electron from a molecule, thus an oxidation process, while LUMO represents the energy necessary to add an electron to a molecule, thus a reduction process.

39

Chapter 5

Experimental Results and Discussion

In this chapter experimental results are presented and discussed. In the first section I examined morphological degradation. Since the morphology of optimized solar cells is not corresponding to thermodynamic equilibrium, changes in morphology are inevitable. As the performance of the devices is dictated by the morphology of the active layers, such changes can exert strong consequences on the performance of the devices. As the goal of my thesis is to photochemically stabilize the active layers, it is of importance to be able to distinguish between the different factors which lower the performance. This section is therefore used as a baseline, allowing distinguishing between the morphologically and photochemically caused decay in performance.

In the second section, I propose a method to photochemically stabilize organic solar cells, by implementing portions of a third component into the active layer. Different classes of stabilizing compounds were tested, and three successful mechanisms could be identified, which allowed a drastic improvement in the lifetime, leading to more than triple accumulated power generation. The first mechanism includes inhibiting the propagation step using hydrogen donors, which compete with the hydrogen abstraction from the polymer. The second one slows down the initiation step using UV absorbers, which prevent the polymer from being excited by absorption of the harmful UV radiation. And the third one prevents the chain branching step using hydroperoxide decomposers, thus competing with hydroperoxide homolysis which would otherwise yield two highly reactive radicals, and producing instead non-radical products. Such ternary blend active layers were optimized by concentration and their long-term stability has been tested under 1 Sun illumination, 45 °C and 15 % relative humidity. The stabilizing effectiveness of tested compounds is thoroughly discussed in terms of their chemical structure, morphological compatibility with the other active layer components and possibility of electronic trap states formation.

40

5.1 Morphological Degradation

Important shortcoming of the organic solar cells is their low charge carrier mobility and lifetime, which gives special importance to obtaining and maintaining the optimal morphology within the device. To balance the charge carrier separation on one side, and charge transport on the other, a continuous path in the form of interpenetrating network of polymer and fullerene is needed [262]. However, as the optimal morphology is not thermodynamically stable, the fullerene molecules in blends with polymeric donors tend to diffuse and recrystallize after deposition, which is especially pronounced with increase in temperature [263]. This process accelerates under thermal exposure, and eventually leads to a morphological destruction of the active layer, including decreased fullerene percolation and reduced charge transport, resulting in decreased performance of solar cell devices [264].

In order to evaluate data on photochemical stability, it is important to distinguish it apart from the morphological degradation. Findings from this part of my thesis therefore serve as a sort of baseline for the photochemical stabilization study which will be discussed in the following section.

Big efforts have already been invested in monitoring morphological degradation using very precise and complex, time demanding methods [265, 266], e.g. X-ray techniques [267-274], AFM [275-278], SEM [32], TEM [279, 280], spectroscopic ellipsometry [281-286], electron tomography [95, 262, 287, 288] and AC chip calorimetry [289]. In this experiment, the attention is drawn to a simple and fast method of determining critical stages of morphology degradation and phase separation, which can be applied for stabilization studies. The applicability of this method is demonstrated on standard P3HT:[60]PCBM solar cells.

The work presented in this section has been partially published in my publication “Methods in determination of morphological degradation of polymer:fullerene solar cells” [290], and “Multiple stress degradation analysis of the active layer in organic photovoltaics” [45].

5.1.1 Tracing the Scale of Phase Separation using Microscopic Techniques A set of films with varying P3HT:[60]PCBM blend ratios (3:2, 1:1, 2:3) and varying annealing times (0, 5, 15 and 30 minutes, 1, 2, 4, 8, 17 and 21 hours) were annealed at 120 °C. The films were evaluated using UV-Vis transmission and photoluminescence spectroscopy, optical and atomic force microscopy.

41

Commonly, combinations of AFM, TEM and other labor extensive techniques are applied to study the scale of phase separation. Here I present a simple, fast and low-cost alternative way to collect the relevant morphology degradation data.

Figure 5.1: Optical microscopy images (magnification x50) of 3:2 (left column), 1:1 (middle column) and 2:3 (right column) P3HT:[60]PCBM solutions, as-cast (first row), annealed at 120 °C for 2 h (second row), for 4 h (third row), for 8 h (fourth row), for 17 h (fifth row) and 21 h (sixth row). The scale bar is always 100 m

Conventional optical microscopy, shown in Figure 5.1, provides direct information whether PCBM clusters have been formed. Once they have been formed, their growth can be conveniently tracked [291-294]. This method can, however, be used only after a certain size of the crystallites has been reached, as the optical microscopes are limited in resolution to about 1 m. For that reason they cannot

42

be used to exactly trace the initial formation of the small PCBM clusters in the early stage of morphological degradation.

Figure 5.1 summarizes the optical microscopy results obtained on the films studied for various P3HT:[60]PCBM blend ratios and annealing times. It is obvious that the 3:2 blend is much more stable than the other two, which are already massively degraded after 2-4 hours of exposure to heat.

In fact, the scale of phase separation can be modeled by taking into account the number and the size of the crystallites formed, as shown on Figure 5.2. The optical microscopy images of films annealed for longer times, corresponding to a later stage of degradation, were processed and the number of clusters and their areas were counted. This reveals the statistical distribution of the PCBM clusters through different stages of degradation. This way the degradation can be followed during cluster growth and coarsening, starting from the point where a large amount of small crystals are formed to the point where those small ones merge into a few large ones.

0 10 20 30 40 50 60

0

250

500

750

1000

1250

1500

1750

No

of c

lust

ers

per

area

[1/

mm

²]

Cluster area [µm²]

21h 17h 8h

1:1 P3HT:PC[60]BM 120°C

Figure 5.2: Statistical distribution of the PCBM crystals growth on the example of 1:1 P3HT:[60]PCBM films, annealed at 120 °C for 8 h, 17 h and 21 h

The small PCBM crystallites grow according to Ostwald mechanism, diffusing out of the original polymer-fullerene blend to eventually form thermodynamically favored micrometer large crystallites and leave behind the PCBM-depleted polymer domains [264, 295, 296].

43

Figure 5.3: Atomic force microscopy images of P3HT:[60]PCBM 1:1 films, (left) as-cast and (right) after annealing at 120 °C for 21 h

Zooming into the smaller length scale, the PCBM-depletion region is investigated by AFM, as shown on Figure 5.3. They reveal a roughening of the surface, confirming the increase in nanocrystallization of P3HT and, parallel to it, diffusion of PCBM out of the blend [275-278, 297]. Of course, it takes more effort to obtain data using this technique.

5.1.2 Modeling the Extent of Phase Segregation using Optical Spectroscopy For all of the films, series of transmission and photoluminescence measurements have been recorded, as shown on Figure 5.4. This technique gives statistical information on the morphological integrity of the polymer:fullerene blend.

It has already been suggested that an increase in photoluminescence is a qualitative measure for the increase in phase separation [298]. This study shows that even transmission spectra per se are able to convey important information concerning the stages of phase separation.

Aggregation of the PCBM clusters implies that its distribution is reduced to a smaller area, so the amount of light that reaches it, and could thus be absorbed, also decreases. This reduction in cross-section area leads to a decrease of the PCBM peak at 325 nm in the absorbance shown on Figure 5.4. At the same time, increased exposure to high temperature increases the crystallinity of P3HT, leading to an increase of the P3HT peak (~525 nm) with the ongoing annealing times [297]. Light scattering is also increased upon the surface segregated growth of PCBM microcrystals, which can be observed in the longer wavelength range in the optical absorbance spectrum (650 nm onwards), below the band gap of the polymer.

44

A good correlation between the transmission and photoluminescence spectra and certain stages of phase separation as detected by optical microscopy was found. A strong correspondence was found for the transition from nanometer to micrometer scale phase separation.

300 400 500 600 700 8000.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8 unannealed

5 min 15 min 30 min

1 h 2 h 4 h 8 h 17 h 21 hA

bsor

banc

e [a

.u.]

Wavelength [nm]

3:2 P3HT:[60]PCBM 120°C

micrometer phase separation

PCBM peak decreases

P3HT peak increases

500 600 700 800 9000.0

0.5

1.0

1.5

2.0 unannealed 5 min 15 min 30 min 1 h 2 h 4 h 8 h 17 h 21 hN

orm

aliz

ed P

L [a

.u.]

Wavelength [nm]

3:2 P3HT:[60]PCBM 120°C

decreased PL quenching

micrometer phase separation

300 400 500 600 700 8000.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8 unannealed

5 min 15 min 30 min

1 h 2 h 4 h 8 h 17 h 21 hA

bsor

banc

e [a

.u.]

Wavelength [nm]

1:1 P3HT:[60]PCBM 120°C

P3HT peak increases

micrometer phase separation

PCBM peak decreases

500 600 700 800 9000.0

0.5

1.0

1.5

2.0 unannealed 5 min 15 min 30 min 1 h 2 h 4 h 8 h 17 h 21 hN

orm

aliz

ed P

L [a

.u.]

Wavelength [nm]

1:1 P3HT:[60]PCBM 120°C

decreased PL quenching

micrometer phase separation

300 400 500 600 700 8000.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

unannealed 5 min 15 min 30 min

1 h 2 h 4 h 8 h

17 h 21 h

Abs

orba

nce

[a.u

.]

Wavelength [nm]

2:3 P3HT:[60]PCBM 120°C

P3HT peak increases

PCBM peak decreases

micrometer phaseseparation

500 600 700 800 9000.0

0.5

1.0

1.5

2.0 unannealed

5 min 15 min 30 min

1 h 2 h 4 h 8 h 17 h 21 h

Nor

mal

ized

PL

[a.u

.]

Wavelength [nm]

2:3 P3HT:[60]PCBM 120°C

decreased PL quenching

micrometer phase separation

Figure 5.4: (left) Absorbance and (right) normalized photoluminescence of the films of 3:2 (top row), 1:1 (middle row) and 2:3 (bottom row) P3HT:[60]PCBM blend ratios, as-cast and annealed in different time intervals from 5 minutes up to 21 h

At the stage of phase separation at which the clusters are growing out of the film by a large number, characteristic changes can be observed in the transmission and photoluminescence spectra. Photoluminescence is drastically increased, and

45

the fullerene peak in the absorbance spectra of those films is almost completely gone.

For the 2:3 and 1:1 blend films between 2 and 4 h of annealing at 120 °C, a micrometer scale phase separation and formation of a PCBM-depleted zone are observed on the optical microscopy images, which can be directly correlated to the grouping of the absorption and photoluminescence curves. The same can be observed with 3:2 blend films, although as the PCBM content is lower, this transition occurs later, between 8 and 17 h.

The question may arise why the PCBM peak disappears at the later stages of degradation. To answer that, a simulation of the measured films using the Beer-Lambert-Bouguer Law was conducted (see Equation 1). I start from the simplified assumption that the film consists of 3 possible phases: the intermixed P3HT:[60]PCBM blend, and the pristine P3HT and PCBM phases. During the process of thermally induced morphological degradation, the areas, with respect to the plane of incident light, of the three phases change, and this can be observed in the absorbance curves. In Figure 5.4 three characteristic groups of absorbance curves are observed, which correspond to different degradation stages. The initial state of the film, corresponding to the curve 1 in Figure 5.4, is graphically shown in Figure 5.5 on the left which represents the well intermixed P3HT:[60]PCBM film. As the PCBM clusters start developing, pristine P3HT areas start appearing around them, as shown in Figure 5.5 in the middle, and as observed in the group 2 in Figure 5.4. In the final stage of degradation, the P3HT and PCBM phases are completely separated, as shown in Figure 5.5 on the right, corresponding to group 3 of the curves in Figure 5.4.

Figure 5.5: Graphical representation of the three stages observed in the thermally induced morphological degradation process: on the left the initial non-annealed well intermixed film (yellow) on the glass substrate (grey), followed by formation of PCBM crystallites (green) and pristine P3HT phase (magenta) amid the well intermixed phase (yellow) as shown in the middle, and finally completely phase separated film consisting of the two pristine materials shown on the right

Considering these three phases, the absorbance of the whole film consists of the sum of the absorbances of each of the phases:

0++=0

dα

total

mixdα

total

FPdαdα

total

FT mixPPPPFF eAA

eA

AAee

AA

II ---- -

Equation 16

46

This absorbance model uses the optical microscopy data, from which the areas of the three phases, Atotal, AF, AP, Amix, can be obtained, and the absorption coefficients of the pristine and blended materials, F, P, mix, obtained using spectroscopic ellipsometry. The thickness of the initial non-degraded film, d0 = Vtotal/Atotal = Vmix/Amix, was determined using the profilometer to be 100 nm. The thicknesses of the pristine phases of the two components are calculated using the volumes of the corresponding phases:

pP

mixtotalP

FF

mixtotalF Af

VVd

AfVV

d--

= ,= Equation 17

where fF and fP are volume ratios, calculated using the densities, P [299] and F [300], of each of the components, and the P3HT:[60]PCBM mass blend ratio in the solution, mF/mP:

PP

FFP

FF

PPF ρm

ρmf

ρmρm

f +1= ,+1= Equation 18

To compare this model with the experimentally obtained absorbance curves, the three phase model from Equation 16 was entered in the expression for absorbance, as noted in Equation 19:

0ln-=

II

Abs T Equation 19

Obtained simulated absorbances are shown in Figure 5.6. Besides some minor deviations due to the scattering of the samples, not considered in the model, these curves reproduce quite well the experimental observations.

300 350 400 450 500 550 600 650 7000.0

0.2

0.4

0.6

0.8

1.0

Sim

ulat

ed A

bsor

banc

e [a

.u.]

Wavelength [nm]

unannealed 2h 4h 8h 17h 21h

Figure 5.6: Simulated absorbance of films at the advanced stage of morphological degradation

47

5.1.3 Observation of Thermodynamically Driven Morphological Changes in OPV Devices Two sets of layer stacks consisting of glass/ITO/PEDOT:PSS/P3HT:[60]PCBM 3:2 were stored for increasing periods of time, at two different conditions, one set at moderate temperature in dark air and the other one, at room temperature in inert glove box atmosphere.

The analysis of the temporal evolution of the reverse current density, as shown in Figure 5.7, brings to light interesting morphological aspects related to the performance of these devices.

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0-100

-50

0

50

100

150

200

250

Reference 1h 2h 4h 8h

Cur

rent

Den

sity

(m

A/c

m2 )

Voltage (V)

Dark Air 50°C P3HT:[60]PCBM

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0-100

-50

0

50

100

150

200

250

Reference 1h 1w 5w

Cur

rent

Den

sity

(m

A/c

m²)

Voltage (V)

Dark Glove Box RT P3HT:[60]PCBM

Figure 5.7: Temporal progression of degradation of P3HT:[60]PCBM cells (left) at moderate temperature in air in dark, and (right) at room temperature in inert atmosphere in dark

It was shown previously that freshly spin coated pre-cathode annealed P3HT:[60]PCBM samples have a PCBM enrichment at the PEDOT:PSS interface, due to the preferential segregation of PCBM at the interfaces with a high surface energy [95, 105, 301]. With time, the PCBM layer will completely wet the interface, that is, a closed layer will be formed [282].

This formation of a wetting PCBM layer causes drastic changes in the band diagram of the device. Within a one dimensional metal-insulator-metal model (MIM), an internal potential drop occurs over the wetting layer, see Figure 5.8. In the forward direction, the injection of electrons and holes into the device is undistorted and can be described by drift and diffusion for both the homogenous active layer and the layer containing the thin wetting PCBM layer.

However, under short circuit conditions and under reverse bias, in presence of wetting layer, possible tunneling effects have to be considered. Under short circuit conditions the internal potential partially drops over the wetting layer, causing a reduction of the gradient in the quasi Fermi potentials across the rest

48

of the film. Nevertheless, the remaining gradient is sufficient for charge carrier extraction. The formed hole barrier at the anode can be overcome by direct tunneling of the charge carriers, if the barrier thickness is sufficiently small. Thus the formation of a thin PCBM-rich wetting layer might result only in a very slight reduction of the short circuit current, which is in agreement with experimental observations. For large reverse bias, the gradients in the HOMO and LUMO level of the PCBM interface layer might become so large that, in combination with the small barrier width, electrons start tunneling from the fullerene HOMO to the LUMO level, thus increasing the current density compared to the homogenous active layer. The tunneling probability through the triangular barrier formed by the applied electric field can be described in terms of Fowler-Nordheim tunneling, and it decreases exponentially with increasing film thickness of the wetting layer.

Figure 5.8: Schematic MIM band diagrams of the active layer under illumination. On the left, a homogenous active layer is shown for cases of (up) positive voltage applied on the anode, (middle) short circuit conditions, (bottom) negative voltage applied. On the right, the same three cases are visualized for the active layer containing a very thin PCBM wetting layer at the anode. The inset in the bottom right diagram focuses in the wetting layer region and shows tunneling of the electrons from the fullerene HOMO to the LUMO

49

But how can the increase of the current density under reverse bias with time be explained? In the early stages, the interface to the anode is already PCBM rich, but polymer percolation paths are present in large amounts. Therefore, the often discussed one dimensional model, as shown in Figure 5.8, is oversimplified and a higher dimensionality model should be applied. The percolation paths reduce the voltage drop near the interface, thus decreasing the band bending and therefore the probability of tunneling from HOMO to LUMO. With time, the closed PCBM enrichment layer will be formed, leading to a state closer to the case of the 1D model, which is then able to fully describe the mentioned tunneling behavior. Thus, the area of the device where the polymer percolation exists and the current increase due to tunneling through the fullerene layer are inversely proportional. As the mobility of PCBM at room temperature is not very high, this process takes a longer time (more than one week). However, with rise in temperature, PCBM can diffuse faster, as shown in Figure 5.7, at 50 °C a closed layer was formed already after 4 h.

5.1.4 Short Summary I showed that characterization of phase scale separation via transmission measurements is very efficient for morphological degradation investigations. This approach can be easily extended to solar cells, where UV-Vis reflectance measurements can be used to obtain the same information. Although optical microscopy provides quick information, optical microscopy still holds two major drawbacks – the inability to give an accurate quantitative measure of the degradation, and its resolution limit which makes it impossible to measure the earlier stages of degradation. AFM offers much more detailed information in the latter case, but such a scan requires more experience and time. My study shows that UV-Vis spectroscopy is as applicable in the earlier as in the later stages of morphological degradation, giving precise information on the differences that occur upon short periods of annealing. Time consuming techniques, such as AFM, are not necessary for successful tracking of the morphological changes in the films – optical spectroscopy methods such as UV-Vis measurements can be used to efficiently determine different stages of polymer-fullerene de-mixing during morphological degradation and aging.

Finally, I report on the formation of a thin wetting PCBM layer at the anode with storage time, accelerated by exposure to moderate temperature. I explain this increase in the current densities for the large reverse voltages by tunneling from the fullerene HOMO to the LUMO, as a result of a large gradient in the electric potential across this layer. This study shows that the morphological degradation of P3HT:[60]PCBM cells at working conditions evolves at a rather slow time scale, allowing to focus in the following section solely on the effects of photochemical stabilization.

50

5.2 Long-term Stabilization using Stabilizing Additives

The lifetime of polymer solar cells is severely limited by photoinduced oxidation, which occurs due to UV-radiation and ingress of oxygen and water into the solar cell device [29-45]. Thus one way to reduce degradation is to prevent water and oxygen diffusion into the device by encapsulation [302-306]. In addition, getter materials can be introduced to further reduce the influence of radicals within the encapsulated device [43, 307-310]. A second way is to effectively reduce the magnitude of photochemical changes by introducing UV-blocking layers within the sealing of the device [311-313] or by selecting active layer materials which are inherently more stabile against oxidation [314-317]. However, it is not possible to fully prevent water or oxygen from entering the device. Materials currently synthesized to achieve high stability lag efficiency-wise behind the modern high-performing polymers [318-321], and vice versa [318, 322]. For additional information on this topic see Appendix A. Thus, further stabilization of the active layer against chemical degradation remains a fundamental problem.

In this study I present a novel way of preventing photooxidative degradation of organic solar cells using stabilizing additives, thus extending their long-term stability. Three main mechanisms of interfering with the photooxidation cycle were tested. In the first part, hydrogen donors were examined, with the task of inhibiting propagation by neutralizing radicals produced during photooxidation. In the second part, UV absorbers were tested, with the task of slowing down the initiation step by absorbing UV radiation and non-radiative deactivation of excited chromophores. In the third part, hydroperoxide decomposers were investigated, which prevent chain branching reactions by decomposing reactive hydroperoxide degradation intermediates to non-radical products.

The work presented in this section has been partially published in my publication “Long-term Stabilization of Organic Solar Cells using Hindered Phenols as Additives” [323], and “Long-term Stabilization of Organic Solar Cells using UV Absorbers as Additives” [324].

5.2.1 Hydrogen Donors Hydrogen donating type of antioxidants, which are common stabilizers for commercial insulating polymers, were blended in different mass concentrations with the polymer:fullerene mixture. This study shows that not every stabilizing additive causes an improvement of the long-term performance of the solar cell device. For the tested reference system P3HT:[60]PCBM, only one of the seven tested hindered phenols, namely octadecyl 3-(3,5-di-tert-butyl-4-hydroxy-phenyl)propionate, has shown the ability to significantly increase the long-term stability under ISOS-3 conditions of encapsulated OPV devices, increasing the

51

accumulated power generation by a factor of roughly four as compared to reference devices with no additive added.

The additives introduced into the P3HT:[60]PCBM system are shown in Figure 5.9 and Table 5.1. These compounds belong to the chemical group of sterically hindered phenols, and they are commercially available stabilizers for a wide variety of insulating polymers, such as polyolefins, rubber, styrenic polymers, polyesters, polyamides, polyacetals, polycarbonates, polyurethanes etc. [106].

Figure 5.9: Chemical structure of the investigated hindered phenols

52

# Chemical Nomenclature 1 2,6-di-tert-butyl-4-methylphenol2,6-di-tert-butyl-4-methylphenol 2 2-2'-methylenebis(6-tert-butyl-4-methylphenol) 3 1,3,5-trimethyl-2,4,6-tris(3,5-di-tert-butyl-4-hydroxybenzyl)benzene 4 pentaerythritol tetrakis (3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionate) 5 octadecyl 3-(3,5-di-tert-butyl-4-hydroxy-phenyl)propionate 6 4-4'-thiobis(6-tert-butyl-m-cresol) 7 2,6-di-tert-butyl-4-((2,2,6,6-tetramethyl-piperidin-4-ylimino)-metyl)-phenol Table 5.1: List of investigated hydrogen donors. For convenience, throughout the manuscript the compounds are represented by the number shown in the first column

In contact with air and light, the polymer reacts with oxygen, which produces different oxidized polymer species and breaks the polymer chain, leading to irreversible failure in functioning of the devices [40, 170, 171]. Recently it has been shown by Hintz et al. that two concurrent mechanisms take place in degradation process, depending on the irradiation conditions [38, 183]. Under the UV radiation, the degradation proceeds as a radical reaction [40, 170, 171] starting at the -carbon of the alkyl side chain, and results with simultaneous degradation of the -conjugated system and the side chain. Under visible light, however, the reaction involves a photosensitized species, possibly singlet oxygen [163, 164, 169, 172, 173] that primarily destroys the -conjugated system of the polymer, leaving the side chain almost unaffected. In white light conditions, both of the mechanisms are active, but since the radical chain mechanism has a higher effectiveness, it dominates even though only a low UV radiation fraction is present [38, 183].

Figure 5.10: The principle of stabilization using hydrogen donors, such as hindered phenols AH represents the hindered phenol

The generalized scheme of degradation inhibition via hindered phenol mechanism [325-328] is depicted in Figure 5.10. In the chain propagation step, the peroxy radical PO2•, that was created by the fast reaction of molecular oxygen with the polymer alkyl radical P•, further reacts with polymer, and by abstracting a hydrogen atom creates a new polymer radical. In presence of hindered phenols (denoted in the figure with AH), the rate of this reaction, defined by the activation energy needed to abstract the hydrogen from the polymer, competes with the rate of abstraction of substantially more abstractable hydrogen from the hindered phenol [325-328]. The stability of the phenoxyl radical created from the additive is influenced by the steric hindrance of the hindered phenols (tert-butyl

53

groups in the 2- and/or 6-position), preventing them from reacting with the polymer and from dimerization with other phenoxyl radicals.

5.2.1.1 Ternary Blends Implementing a third component (additive) into the active layer of organic solar cells, two obstacles may be encountered.

Electronic Trap States As organic solar cells are inherent electronic optoelectronic devices, the energetic levels (HOMO/LUMO) of each of the constituents of the bulk-heterojunction are of fundamental importance for the device performance. Besides the LUMO (HOMO) offset necessary for the photoinduced charge transfer between the polymer and fullerene [72, 329, 330], the energetic position of any additional blend constituent might impair the maximum reachable power conversion efficiency (PCE) [331-339]. Gap states / trap states Etrap within the effective band gap of the photoactive layer, Donor

HOMOAcceptorLUMOeffgap EEE -=, , are centers for non-geminate recombination of

free charge carriers. The recombination rate of electrons Re,trap (holes Rh,trap) via trapped holes (electrons) can be given in the form:

trappedee

trape npεμqβ

R =, trappedhh

traph pnεμqβ

R =, Equation 20

where µ is the charge carrier mobility, n (p) the electron (hole) density, and ptrapped (ntrapped) the density of trapped holes (electrons). The dimensionless factor corresponds to the probability of a free charge carrier to find a trap level and is therefore strongly morphology dependent. A strong chemical incompatibility and resulting decomposition of the blend matrix might lead to the formation of large agglomerates with high trap density, but at the same time in a reduction of the probability for free charge carriers to find those traps as the effective interface area decreases. This is due to the localized charge transport via hopping, which is in contrast to delocalized band transport of free charges in classical semiconductor physics, e.g. silicon solar cells. For this reason, the well-known Shockley-Read-Hall (SRH) recombination [340, 341] found in classical semiconductors cannot be used in a one to one manner on organic solar cell devices, but might be used to give useful guidance.

Similar to the case of SRH recombination, photogenerated charges, generated at rate G, are captured by localized trap states with the rates Re.trap and Rh.trap, and are thermally emitted with rates Ge,trap and Gh,trap back to the LUMO and HOMO of the active layer, see Figure 5.11.

54

Figure 5.11: Trapping of electrons and holes produced by the photoinduced charge transfer by a trap level Etrap with the rates Re,trap and Rh,trap. Electrons (holes) occupying the localized trap state, can be thermally emitted back to the acceptor LUMO (donor HOMO) with rate Ge,trap (Gh,trap)

Following Boltzmann statistics, the generation from trap states or de-trapping Gtrap, depends exponentially on the energetic position of the trap state with respect to the HOMO of the donor (that is, LUMO of the acceptor) of the photoactive material:

TkEE

nGB

trapAcceptorLUMO

trapetrape exp,,

TkEE

pGB

DonorHOMOtrap

traptraph exp,

Equation 21

Thus at room temperature holes or electrons which are trapped in an energetic state near the middle of the effective band gap or just far away from either the donor HOMO or acceptor LUMO, most likely stay localized until their recombination with a free charge carrier. It follows that trap states close to the middle of the band gap will possibly lead to the biggest loss of free charge carriers and a reduction of the maximum short circuit current and open circuit voltage.

0.0 0.5 1.0 1.5 2.0 2.5 3.0

0

1

2

3

4

5

6

7

8

Jsc (

mA/c

m²)

Concentration (%)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

PCE

(%)

0.0 0.5 1.0 1.5 2.0 2.5 3.0

0

100

200

300

400

500

600

Voc (

mV)

Concentration (%)

0

10

20

30

40

50

60

FF (

%)

Figure 5.12: Photovoltaic parameters in dependence of concentration of Additive #7 in the active layer: (left) short circuit current (magenta) and efficiency (green); (right) open circuit voltage (green) and fill factor (magenta). The parameters of the reference device (with no additive added) are represented by respectively colored lines

55

Figure 5.12 shows the photovoltaic parameters in dependency of concentration of Additive #7. The strong exponential decay in all of the parameters even at relatively small concentration (up to 3 %), suggest that the cause might be the introduction of a trap level.

The cyclic voltammogram of this compound, shown in Figure 5.13 (left), revealed a -4.06 eV LUMO level, and an additional oxidation peak with an onset at -4.6 eV. As the optical band gap was estimated from the Figure 5.13 (right) to 2.8 eV, thus giving a HOMO level of -6.89 eV, the small oxidation peak is attributed to impurities [342-344] (various synthesis residue catalysts [345-347] and defects [348]) within the additive.

-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8-0.18

-0.16

-0.14

-0.12

-0.10

-0.08

-0.06

-0.04

-0.02

0.00

0.02

Cur

rent

(A

)

Potential (V vs. Ag/AgCl)

Eredonset = -0.29 V

Eoxonset = 0.25 V

0.0 0.5-0.01

0.00

0.01

200 300 400 500 6000.0

0.2

0.4

0.6

0.8

1.0

Abs

orba

nce

(a.u

.)

Wavelenght (nm)

Eabsonset

= 2.83 eV

Figure 5.13: (left) Cyclic voltammogram of Additive #7 in a dichloromethane solution containing 0.1 mol/L of the supporting electrolyte n-tetrabutyl-ammonium hexafluoro-phosphate (TBAPF6). The potential is given with respect to the Ag/AgCl reference electrode, the scan rate was 50 mV/s; (right) absorption spectrum was used to determine the optical band gap. UV-Vis absorption was measured on thin films spin coated on a CaF2 substrate

Note, in Figure 5.14, that the trap level is ~0.4 eV lower than the LUMO of PCBM and 0.4 eV bigger than the HOMO of P3HT. Thus the Additive #7 forms deep trap states within the effective band gap of the active layer. Already at a low concentration (0.3 %) of added Additive #7, solar cells have a power conversion efficiency prior to degradation, PCE0, of 0.02 % (compared to 1.48 % for the reference device without additives), due to a significant reduction of the open circuit voltage to 109 mV compared to 389 mV of the reference solar cells without additives. This reduction of the Voc is in perfect agreement with the considerations above, which predict a reduction of the open circuit voltage for traps near the band gap middle, and visualizes the importance of the knowledge on the energetic levels of each of the blend constituents. A morphological problem due to the addition of the additives could be excluded in all of the cases, see discussion below.

56

Figure 5.14: Energetic position of the electrochemical level of Additive #7, in comparison to the HOMO and LUMO level of the photoactive blend P3HT:[60]PCBM. Values are given with respect to the vacuum energy. The dotted level in the gap of Additive #7 represents the trap level, located within the effective active layer band gap marked in green

Morphology Issues The second highly important point that has to be taken into account when analyzing the solar cell performance is the influence of additional third compound on the morphology of the binary blend.

In order to achieve an optimum trade-off between charge separation, charge transport and charge extraction, often small amounts of stabilizing additives are used. Although their role is limited only to interaction with the blend constituents in liquid phase, and upon drying they are expected to have left the active layer, they can drastically alter the blend morphology [87, 89, 91, 349-353]. In this study, solid compounds are added which remain in the active layer, thus constituting a ternary blend. The thermodynamic interactions between the components in a ternary blend are significantly more complex, thus changes in morphology, which will in turn affect the performance of the devices, might occur [354-357].

Figure 5.15 shows the experimentally measured photoluminescence signal, as well as the AFM (Atomic Force Microscopy) topography of the reference P3HT:[60]PCBM film compared to the film containing one of the hindered phenols (Additive #7), both measurements obtained prior to degradation.

The reason of the slightly lower photoluminescence signal of the film containing Additive #7, compared to the reference and the rest of the films with hindered phenols, is obviously not in morphology, as shown by AFM imaging. Two possible factors might explain this. First, its band gap, unlike that of the rest of the additives (~4 eV), reaches further in the visible spectra (2.8 eV), as shown in Table B 1 in Appendix B. As the normalization was done with respect to the excitation laser wavelength, and the absorption of the film in that region is additionally higher due to the absorbance of the additive, the normalization might result in an underestimation of the photoluminescence. The other possibility is that the additive is capable of acting as an additional excitation quencher.

57

600 700 800 900 10000.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16N

orm

aliz

ed P

L (a

.u.)

Wavelength (nm)

Reference #1 #2 #3 #4 #5 #6 #7

Figure 5.15: (left) Photoluminescence signal of P3HT:[60]PCBM non-degraded films containing each of the seven hindered phenols (concentration as noted in Table 5.2), as well as the reference without any additive. The PL excitation wavelength was 445 nm (~2.78 eV), which is below the optical gap of each of the additives; (right) AFM topography image of the reference film and of the P3HT:[60]PCBM film containing Additive #7

This interpretation is further supported by a fill factor of over 50 % for the solar cells containing Additive #7 (see Table B 2 in Appendix B), which is comparable to the reference solar cell. In case the additive would act as compatibilizer for P3HT and PCBM, causing a finer intermix of both components across the blend, a much smaller FF would be expected, due to a drastic increase of the recombination of charge carriers or the buildup of space charges [243, 358-360]. The relatively unchanged morphology of the blend, especially the degree of polymer order was further supported by UV-Vis measurements. Nevertheless, in general, compounds other than those reported within this manuscript, might cause significant morphological changes, and thus affect the solar cell performance more drastically.

5.2.1.2 Bleaching Blend films containing one of the seven hindered phenols and a reference film without additive, were prepared on CaF substrates. Freshly prepared and degraded films were investigated via UV-Vis and FTIR spectroscopy to elucidate the effect of the antioxidant on the blend degradation upon direct illumination in ambient. Note that the active layer was not shielded from UV-radiation during aging. Figure 5.16 shows the normalized absorption spectra of all investigated films prior to degradation in air, as well as the relative absorption change upon film aging. In agreement with the morphology results reported above, the polymer order within all films stays unaltered independent of the stabilizing additive compared to the reference device, which is manifested in a nearly

58

identical ratio of the A0-0 and A0-1 absorption peaks, and a good overlap of the absorption spectra in the wavelength range between 300-800 nm after normalization to the maximum polymer absorption. Since all additives have shown large optical band gaps (~4 eV, detailed information can be found in the Appendix B in Table B 1), the blend absorption for each of the investigated films only differs in the high energy region.

200 300 400 500 600 700 8000.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

#1 #2 #3 #4 #5 #6 #7

A0-0

Nor

mal

ized

Abs

orba

nce

(a.u

.)

Wavelength (nm)

Reference

A0-1

200 300 400 500 600 700 800-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

#1 #2 #3 #4 #5 #6 #7

Reference(

Abs

orba

nce(

120h

) -

Abs

orba

nce(

0h)

) /

Abs

orba

nce

(0h,

500

nm)

(a.

u.)

Wavelength (nm)

Figure 5.16: (left) Normalized UV-Vis absorption spectra of the non-degraded films with additives and one reference with no additive on CaF substrates. Normalization was done with respect to the absorption peak at around 500 nm; (right) relative change in absorption of the films after 120 h of continuous 1000 W/m² illumination in ambient. Note that all spectra were normalized to the absorption peak at 500 nm of the corresponding non-degraded film

While there is no significant difference observable for the freshly prepared films, this changes drastically upon degradation. After 120 h of film aging under 1000 W/m² continuous illumination in air, all of the films show significant photo-bleaching, see Figure 5.16 (right). The bleaching of the ground state polymer absorption (S0-S1 transition) due to formation of a charge transfer complex between O2 and the polymer in early degradation stages [191, 361], as well as due to the creation of a photo-oxidized polymer species [29, 35, 36, 42] can be observed in the wavelength range 450-650 nm. Below 450 nm the “bleaching” is a consequence of a simultaneous morphological degradation including the formation of large fullerene clusters at elevated temperatures [290] and additional photochemical degradation of the fullerene derivative [44, 362]. The newly formed absorption band (above 650 nm) upon illumination was recently interpreted as the high energy electronic absorption (P2 band) of P3HT corresponding to the D2-D0 transition of the polymer radical cation [193, 361, 363, 364]. A second, more relevant, interpretation is the increase of light scattering due to morphological changes and the formation of PCBM clusters during degradation. Figure 5.17 shows the optical microscopy images of the reference film without the additive prior to and upon degradation, to support this conclusion.

59

0 5 10 15 20 25 300.0

5.0x10-4

1.0x10-3

1.5x10-3

2.0x10-3

2.5x10-3

3.0x10-3

3.5x10-3

4.0x10-3

Reference Degraded

Num

ber

of C

lust

ers

per

Are

a (µ

m-2

)

Cluster Size (µm2)

Figure 5.17: Optical microscope images of the reference film: (left) as-cast and (right) after 5 days of degradation under illumination (1000 W/m², ambient). Zoom 100x; (bottom) Frequency count of the cluster size for both samples

All of the films are significantly bleached upon degradation. As all of the tested additives are known to work as anti-oxidants in other polymer systems, these results might be somewhat unexpected, since photo-bleaching is addressed to the breaks in conjugation [36, 40, 170, 180, 214], and in case of P3HT it was found that the magnitude of absorption decrease is directly proportional to the number of oxidized thiophene rings [38].

All active layers containing phenols suffered from a significantly higher bleaching in the visible wavelength range compared to the reference film without any additive. This suggests increased thiophene ring destruction in the presence of the additives. Indeed, it was reported on the contribution of some products of consumption of hindered phenols that arise during polymer lifetime as a consequence of reactions with alkylperoxy radicals, principally the ultimate transformation products quinone methides [365-369], to discoloration of stabilized plastics during aging.

60

It was shown in a study by Hintz et al., that depending on the radiation wavelength, two different degradation mechanisms will take place, each attacking different part of the polymer [183]. Hindered phenols are preventing the radical degradation pathway [180] which affects both alkyl side chains and the conjugated -system, but it has no effect on the degradation via photosensitized species which primarily attacks the conjugated -system [172]. This explains why films that are protected by hindered phenols from the highly effective free radicals attack, still exhibit photobleaching.

It was already reported with MDMO-PPV cells, that the loss in absorbance is not the main responsible for the decrease in performance upon photooxidative degradation [370]. As photo-bleaching is subject to a loss in conjugation and thus backbone degradation, the side chain reactions do not contribute to the visual impression of degradation which is observable in simple UV-Vis measurements, however they will be of significant electronic importance.

This becomes clear considering two things. First, the absorption and thus extractable current density is more or less linearly dependent on the concentration of absorbers and their absorption coefficient. Second, the recombination rate of free charge carriers and excitons depends on the probability of finding traps or defects. It can be assumed that radicals and later stage degradation products lead to exciton trapping and recombination, due to a strong electron pulling effect, thus the number of extractable charges is proportional to the probability that no traps can be found. In case of homogenous distributed traps (volume concentration ct), the latter one can be expressed by the Poisson probability:

tVcep /1= Equation 22

where V is the volume accessible by the exciton via diffusion. Indeed, Bauld et al. found an exponential relationship between the solar cell efficiency and paramagnetic defect (radical) density [371].

By these considerations I want to emphasize that the visual appearance is not necessarily a good indication for good respectively bad device performance. Additional measurements yielding information about the fraction of reaction products are therefore of utmost importance. Furthermore, optical results should be always compared to the actual device performance.

5.2.1.3 Long-term Stability After investigating the optoelectronic and morphological properties of the P3HT:[60]PCBM films containing hindered phenols, solar cells containing each one of the seven antioxidants were tested with respect to the stability enhancement of the solar cells compared to reference solar cells without

61

additives. Encapsulated pre-cathode annealed solar cells were stressed for 150 h under accelerated degradation conditions.

In Figure 5.18 the solar cell parameters during continuous illumination of 1000 W/m² of the reference cell without additives and the best stabilized cell containing Additive #5 are compared.

In the investigated samples only two time regimes were observed, thus a bi-exponential decay in the form:

22

11 expexp)(

tta

ttatPCE Equation 23

was used to quantitatively describe the time dependence of the photovoltaic parameters during degradation [37].

0 20 40 60 80 100 120 140 1600.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6 Additive #5 Additive #5 fit Reference Reference fit

PCE

(%)

Time (h)

0 20 40 60 80 100 120 140 1600

1

2

3

4

5

6

7J sc

(m

A/c

m2 )

Time (h)

Additive #5 Additive #5 fit Reference Reference fit

0 20 40 60 80 100 120 140 1600.0

0.1

0.2

0.3

0.4

Voc (

V)

Time (h)

Additive #5 Additive #5 fit Reference Reference fit

0 20 40 60 80 100 120 140 1600

10

20

30

40

50

60

FF (

%)

Time (h)

Additive #5 Additive #5 fit Reference Reference fit

Figure 5.18: Long-term PCE, Jsc, Voc and FF of solar cells with no additives and containing Additive #5. The solar cells were measured continuously during 1000 W/m² illumination of a metal halide lamp. The time dependence of the parameters was fitted using a bi-exponential decay curve

62

Two values, tburn-in and t80, used to determine the burn-in and the lifetime of the solar cell [372], are extracted from the fitted curves for the purposes of evaluating the stability of the solar cells, as shown in Figure 5.19. The burn-in of the PCE is characterized by a fast decay, described by the time constant t1, while t2 corresponds to a slow degradation process of the solar cell following the burn-in.

However, there is no exact rule on how to derive these parameters, and the burn-in period is commonly approximated by a point in time where the dynamics of the decay changes. In this study, the burn-in time is analytically defined as the intersection between the first derivations of the two exponential contributions of the fitted curve:

21

12

21

12lntt

tttAtAt inburn

Equation 24

Figure 5.19: Long-term stability curve with characteristic parameters

In order to interpret the effect of the additive on the stabilization of the OPV device, I defined an additional parameter, APG, which can be calculated to reflect commercially relevant information concerning the power which can be generated by the device. Accumulated power generation (APG) is calculated as the integral of extrapolation of PCE measured under continuous illumination (1 kW/m²) for a given period of time (two years), as depicted in Figure 5.20. By incorporating the initial and final power conversion efficiency in one parameter, it takes into consideration that initially slightly lower power conversion efficiency devices with a slower decay of the PCE, can compensate their initial low performance with more power accumulation in their lifetime.

63

Figure 5.20: Graphical representation of Accumulated Power Generation APG

In Table 5.2 the determined long-term performance of the solar cells with and without additives is given in form of bi-exponential fit parameters (a1, a2, t1 and t2), percentage ratio of the final and initial efficiency PCEend /PCE0, burn-in (tburn-

in) and lifetime (tlife) values and accumulated power generation (APG).

# Conc. [%]

PCE0 [%]

a1 a2 t1 [h]

t2 [h]

tburn-

in[h] tlife [h]

PCEburn-

in [%] PCEend [%]

PCEend/ PCE0 [%]

APG2yrs [kWh/m2]

1 20 1.26 0.36 0.57 14.0 331 40 61 0.53 0.34 27 194 2 20 1.30 0.38 0.63 15.1 321 40 57 0.58 0.37 29 208 3 10 0.87 0.34 0.24 9.4 186 33 30 0.21 0.10 11 48 4 20 1.55 0.62 0.55 10.7 170 33 28 0.48 0.20 13 100 5 20 1.57 0.40 1.05 22.3 742 58 142 1.00 0.84 53 788 6 20 1.23 0.53 0.47 21.5 289 63 49 0.41 0.26 21 147 7 0.03 1.25 0.38 0.46 14.4 354 45 65 0.42 0.29 23 168 R 1.48 0.59 0.58 12.2 434 45 85 0.54 0.39 27 259 Table 5.2: Solar cell long-term performance for devices with and without hindered phenols. Note that optimization was done with respect to the maximum obtainable additive concentration yielding a PCE which is not smaller than 2/3 of the reference’s PCE (see Table B 3 in Appendix B). Also note that the cells were pre-cathode annealed. Shown are the parameters of the bi-exponential fit, a1, a2, t1 and t2, as well as the commonly used tburn-in and tlife. The power conversion efficiency after the burn-in (PCEburn-in) and the power conversion efficiency after the end of the test period of 168 h (PCEend) are shown for completeness. The extrapolated accumulated power generation after two years of continuous illumination with 1 kW/m² (APG2yrs) is used as assessment for the stabilization effect. Highlighted is the row with the devices having APG value higher than the reference, stabilized with Additive #5. R denotes the reference device

It becomes obvious that only one of the seven tested hindered phenols is suitable for application as stabilizer for organic solar cell devices. The use of octadecyl 3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionate, also known under the product name Arenox A76, lead to an increase of the accumulated power generation by a factor of 3 compared to the reference device without additive, while all other tested compounds did not improve the stability of the active layer.

The increase of the APG obtained with Additive #5 is caused by a simultaneous increase of the burn-in time as well as an overall slower decay of the PCE

64

(compare t1 and t2 values with those of the reference device and all other additives). Note also that the ratio a1/a2 is significantly smaller compared to all other devices (with Additive #5 ~0.4 compared to ~1 for all other cases). This is equivalent to a significantly decreased loss of the PCE during the burn-in period. Furthermore, this result indicates that the dominant degradation mechanism of the active layer might have changed compared to the reference device.

Figure 5.21: Essential bimolecular processes involved in hindered phenols stabilization mechanism (Adapted from Pospisil et al. [124])

The activity mechanism of hindered phenols is based on hydrogen transfer from the phenolic hydroxy group to alkylperoxyls derived from polymers oxidizing by a chain breaking mechanism [124, 373]. The primary transformation products are phenoxyl radicals A•, which play a key role in the chemistry of phenolic antioxidants, and are the precursors of all further transformation products. In the

65

course of aging, the activity of the additive gets depleted, in interactions with alkylhydroperoxides, photolysis, sensitized photooxidation and impurities (such as transition metal ions) [373-376]. Essential bimolecular processes, as depicted in in Figure 5.21, include disproportionation to quinone methide (QM) and starting phenol (InH) (route a), that is, regeneration of antioxidant and formation of R• trap (route d and e); C-O and C-C couplings leading to aryloxycyclohexadienones (ArO-CHD) and phenolic dimers (HIn•-•InH); and recombination with RO2• to alkylperoxycyclohexadienones (ROO-CHD) (route c) [124].

The transformation products of the phenols can also possess antioxidant or even radical scavenging abilities, as it is in the case of propionate type phenolics (various derivatives of 3-(3,5-di-tert-butyl-4-hydroxyphenyl) propionic acid), such as Additive #4 and #5, which makes them superior to the other types of hindered phenols. The effect is based on an intermolecular rearrangement and aromatization of the conjugated QM system [367, 373].

As shown in Figure 5.22, the initial additive is prone to aromatization to a phenolic cinnamate, a strong antioxidant. In reactions with alkylperoxy radicals, dimeric quinone methide is formed, which can rearrange to a phenolic cinnamate dimer, which is capable of scavenging alkylperoxyls. In the ultimate phase, a conjugated quinone methinoide dimer is formed. Analogous transformations also take place in polynuclear propionate phenolics, such as Additive #4 [373]. However, the other tested compound from this group, Additive #4, exhibited a much lower stabilization effect, which is attributed to its bulkiness which limits its diffusion through the active layer.

Figure 5.22: Transformation products of propionate type hindered phenols (Adapted from Pospisil et al. [373])

Similar is observed in the subclass of partially hindered benzyl phenols, which have substituents at least in 4- (or 2- or 6-) position having H atoms on a C atom vicinal to the phenyl group [106, 367, 373, 377, 378], such as Additive #1, #2 and #3. Since the hydroxyl group of the phenol is the active hydrogen donor, one would expect that the stabilizing effect of an additive is in direct correlation with the number of available phenol groups within the additive. However, stronger

66

efficiency loss after 1 week of degradation of Additive #3, as compared to Additive #1 and #2, as shown in Table 5.2, suggests that the stabilization effect can be limited by the bulkiness of the stabilizer.

From the considerations in previous sections and above mentioned structural characteristics, it can be suggested that the increase in long term performance using Additive #5 is occurring via a reduction of radical states and reduced exciton recombination. However, the origin of the PCE decay, the degradation mechanism, cannot be resolved from the PCE measurements alone, and other measurements are needed. To mention here are imaging techniques like photo-/electro- luminescence imaging, dark-lock in thermography [30, 379, 380], as well as Fourier transform infrared or UV-Vis spectroscopy [29, 33-35, 38, 40-42] and photoelectron spectroscopy [36, 381], or time-of-flight secondary ion mass spectrometry analysis in conjunction with isotopic labeling [32, 39, 382, 383], in order to resolve if the degradation of the electrodes, photo-oxidation of the photoactive layer or the formation of defect states are the main degradation factors. To confirm this, the reference and degraded (1.5 AM illumination in air) samples, as measured with FTIR (Fourier Transform Infrared Spectroscopy), are shown in Figure 5.23.

As identified by Manceau et al. (see Figure 3.10), the peak which arises around 3460 cm-1 in the spectra of degraded samples, is attributed to hydrogen-bonded OH-stretching of alcohols and hydroperoxides [180, 384]. These products correspond to the products formed in the chain propagation step, in reaction of peroxy radicals with the polymer, which yields hydroperoxides and alkyl radicals, as depicted in Figure 5.10. As seen in Figure 5.23 (left), the intensity of these peaks in the sample containing Additive #5 is lower, which is consistent with the above mentioned characteristic of partially hindered phenols to form quinone methides, which by reacting with alkyl and peroxy radicals in the chain propagation step can reduce the amount of hydroperoxides formed. As a consequence, less of the reactive alkoxy and hydroxyl radicals will be formed in chain branching step, and those which will be formed will partially also react with the additive-borne quinone methides. This is confirmed by spectra in the region 1900-1550 cm-1, where it can be observed that the samples containing Additive #5 have an overall lower intensity in the carbonyl region, which is where the contributions of oxidative products arising from the decomposition of hydroperoxides [180], including esters, anhydrides, aliphatic and aromatic ketones and carboxylic acids, are located. In concordance, in the CH-stretching region, the intensity ratio of reference and degraded samples, as observed in the interval from 3100 to 2700 cm-1 [172, 180, 385], indicates a reduction of side chains degradation in the presence of Additive #5.

67

4000 3800 3600 3400 3200 3000 28000.000

0.005

0.010

0.015

0.020

0.025Abs

orba

nce

(a.u

.)

Wavenumber (cm-1)

Additive #5 initial Additive #5 degraded Reference initial Reference degraded

3100 3000 2900 2800 2700 2600 25000.000

0.005

0.010

0.015

0.020

0.025

0.030

Abs

orba

nce

(a.u

.)

Wavenumber (cm-1)

Additive #5 initial Additive #5 degraded Reference initial Reference degraded

2000 1900 1800 1700 1600 1500 1400 13000.000

0.005

0.010

0.015

0.020

0.025

Abs

orba

nce

(a.u

.)

Wavenumber (cm-1)

Additive #5 initial Additive #5 degraded Reference initial Reference degraded

Figure 5.23: FTIR spectra of the samples containing Additive #5 and reference samples with no additive: (left) hydroxyl region, (right) zoomed up CH-stretching region, and (bottom) carbonyl regions

68

5.2.2 UV Absorbers The second tested class of polymer stabilizers are the UV Absorbers. In contrast to the chemical stabilization of hindered phenols, the protection mechanism of UV absorbers is of entirely physical nature and based on the strong absorbance of the destructive UV radiation (290-400 nm) and its rapid dissipation in form of heat [386]. Dissolved in the polymer matrix, these additives compete with the light-induced reactions of the polymer [106], as shown in Figure 5.24. By absorption of light, UV absorbers get transformed into an excited state, which is by rapid intramolecular processes, deactivated and returned to ground state. Consequently, the energy imposed on the polymer matrix cannot initiate photooxidation.

Figure 5.24: The principle of stabilization using UV absorbers. RH denotes the non-degraded polymer

Figure 5.25 shows the most important classes of UV absorbers, which were tested in this study. They can be divided into two groups, the phenolic and the non-phenolic compounds. Phenolic compounds can be subdivided in those forming O-H-O bridges, such as Additive #1 and #5, and those forming O-H-N bridges, such as Additive #2 and #6, while the non-phenolic compounds include Additive #3 and #4.

# Chemical Nomenclature Type 1 2-hydroxy-4-(octyloxy) benzophenone o-hydroxybenzophenone 2 octrizole 2-(2-hydroxyphenyl)benzotriazole 3 octyl-4-methoxycinnamate cinnamate 4 etocrilene cyanoacrylate 5 phenyl salicylate salicylate

6 2-(4,6-diphenyl-1,3,5-triazin-2-yl)-5-[(hexyl)oxy]-phenol 2-(2-hydroxyphenyl)-1,3,5-triazine

Table 5.3: List of investigated UV absorbers. For convenience, throughout the manuscript the compounds are represented by the number shown in the first column

69

Figure 5.25: Chemical structures of the investigated UV absorbers

The common structural element of the phenolic compounds is the strong intramolecular hydrogen bond (IMHB). This bond is formed between the hydrogen atom of the phenol’s hydroxy group and a highly electronegative atom present in the same molecule, such as oxygen or nitrogen, resulting in the cyclization of the molecule in the ground state S0 [387, 388].

Figure 5.26: Scheme of Excited State Intramolecular Proton Transfer (ESIPT) in phenolic UV absorbers, with example of 2-hydroxybenzophenone transformation

70

Upon absorption of light, singlet state S1 is formed, as shown in Figure 5.26, and although the molecule remains in the phenol form, the acidity of the phenolic group is strongly enhanced, compared to S0 [388-391]. The basic part of the molecule is either the heterocyclic nitrogen or oxygen atom [389, 391]. Since the acidic and basic sites are in close proximity, upon electric excitation, the excited state intramolecular proton transfer takes (ESIPT) place [388, 392].

The electronic distribution in the first excited singlet state S1 favors a fast proton transfer to the heteroatom, and accounts for proton-transferred excited singlet state of the tautomer form S1’ with strong IMHB [393]. The excited tautomer S1’ is believed to dissipate the excitation energy by a rapid internal conversion (IC), a radiationless non-degradative process, accounting for the ground state of the tautomeric form S0’.

The last step of ESIPT, involves the proton jumping back (a reverse proton transfer) thus regenerating the original ground state S0 of the phenolic form. The transformation S0S1S1’S0’S0 occurs on an ultrafast picosecond timescale for highly photostabile UV absorbers [394]. The fast physical process S1’S0

accounts for the long service time of UV absorbers [388].

This mechanism can be repeated as long as the IMHB and planarity of the stabilizer remain intact [388, 395]. ESIPT is therefore the key in rapid deactivation of excited UV Absorber molecules [388, 390, 393, 394, 396-399].

Ternary Blending Although UV absorbers are known to successfully stabilize insulating plastic materials such as polyamides, PVC or polycarbonates, in order to have them implemented into active layers of solar cell devices, additional conditions have to be met. Since organic solar cells are optoelectronic devices, interference with the photoinduced charge transfer will greatly affect their performance. This will occur if the energetic levels (HOMO/LUMO) of implemented stabilizer is within the effective band gap Egap,eff of the photoactive layer [331-339], as such trap levels act as centers for non-geminate recombination of free charge carriers. All tested UV-absorbing compounds were characterized via cyclic voltammetry, and no energy levels were found within the effective band gap of the polymer/fullerene system (for details see Table B 4 in Appendix B). Thus formation of unfavorable energetic traps, upon ternary blending of one of the additives to the binary photoactive blend, is thus unlikely.

Besides electronic traps, the presence of third component in the active layer can cause inauspicious morphology, which can in turn impair the performance [354-

357]. Films of ternary blended active layers exhibited a comparable photoluminescence signal to that of the reference film, as shown in Figure B 1 in Appendix B, which indicates that all films have a comparable morphology. In the

71

case of Additive #3, the signal is somewhat reduced compared to the reference. As the band gap of this compound is low, (around 3.4 eV, see Table B 4 in Appendix B), the decrease in the PL cannot be caused by the underestimation of the signal within the normalization. Fill factor of these devices, comparable to that of the reference (both around 55 %, see Table B 5), also excludes the possibility of compatibilizing activity of Additive #3. Thus the possible explanation remains that this compound is capable of acting as an additional excitation quencher. Indeed, reports in the literature have attributed the photostabilization of plastics using cinnamic acid esters to be based on a combined UV radiation filter effect and quenching of the excited states [400, 401].

5.2.2.1 Bleaching Blend films containing one of the seven UV absorbers and a reference film without additives, were prepared on CaF substrates, and investigated via UV-Vis and FTIR spectroscopy, both freshly prepared and upon illumination in ambient. Figure 5.27 shows the normalized absorption spectra of all investigated films prior to degradation in air, as well as the relative absorption change upon film aging. Nearly identical ratio of the A0-0 and A0-1 absorption peaks, and a good overlap of the absorption spectra in the wavelength range 300-800 nm after normalization to the maximum polymer absorption further confirm that the presence of additives did not significantly interfere with the morphology of the active layers. As the investigated additives have a large optical band gaps (see Table B 4 in Appendix B), the blend absorption of the films differs only in the high energy region.

200 300 400 500 600 700 8000.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

#1 #2 #3 #4 #5 #6

Nor

mal

ized

Abs

orba

nce

(a.u

.)

Wavelength (nm)

Reference

200 300 400 500 600 700 800

-0.6

-0.4

-0.2

0.0

0.2

#1 #2 #3 #4 #5 #6

( Abs

orba

nce(

120h

) -

Abs

orba

nce(

0h)

) /

Abs

orba

nce(

0h,

500n

m)

(a.u

.)

Wavelength (nm)

Reference

Figure 5.27: (left) Normalized UV-Vis absorption spectra of the non-degraded films with additives and one reference with no additive on CaF substrates; (right) relative change in absorption of the films after 120 h of continuous 1000 W/m² illumination in ambient. Note that all spectra were normalized to the absorption peak at 500nm of the corresponding non-degraded film

72

Upon degradation (120 h of aging under 1000 W/m² continuous illumination in air), the films have photo-bleached significantly. Additionally an increase in scattering is observed, which can be attributed to the morphological changes and formation of PCBM clusters, as indicated by the apparent absorption increase above 700 nm, see Figure 5.27 (right).

Stronger bleaching of the active layer in the visible wavelength range, compared to the reference film without the additive, suggests increased thiophene ring destruction in the presence of some additives. Thiophene ring can get destroyed via two mechanisms, a free radical pathway and a pathway involving photosensitized species (e.g. singlet oxygen). The mechanism of UV absorbers involves stabilization against the radical initiated decay, but the decay can still occur via formation of photosensitized species [172, 183]. The side chain reactions do not contribute to the visual impression of degradation which is observable in simple UV-Vis measurements, however they are of significant electronic importance, as explained in the previous section.

5.2.2.2 Long-term stability Solar cells containing each of the seven antioxidants were tested with respect to the stability enhancement of the solar cells, as compared to reference solar cells without additives. Encapsulated pre-cathode annealed solar cells were stressed for 150 h under accelerated degradation conditions.

In all of the investigated devices, two regimes were observed, thus the parameters were fitted using a bi-exponential decay. In Table 5.4 the determined long-term performance of the solar cells with and without additives is given in form of the characteristic bi-exponential fit parameters (t1, t2, a1, a2), burn-in (tburn-in) and lifetime (tlife) values, APG and percentage ratio of final and initial efficiency.

# Conc. [%]

PCE0 [%]

a1 a2 t1 [h]

t2 [h]

tburn-

in[h] tlife [h]

PCEburn-

in [%] PCEend [%]

PCEend/ PCE0 [%]

APG2yrs [kWh/m2]

1 20 1.21 0.34 0.59 15.5 392 43 74 0.55 0.38 32 237 2 10 1.83 0.64 0.71 10.0 318 35 62 0.66 0.42 23 232 3 20 1.36 0.47 0.53 21.0 566 69 109 0.49 0.39 29 310 4 20 1.54 0.54 0.61 22.0 512 70 96 0.56 0.44 29 324 5 10 1.42 0.54 0.64 11.8 280 37 69 0.58 0.35 25 186 6 10 1.52 0.46 1.05 11.1 895 40 189 1.02 0.87 57 945 R 1.48 0.59 0.58 12.2 434 45 85 0.54 0.39 27 259 Table 5.4: Solar cell long-term performance for devices with and without UV absorbers. Note that optimization was done with respect to the maximum obtainable additive concentration yielding a PCE which is not smaller than 2/3 of the reference PCE (see Table B 6 in Appendix B). Highlighted is the row with the devices having APG values higher than the reference, stabilized with Additive #6. R denotes the reference device

73

In Figure 5.28 the solar cell parameters during continuous illumination of 1000 W/m² of the reference cell without additives and the best stabilized cell containing Additive #6 are compared.

0 20 40 60 80 100 120 140 1600.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6 Reference Reference fit Additive #6 Additive #6 fit

PCE

(%)

Time (h)

0 20 40 60 80 100 120 140 1600

1

2

3

4

5

6

7

J sc (

mA/c

m2 )

Time (h)

Reference Reference fit Additive #6 Additive #6 fit

0 20 40 60 80 100 120 140 1600.0

0.1

0.2

0.3

0.4

0.5

Voc (

V)

Time (h)

Reference Reference fit Additive #6 Additive #6 fit

0 20 40 60 80 100 120 140 1600

10

20

30

40

50

60FF

(%

)

Time (h)

Reference Reference fit Additive #6 Additive #6 fit

Figure 5.28: Long-term PCE, Jsc, Voc and FF of solar cells with no additives and containing Additive #6. The solar cells were measured continuously during 1000 W/m² illumination of a metal halide lamp. The time dependence of the parameters was fitted using a bi-exponential decay curve

The causes of differences in stabilizing effect between the tested compounds can be sought in morphological issues concerning the homogenous diffusion of additives into the active layer, and most importantly in the mechanism specifics of each of the tested compounds including their possible loss pathways.

Salicylates, such as Additive #5, act by ESIPT mechanism involved in original structure and via 2,2’-dihydroxybenzophenone resulting from its Photo Fries rearrangement (homolytic cleavage which gives a pair of radicals which subsequently recombine within the cage) [387]. If the two radicals drift apart before recombining, the phenoxy radical may abstract a hydrogen atom from the polymer to form a phenol, or if oxygen is present the radicals may react with it,

74

reducing the rearrangement product and the effectiveness of the additive [387, 402]. This is manifested in the inferior effectiveness of Additive #5 with respect to benzophenone type Additive #1.

Although benzophenone derivatives (Additive #1) are lightfast and have an effective ESIPT [387, 400], the effectiveness of Additive #1 is not optimal. This can be explained by losses during stabilization due to its activity as free-radical scavengers [403], which due to their inability to reform themselves, disrupts ESIPT thus depleting the active UV absorber.

Similar losses involving abstraction of hydrogen from phenolic hydroxyl in presence of free radicals occur in benzotriazoles (Additive #2) [326, 373, 404-407]. However, in case of Additive #2, an additional structural characteristic might account for its lower effectiveness. Absence of bulky ortho-substituents to phenolic hydroxyl enables “opening” of the intramolecularly H-bonded six-membered ring, which allows the rotation of the hydroxy group and enables intersystem crossing to excited long-living triplet form [326, 373, 408]. This leads to depletion of the absorbers and sensitization of the polymer and photooxidation [394], thus lowering the stabilization effectiveness.

IMHB in triazines (such as Additive #6) is stronger than that in benzotriazoles (Additive #2) [393, 398, 409, 410]. This results from a longer O-H distance and shortened N-H distance and definitely a much more linear H-bond [398]. Rotation around the C–C bond (connecting triazine and phenol group) allows the switching of H-tunneling to either of the two closest N atoms. This gives them an entirely ‘closed-ring’ structure because of strong IMHB [411], enabling effective ESIPT after excitation, thus explaining their superior effectiveness.

The two other non-phenolic compounds, of cinnamate and cyanoacrylate class (Additive #3 and #4), manifested a stronger stabilizing effectiveness, as compared to the most of phenolic compounds which suffer from free-radical facilitated depletion.

The increase of the APG, obtained with Additive #6, by factor of over 3.6, is governed by an overall slower decay of the PCE, as reflected in t2. Unlike stabilization via hindered phenols, no decrease of the burn-in period was observed. Note that the ratio a1/a2 is significantly smaller compared to all other devices (Additive #6 ~0.4 compared to ~1 for all others), equivalent to a significantly decreased loss of the PCE during the burn-in period. This indicates stabilization of the active layer due to the reduced formation of radicals, facilitated by the UV filtering of the additive. However, it does not eradicate the formation of some radicals, and their reactions with the active layer are not inhibited by this additive, resulting in a burn-in period comparable to that of the reference devices. This is suggested by the FTIR spectra, shown in Figure 5.29.

75

4000 3800 3600 3400 3200 3000 28000.000

0.005

0.010

0.015

0.020

0.025Abs

orba

nce

(a.u

.)

Wavenumber (cm-1)

Additive #6 initial Additive #6 degraded Reference initial Reference degraded

3100 3000 2900 2800 2700 2600 25000.000

0.005

0.010

0.015

0.020

0.025

0.030

Abs

orba

nce

(a.u

.)

Wavenumber (cm-1)

Additive #6 initial Additive #6 degraded Reference initial Reference degraded

2000 1900 1800 1700 1600 1500 1400 13000.000

0.005

0.010

0.015

0.020

0.025

Abs

orba

nce

(a.u

.)

Wavenumber (cm-1)

Additive #6 initial Additive #6 degraded Reference initial Reference degraded

Figure 5.29: FTIR spectra of the samples containing Additive #6 and reference samples with no additive: (left) hydroxyl region, (right) zoomed up CH-stretching region and (bottom) carbonyl regions

UV is only a small portion of UV-Vis radiation that reaches the sample. By deactivating that component via UV absorbers, only a small amount of incoming photons will be affected, which explains the very modest decrease in the formation of hydroperoxides. However, the homolytic splitting of hydroperoxides requires the high energy light, thus UV filtering will reduce the formation of resulting OH• and PO• radicals. Such formed OH• radicals undergo cage reactions which, in presence of light and oxygen, result in formation of carboxylic acids. By UV filtering, this process gets attenuated, which explains the reduced intensity of carboxylic acids signal around 3240 cm-1 [180]. Due to that, the formation of all further oxidation products (see Figure 3.10) evolving from carboxylic acids is reduced, thus slowing down the decay of the devices, as observed in carbonyl region of the spectrum (1900-1550 cm-1). Accordingly, the intensity ratio of reference and degraded samples in the CH-stretching region (3100-2700 cm-1) confirms that the stabilizing mechanism of Additive #6 reduces the side chains degradation.

76

5.2.3 Hydroperoxide Decomposers Hydroperoxides formed in the propagation step are the source for the further oxidative degradation of polymer. Homolytic split of hydroperoxides results in formation of two extremely reactive alkoxy and peroxy radicals, RO• and •OH. In the presence of some organosulfur, organophosphorus and “quencher” compounds, the rate of the homolytic split reaction can be reduced. These compounds effectively decompose the hydroperoxides into non-radical products, as depicted in Figure 5.30. Compounds tested in this study are listed in Table 5.5 and their structures are shown in Figure 5.31.

Figure 5.30: The principle of stabilization using hydroperoxide decomposers, such as organophosphorus compounds. P represents phosphorus and Ar the aryl chain

# Chemical Nomenclature Type 1 3,3'-thiodipropionic acid dimyristyl ester organosulfur 2 triphenyl phosphite aromatic organophosphorus 3 phosphorous acid triisodecyl ester aliphatic organophosphorus

4 2-2'-thiobis(4-tert-octylphenolate)-n butylamine nickel (II) Ni chelate (quencher)

5 AdvapakTM NEO-1120 blocked thiols Table 5.5: List of investigated hydroperoxide decomposers. For convenience, throughout the manuscript the compounds are represented by the number shown in the first column

Figure 5.31: Chemical structure of the investigated hydroperoxide decomposers

77

The long-term performance parameters of encapsulated pre-cathode annealed solar cells containing each of the five antioxidants were stressed for 150 h under accelerated degradation conditions are reported in Table 5.6.

# Conc. [%]

PCE0 [%]

a1 a2 t1 [h]

t2 [h]

tburn-

in[h] tlife [h]

PCEburn-

in [%] PCEend [%]

PCEend/ PCE0 [%]

APG2yrs [kWh/m2]

1 3 2.43 0.67 1.28 10.8 301 30 57 1.20 0.39 30 393 2 20 1.10 0.31 0.53 15.3 342 41 63 0.49 0.32 29 186 3 10 1.13 0.48 0.52 12.3 340 41 63 0.48 0.32 28 183 4 3 0.78 0.20 0.44 16.0 372 39 69 0.41 0.28 36 167 5 20 1.06 0.31 0.44 21.2 808 72 149 0.41 0.36 34 362 R 1.48 0.59 0.58 12.2 434 45 85 0.54 0.39 27 259 Table 5.6: Solar cell long-term performance for devices with and without hydroperoxide decomposers. Encapsulated pre-cathode annealed solar cells were stressed for 150 h under accelerated degradation conditions. Note that optimization was done with respect to the maximum obtainable additive concentration yielding a PCE which is not smaller than 2/3 of the reference PCE (see Table B 7 in Appendix B). Highlighted are the rows with the devices having APG values higher than the reference, stabilized with Additive #1 and #5. R denotes the reference device

The lowest APG was obtained using Additive #4, which is classified as a quencher. Although the name originates from the times when metal complexes, such as nickel chelates, were thought to stabilize polymers based on energy transfer from an excited chromophore to a quencher before the former can undergo photochemical reactions [126, 412-416], the main stabilizing mechanism of these compounds is hydroperoxide decomposition and radical scavenging [126, 417-

419]. The high ratio between the final and initial efficiency indicates a stabilization effect, however due to the initially low efficiency, the APG remains lower than that of the reference devices. In agreement with the photoluminescence measurements (see Figure B 2 in Appendix B), the decreased efficiency is a result of the quenching properties of this compound, which obstructs the photoinduced charge transfer, essential to the functioning of the organic solar cells.

The mechanism of phosphite based compounds, such as Additive #2 and #3, consists of reducing hydroperoxide to corresponding alcohols, in which they get transformed into phosphates [146, 420, 421]:

33 )(=+')(+' ORPOOHPORPOOHP → Equation 25

This mechanism is stoichiometric, meaning that 1 molecule of additive can only decompose one hydroperoxide molecule. This explains why their stabilizing effectiveness is not as high compared as of the organosulfides [422].

The mechanism of organosulfide compounds includes transformation of two molecules of hydroperoxides into alcohols, according to the following reactions:

78

++ POHRSRPOOH →-- Equation 26

+POOH +POH→ Equation 27

After formation of sulfoxide in Equation (26), other reactions may take place, involving sulfenic acid [422, 423]:

'=+ → 2 RCHCHOHSR --- Equation 28

This is followed by reactions yielding thiosulfinates and thiosulfoxylic acid:

→-- OHSR2 OH2+ Equation 29

RSSOHCHCHR +=' 2-→ Equation 30

Sulfur containing acids, and sulfur dioxide and trioxide which are formed as end products of oxidations are especially effective hydroperoxide decomposers [423], adding to the stabilizing effectiveness of this type of compounds. The stabilizing mechanism of both of the compounds that showed an increase in APG compared to the reference devices, Additive #1 and #5, involve this mechanism.

Additive #5 is a polyvinyl chloride (PVC) stabilizer, based on blocked thiol mechanism [424]. Stabilization of PVC is somewhat specific compared to other plastic materials, as it is strongly influenced by the presence of chlorine atom in its structure. There are two processes taking part simultaneously during its photooxidation: one is dehydrochlorination leading to formation of polyene sequences and crosslinking, and the other one is the oxidative radical chain process and chain scission [425-429]. In the course of aging, HCl gets eliminated from the polymer, forming a redox complex with hydroperoxide which accentuates the degradation process [111]. However, due to its dominant organosulfide character, the stabilizing activity, as in the case of Additive #1, involves decomposition of hydroperoxides into alcohols thus preventing its homolytic split into reactive radicals [425, 430], according to the general mechanism depicted in Equations (26)–(30).

79

4000 3800 3600 3400 3200 3000 28000.000

0.005

0.010

0.015

0.020

0.025Abs

orba

nce

(a.u

.)

Wavenumber (cm-1)

Additive #1 initial Additive #1 degraded Additive #5 initial Additive #5 degraded Reference initial Reference degraded

3100 3000 2900 2800 2700 2600 25000.000

0.005

0.010

0.015

0.020

0.025

0.030

Abs

orba

nce

(a.u

.)

Wavenumber (cm-1)

Additive #1 initial Additive #1 degraded Additive #5 initial Additive #5 degraded Reference initial Reference degraded

2000 1900 1800 1700 1600 1500 1400 13000.000

0.005

0.010

0.015

0.020

0.025

0.030 Additive #1 initial Additive #1 degraded Additive #5 initial Additive #5 degraded

Abs

orba

nce

(a.u

.)

Wavenumber (cm-1)

Reference initial degraded

Figure 5.32: FTIR spectra of the samples containing Additive #1, Additive #5 and the reference sample with no additive: (left) hydroxyl region, (right) zoomed up CH-stretching region and (bottom) carbonyl region

Long-term stability of the photovoltaic parameters obtained by devices stabilized by Additive #1 is shown in Figure B 4 in Appendix B. Although the APG is increased, the decay dynamics of neither of the parameters has been slowed down. This is further confirmed by the FTIR measurement in Figure 5.32, which shows no attenuation of the signal in the peroxide or the region of P3HT oxidative products, thus confirming the absence of stabilization effect. The increase in APG in this case, occurs not due to the stabilization of the system but mere due to an increase in initial efficiency. The increase in Voc and FF, but a constant Jsc, of devices prior to degradation, as shown in Table B 8 in Appendix B, point to differences in the interfacial morphology as a cause for the efficiency increase. A similar effect is observed when devices are annealed after the cathode deposition, preventing the formation of a thin P3HT surface layer at the interface with the cathode, which results in a significantly higher Voc [45, 431-437]. Possibly, the Voc increase might even be a result of formation of a hole blocking layer consisting of the additive at the cathode interface, similar to the performance improvement upon the introduction of a TiOx as a hole blocking

80

layer [438]. Further PES measurements should be conducted in order to conclude on the origin of the observed Voc increase into more detail.

Devices containing Additive #5, on the other hand, manifest an increase in APG in spite of the initially reduced efficiency. The decay dynamics of the parameters is significantly slowed down, especially the Jsc and efficiency, as shown in Figure B 4 in Appendix B, leading to a higher long-term efficiency which compensates the initially weaker performance, and eventually results in an overall higher accumulated power generation. FTIR measurements of degraded film containing Additive #5, as shown in Figure 5.32, reveal a strongly decreased signal belonging to both hydroperoxide (3460 cm-1) and resulting oxidative products (carbonyl region in the interval 1900-1550 cm-1), as compared to the degraded films containing no additive. This confirms the proposed hydroperoxide decomposer mechanism, where POOH, which is otherwise homolytically split forms two extremely reactive radicals RO• and •OH which take part in further radical chain oxidation of the active layer (see Figure 3.10), is instead decomposed into non-reactive products. The reduction in CH-stretching signal of samples containing additives indicates that the non-radicals products formed from polymer side chains are mostly simple alcohols which, at working temperatures, evaporate out of the films.

Although the stabilizing efficiency of hydroperoxide decomposing additives is not as high as that of hindered phenols and UV absorbers, they do show a slight stability improvement without compromising the performance of the devices. This opens up a possibility for heterosynergistic blends with additives of other stabilizing functionalities. Specifically, blends combining the stabilizing effect of hydrogen donors, such as phenols or amines, and the hydroperoxide decomposition properties of organophosphorus [106, 439-442] and organosulfur [106,

327, 443] compounds have been successfully employed in stabilization of polypropylene, polystyrene and polyolefins.

81

Chapter 6

Summary and Outlook

In the last two decades organic photovoltaics have drawn a lot of attention in the academic circles. The efficiency of the devices is already as high as 12 %, and as the production scales up the price per generated power is predicted to decrease significantly below their inorganic counterparts. Nevertheless, in order to attract commercial interest, the long-term stability of devices still needs to be improved. In my thesis I investigated the stabilization of organic solar cells, by blending an additional third component with stabilizing properties into the active layer. The highlights of this study are briefly summarized below.

I successfully demonstrated on P3HT:[60]PCBM model devices, that ternary blending photoactive layers of organic solar cells with antioxidants and UV absorbers can be applied to substantially prolong the lifetime of organic solar cells. Drastic improvements in lifetime of P3HT:[60]PCBM devices were achieved, increasing the accumulated power generation by more than triple.

A highly successful stabilization was achieved with propionate hindered phenols, which mechanism includes hydrogen donation and free radicals scavenging. Upon addition of octadecyl 3-(3,5-di-tert-butyl-4-hydroxy-phenyl) propionate, the long time performance, as tracked in situ under ISOS-3 conditions, could be largely enhanced via a reduction of the burn-in loss and prolonged burn-in time. Compared to a reference cell prepared in a standard manner and without stabilizer, the efficiency of the additive containing device after its burn-in period was still twice as high. This could be attributed to a reduction of the number of formed radicals upon degradation, which is reflected in a decreased losses of short circuit current density and open circuit voltage during this period. Considerations on exciton recombination statistics, based on the Poisson probability of generated excitons finding radicals formed during degradation, further support this statement and reveal the importance of reducing the radical formation within the device.

A similar reduction in formation of radicals was also achieved using triazine UV absorbers, which is attributed to their ability to absorb the harmful UV portion of radiation and deactivate it non-radiatively. Due to their physical nature of device stabilization they are ideal synergists with hindered phenols which could lead to further improvements in the stabilization of OPV devices.

82

While the stabilizing effect of antioxidants and UV absorbers on classical insulating plastic polymers, such as polypropylene, polyethylene or polyvinylchloride, is well known, it is this study that demonstrates for the first time that they can be successfully used in organic solar cells. However, the material selection cannot be directly translated from the insulating systems to conductive OPV polymers, due to the comparably complex interplay between morphology of the photo active blend and charge separation and extraction, as well as the crucial impact that additional energetic levels, caused by introducing a third compound and its own energetic properties, can have on the device physics. In this sense, I thoroughly investigated over 20 test compounds of different stabilizer classes, ternary blended in the model P3HT:[60]PCBM active layer system, using microscopic and spectroscopic methods. Special emphasis was given on detecting possible trap-forming energetic levels causing increased recombination, and unfavorable morphology with too fine or too coarse phase separation between the donor and acceptor, leading to a reduced exciton separation or charge extraction.

I proved that by carefully chosen additives, the device stability can be increased without making compromises with the device performance. Although the presented results are based on blends of P3HT:[60]PCBM, the presented approach can easily be extended onto other polymer:fullerene systems, provided good compatibility between the active layer components and the stabilizer compound, which is necessary in order to gain favorable morphology and energy levels alignment, essential for the proper functioning of the devices. As long as the stabilizing compound adheres to these two requirements, there are no impediments for application of stabilizers to other polymer systems.

This proof of principle study opens the doors for further stability studies on ternary blend polymer:fullerene:additive active layers, including in-situ FTIR and UV-Vis spectroscopy, which would give a deeper insight into the kinetics of the stabilization mechanism. It would be of great interest to conduct such experiments in a controlled environment, including adjustable UV-Vis radiation wavelength and intensity, temperature and oxygen and humidity fractions in the atmosphere, which could reveal more details on the degradation processes present in OPV polymer systems. To further examine the influence of chemical compatibility of the stabilizer with the polymer:fullerene system, the study could be expanded to include testing on a variety of different polymer structures.

The most important aspect of these findings lies in the potential of the proposed stabilizing route in the commercialization of flexible organic solar cells, as it offers a way to produce highly reliable devices, regardless of the innate photochemical stability of the polymer and fullerene, by a simple addition of small portions of a cheap third component.

83

Appendix A

Supporting degradation study

Systematic Study of Active Layer Degradation under Exposure to Multiple Environmental Stresses

Their organic nature makes organic solar cells especially sensitive to light, oxygen, heat and water. As all of the mentioned factors are present in normal working environment of the solar cells, big efforts have to still be undertaken to understand the degradation mechanisms and to find ways to reduce these effects [43, 46, 47].

In this study stabilities of two different polymers, the anthracene containing PPV-type copolymer poly(p-phenylene-ethynylene)-alt-poly(p-phenylene-vinylene) (referred to as AnE-PVcc) [444-448] and a representative of the 3rd generation materials poly[[9-(1-octylnonyl)-9H-carbazole-2,7-diyl]-2,5-thiophenediyl-2,1,3-benzothiadiazole-4,7-diyl-2,5-thiophenediyl] (in further text, PCDTBT) [86, 449] are investigated and compared with the reference poly (3-hexylthiophene) P3HT devices [450]. The structures of the polymers are shown in Figure A 1. The peak efficiencies of the optimized solar cells devices were 1.75 % for pre-cathode annealed P3HT:[60]PCBM (3:2), 5 % in case of PCDTBT:[70]PCBM (2:3) and 4 % in case of AnE-PVcc:[60]PCBM (1:1).

Since solar cells are multi-layered structures, the decay in their performance reflects not only the degradation of the photoactive layer, but of each of its constituent layers. As degradation of the cathode exerts a strong influence on the performance, which can overshadow the effects of photoactive layer degradation, the stability testing of the material systems were performed on layer stacks prior to the cathode deposition.

Stability comparison test consists of selective exposure of glass/ITO/PEDOT:PSS/active layer stacks to different combinations of stresses, starting from one single stress to which new stresses were iteratively added one by one. Testing conditions included ambient and inert atmosphere, at different temperatures, in the dark and exposed to different types of illumination. This allowed a comparison of how harshly different polymers get affected by

84

individual stresses. The influence of temperature on the blend morphology, and the influence of UV radiation in absence of water and oxygen, could be directly investigated by applying each of the two stress factors in inert atmosphere of a glove box. By addition of air to the combination of stresses, oxygen related degradation could be investigated. To elucidate the influence of humidity, samples were exposed to a temperature below the water boiling point in dark air, thus including the presence of water vapors. Finally, all of the stresses combined along with UV-Vis radiation of 1 Sun was used to test the devices under ambient conditions.

Figure A 1: Scheme of the probed layer stack, along with the chemical structures of the tested polymers: poly (3-hexylthiophene) or P3HT, poly(p-phenylene-ethynylene)-alt-poly(p-phenylene-vinylene) copolymer or AnE-PPVcc, and poly[[9-(1-octylnonyl)-9H-carbazole-2,7-diyl]-2,5-thiophenediyl-2,1,3-benzothiadiazole-4,7-diyl-2,5-thiophenediyl] or PCDTBT

In the following, solar cell parameters of devices exposed to different combinations of stresses will be discussed. Each data point represents the average value, and the standard deviations are included in bars. The plotted error bars include the variations in the performance of cells contained on one substrate, caused by the possible inhomogeneities of the films and of the solar simulator lamp used to measure the devices. It is to note that each data point was measured on a different device, so an additional device-to-device fabrication error should be taken into account.

85

The work presented in this section has been partially published in my publication “Multiple stress degradation analysis of the active layer in organic photovoltaics” [45].

a) Thermally Accelerated Morphological Degradation To compare the extent of morphological degradation of the three material systems, the samples were exposed to 150 °C in inert atmosphere (nitrogen glove box). As shown in Figure A 2, the performance of P3HT:[60]PCBM devices stays unaltered in the first 8 h, as expected from the spectroscopic and microscopic measurements previously reported [290, 293, 296]. However, the current of AnE-PV and PCDTBT devices slowly drops over time, causing a loss in efficiency. This is because, unlike P3HT devices, which benefits from exposures to temperatures at around 150 °C due to of the increase in coherence length due to the growth of the crystallites [451], some other polymer devices, such as PCDTBT, suffer from the reduction of the coherence length of the -stacking caused by annealing, which increases the disorder in their electronic structure [452].

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

Nor

mal

ized

Jsc

Time (h)

P3HT PCDTBT AnE-PV

Dark Glovebox 150°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Nor

mal

ized

Voc

Time (h)

P3HT PCDTBT AnE-PV

Dark Glovebox 150°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Nor

mal

ized

FF

Time (h)

P3HT PCDTBT AnE-PV

Dark Glovebox 150°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Nor

mal

ized

PCE

Time (h)

P3HT PCDTBT AnE-PV

Dark Glovebox 150°C

Figure A 2: Influence of elevated temperature (150 °C) in absence of air in the dark: Jsc, Voc, FF and PCE. Parameters are normalized with respect to initial values

86

b) Photolysis by UV Radiation To separately investigate just the effect of illumination, the devices were exposed to the most destructive type, from the UV region [38, 183, 347], at room temperature in inert atmosphere.

Performance of the devices decays exponentially, as seen in Figure A 3, due to the exponential decrease of the Jsc. Interestingly, Voc decays much slower, and in case of P3HT it even increases to a plateau value of 0.6 V. However, following the initial increase in Voc, a sharp decay follows, indicating the complete destruction of the polymer in case of P3HT [174, 214, 453]. At the same time, although to a reduced capacity, the PCDTBT and AnE-PV based devices continue functioning.

0 5 10 15 20 25 30

0.0

0.2

0.4

0.6

0.8

1.0

Nor

mal

ized

Jsc

Time (h)

P3HT PCDTBT AnE-PV

UV Irradiation Glovebox RT

0 5 10 15 20 25 30

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Nor

mal

ized

Voc

Time (h)

P3HT PCDTBT AnE-PV

UV Irradiation Glovebox RT

0 5 10 15 20 25 30

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Nor

mal

ized

FF

Time (h)

P3HT PCDTBT AnE-PV

UV Irradiation Glovebox RT

0 5 10 15 20 25 30

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Nor

mal

ized

PCE

Time (h)

P3HT PCDTBT AnE-PV

UV Irradiation Glovebox RT

Figure A 3: Influence of UV radiation at room temperature in inert atmosphere : Jsc, Voc, FF and PCE. Parameters are normalized with respect to initial values

A possible explanation for the increase in Voc could be that the film gets heated up by the UV lamp, which alters the morphology, and consequently increases the Voc. However, as shown in Figure A 2, no increase in Voc was observed under even longer exposure to elevated temperature. To elucidate the origin of this

87

unexpected increase in Voc, photoelectron spectroscopy measurements (PES) were conducted on P3HT:[60]PCBM layer stacks after being exposed to UV light in inert atmosphere for 16 minutes.

292 290 288 286 284 2820

5000

10000

15000

20000

25000

30000

292 290 288 286 284 282

Reference UV glove box UV UHV

C1s

Inte

nsity

(co

unts

/s)

XPS (mon. AlKa)hn = 1486.7 eV

x20

Binding energy (eV)

170 168 166 164 1620

500

1000

1500

2000

2500

3000

3500

Reference UV glove box UV UHV

XPS (mon. AlK)h = 1486.7 eV

S2p

Inte

nsity

(cou

nts/

s)

Binding energy (eV)

Figure A 4: C1s and S2p core level spectra of as prepared and UV irradiated P3HT:[60]PCBM films measured using XPS. No modification in chemical composition could be observed after UV degradation

Based on core level spectra, as shown in Figure A 4, no modification of the surface chemical composition occurred, since the carbon and sulfur content remained unchanged and no variation in the related peak shapes was observed.

The analysis of the surface electronic structure by ultraviolet PES revealed distinct changes caused by UV irradiation. The work function of the P3HT:[60]PCBM film was determined to be 4.0 eV which remained unchanged after UV irradiation in the glove box, and also after exposure to the higher intensity Hg lamp attached to the vacuum chamber. Nevertheless, noticeable changes were found for the occupied electron states close to the Fermi level (EF). Figure A 5 includes He II spectra prior to and after exposure to UV radiation. Changes are observed for the electron density at the HOMO level and deeper lying electron states. Exposure to UV radiation leads to an increase in electron density between 2 and 4 eV. The most relevant for changes in the electrical device characteristics is the shift of HOMO level away from the Fermi level. The HOMO level of the as prepared film, as determined by linear extrapolation of the electron emission, is 0.55 eV below EF. It is to mention here that the HOMO levels obtained by UPS always slightly differ from the values measured by electrochemical methods [454]. The sample exposed to UV radiation in the glove box reveals a ~50 meV shift of the HOMO level, while when exposed to harsh illumination in the vacuum chamber, this effect is even more severe, resulting in a shift of the HOMO onset by 150 meV.

88

5 4 3 2 1 05 4 3 2 1 0

Nor

mal

ized

inte

nsity

Binding energy (eV)

UPS (He II)h = 40.8 eV

EF

Reference UV glove box UV UHV

Figure A 5: (left) Ultraviolet photoelectron spectra of the near HOMO region for as prepared and UV irradiated P3HT:[60]PCBM films; (right) schematic of the electronic properties of P3HT:[60]PCBM visualizing the shift of the HOMO level related to degradation by UV radiation

The observed shift of the HOMO level upon irradiation, and the increase of the P3HT band gap at the Al-cathode which forms a hole blocking layer thus reducing the recombination of electrons and holes at the cathode, explains the observed increase in open circuit voltage of the UV irradiated devices.

It has to be taken into account that the experiment was performed on pre-cathode annealed P3HT:[60]PCBM devices, where Voc is lower (0.4 V) compared to post-cathode (0.6 V) annealed devices. This is because a very thin face-on P3HT layer is formed on the surface during annealing in absence of the top aluminum cathode, which then acts an electron transport barrier from the active layer to the cathode [432].

UV-Vis radiation has been reported to cause changes in P3HT, primarily via a reduction of the conjugation length [214]. The shift of the valence band away from the Fermi level observed by UPS can be interpreted as the consequence of destruction of the thin face-on P3HT layer by UV irradiation. In this way a more selective contact is formed, due to the reduction of hole mobility in the damaged P3HT layer and a larger band gap of this layer, resulting in the contact becoming a hole blocking layer. This gives rise to an increase of Voc, which reaches the value of the post-cathode annealed devices.

The remains of this damaged layer constitute a blocking layer, which is manifested in the observed S-shape of the J(V) characteristics [455, 456], as shown in Figure A 6.

89

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0-100

-50

0

50

100

150

200

250

Reference 4min 8min 16min 32min

Cur

rent

Den

sity

(m

A/c

m²)

Voltage (V)

UV Radiation Glove Box P3HT:[60]PCBM

0.0 0.1 0.2 0.3 0.4 0.5 0.6-10

-8

-6

-4

-2

0

Figure A 6: Temporal progression of degradation of P3HT:[60]PCBM cells under UV radiation in inert atmosphere at room temperature. Inset shows the zoomed in S-shaped region of the J(V) curves

c) Thermo-Oxidative Degradation To investigate the influence of oxygen, the experiment from Section a) was performed in air. Due to the high temperature above the boiling point of water, the effect of water vapor could be excluded. Compared to devices shown in Figure A 2, which were exposed to high temperature in inert atmosphere, the degradation in the presence of oxygen shown in Figure A 7, accelerated for all three devices.

The strongest affected are the P3HT devices, where the current drops to zero within the first 4 hours of exposure. Compared to pure thermal stress, in the same period of time, the devices maintained a steady performance, thus this drop in performance can be attributed to interactions with oxygen.

Although PCDTBT and AnE-PV based devices initially decay faster than P3HT devices, after longer exposure they retain up to 20 % of the starting Jsc value at the time when the P3HT device no longer functions. It has to be taken into consideration that the thermally initiated morphological degradation component per se, as shown in the Figure A 2, already causes a drop in the performance of PCDTBT and AnE-PV devices, which is especially present in the initial hours of degradation.

90

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

Nor

mal

ized

Jsc

Time (h)

P3HT PCDTBT AnE-PV

Dark Air 150°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Nor

mal

ized

Voc

Time (h)

P3HT PCDTBT AnE-PV

Dark Air 150°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Nor

mal

ized

FF

Time (h)

P3HT PCDTBT AnE-PV

Dark Air 150°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Nor

mal

ized

PCE

Time (h)

P3HT PCDTBT AnE-PV

Dark Air 150°C

Figure A 7: Influence of oxygen in the dark at high temperature (150 °C): Jsc, Voc, FF and PCE. Parameters are normalized with respect to initial values

To define whether there is a chemical interaction taking place or is it a mere adsorption of oxygen, the degraded samples can be exposed to a short annealing at elevated temperature in inert atmosphere. However, it has to be noted that this approach is only possible for P3HT devices, as exposure to temperature, in the case of AnE-PV and PCDTBT leads to a significant morphological degradation. If the losses in performance are reversible, the dominating process can be identified as a formation of weak charge transfer complexes between P3HT and O2 [187, 189, 193]. However if the observed loss is irreversible, it indicates formation of covalent bonds between oxygen and the polymer [180, 457]. As shown in Figure A 8, the losses in the performance could not be reversed by annealing treatment, pointing to oxidation as the dominant degradation mechanism.

91

0.0 0.1 0.2 0.3 0.4

-8

-6

-4

-2

0

Cur

rent

Den

sity

(m

A/c

m²)

Voltage (V)

Reference DA 150°C 2h DA 150°C 2h A

Dark Air 150 °C P3HT:[60]PCBM

Figure A 8: J(V) curves of the fresh reference cell, a thermooxidatively degraded cell (DA 150 °C 2 h) and a degraded cell which was afterwards annealed for 10 minutes in inert atmosphere at 150 °C (DA 150 °C 2 h A)

d) Thermo-Oxidative Degradation in Presence of Humidity To investigate the influence of humidity, the devices were exposed to 50 °C, which is well below the boiling point of water. As shown in Figure A 9, all three polymers decay slower compared to the degradation at elevated temperature as described in Section c), in accordance with Arrhenius law [38, 106, 458, 459]. As expected, PCDTBT and AnE-PV devices are degraded to a lesser extent compared to high temperature degradation, due to less heat-induced morphological disordering.

In Section a) it was observed that, within the probed time, P3HT devices are stabile under exposure to high temperatures in absence of air. However, in this part of the experiment, when exposed to air at moderate temperature, P3HT decayed. Taken into consideration that its performance could be fully restored upon short thermal annealing in inert atmosphere, as shown in Figure A 10, the oxidation could be ruled out as the dominant mechanism, leaving two reversible processes to account for the losses, oxygen doping [187, 189, 191] and humidity induced degradation.

92

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

Nor

mal

ized

Jsc

Time (h)

P3HT PCDTBT AnE-PV

Dark Air 50°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Nor

mal

ized

Voc

Time (h)

P3HT PCDTBT AnE-PV

Dark Air 50°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Nor

mal

ized

FF

Time (h)

P3HT PCDTBT AnE-PV

Dark Air 50°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Nor

mal

ized

PCE

Time (h)

P3HT PCDTBT AnE-PV

Dark Air 50°C

Figure A 9: Influence of oxygen at moderate temperature (50 °C) in the dark in presence of humidity: Jsc, Voc, FF and PCE. Parameters are normalized with respect to initial values

0.0 0.1 0.2 0.3 0.4 0.5

-8

-6

-4

-2

0 Reference DA 50 °C 30m DA 50 °C 30m A

Cur

ren

t D

ensi

ty (

mA

/cm

2 )

Voltage (V)

Dark Air 50°C P3HT:[60]PCBM

Figure A 10: J(V) curves of the fresh reference cell, a cell degraded at moderate 50 °C temperature in the dark air with humidity present (DA 50 °C 2 h), and a degraded cell which was afterwards annealed for 10 minutes in inert atmosphere at 150 °C (DA 50 °C 2 h A)

93

As oxygen doping includes formation of weak charge transfer complexes between active layer and the O2 which diffuses into it, this process will be less prominent in crystalline organic materials, as they present a greater barrier for oxygen penetration [109, 111, 460, 461]. As P3HT is a crystalline polymer, while AnE-PV and PCDTBT are not, the oxygen will be able to penetrate their amorphous films much easier, thus doping them stronger than more crystalline P3HT active layers.

The charge transfer complex can act as a precursor for formation of a charge pair consisting of a positively charged radical cation on the polymer chain and superoxide radical anion [127, 191, 361], which are the main intermediates of the radical chain oxidation of polymer [171, 180]. It is therefore reasonable that the two tested amorphous polymers are stronger affected by oxidation, in contrast to semicrystalline P3HT which is under these conditions almost completely unoxidized, as shown in Figure A 9.

Humidity, on the other hand, mainly impacts the PEDOT:PSS layer and its interfaces, thus affecting all three devices to the same extent. As reported in [462], due to the hygroscopic nature of PEDOT:PSS, water molecules are attracted and retained by it. It has been shown that even encapsulated devices, in which only the lateral side of the cells (~40 nm) containing PEDOT:PSS is exposed to the atmosphere, results in a strong decay in performance due to the hydration of PEDOT:PSS and subsequent oxidation of the electrodes [302, 463]. As unencapsulated devices prior to evaporation were tested, the humidity can enter the devices throughout the whole active layer area. Since the tested layer stacks were degraded in absence of cathode, the primary contributions of the humidity to the decay can be attributed to the formation of insulating domains at the interface with the active layer, resulting from the reaction of acidic species in PSS with water [464], and from presence of Indium in the PEDOT:PSS, which has been etched from ITO in acid aqueous environment created in contact of PSS with water [465].

e) All Influences Combined – Photo-Thermo-Oxidative & Humidity Degradation Ultimately, the stack was exposed to visible light (1 Sun) in air at 50 °C. This condition combines at the same time the effects of light, oxygen, humidity and heat, corresponding to the common operating conditions.

As seen in Figure A 11, the short circuit current (Jsc) of all three devices decays exponentially, which is much faster than in the dark (compare with Figure A 9), while Voc decays at the similar pace as in the dark.

94

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

Nor

mal

ized

Jsc

Time (h)

P3HT PCDTBT AnE-PV

UV-vis Light Air 50°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Nor

mal

ized

Voc

Time (h)

P3HT PCDTBT AnE-PV

UV-vis Light Air 50°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Nor

mal

ized

FF

Time (h)

P3HT PCDTBT AnE-PV

UV-vis Light Air 50°C

0 2 4 6 8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Nor

mal

ized

PCE

Time (h)

P3HT PCDTBT AnE-PV

UV-vis Light Air 50°C

Figure A 11: Influence of UV-Vis radiation (1 Sun), oxygen, humidity and moderate temperature (50 °C): Jsc, Voc, FF and PCE. Parameters are normalized with respect to initial values

It is interesting to note that P3HT devices are slightly more stabile than the other two polymers. The stability of organic semiconducting polymers in contact with oxygen is related to their ionization potentials, namely HOMO levels lower than -5.2 eV are needed to ensure air stability [466, 467]. According to that, one would expect both AnE-PVcc and PCDTBT to exhibit higher stability, as their HOMO levels are -5.3, -5.5, compared to P3HT with -5.2 eV HOMO level. However, many examples from the literature show that certain structural features are even more decisive for the stability of polymers [318, 468, 469].

The superior stability of polymers containing thiophene moieties compared to those containing carbazole (as in case of PCDTBT) and dialkoxybenzene moieties (found in AnE-PV) can be explained by absence of readily cleavable bonds [318] such as C-O (present in dialkoxybenzene group) [42, 170, 171, 318], and C-N (present in carbazole group) [318, 470, 471]. Except for the cleavable C-O bond which is responsible for detachment of the ether groups from the benzene ring, another contribution to the low stability of AnE-PV represents the vinyl bond [42, 170, 171].

95

0.0 0.1 0.2 0.3 0.4-10

-8

-6

-4

-2

0 Reference SA 50°C 30min SA 50°C 30min A

Cur

rent

Den

sity

(m

A/c

m2 )

Voltage (V)

UV-Vis Radiation Air 50°C P3HT:[60]PCBM

Figure A 12: J(V) curves of the fresh reference cell, a cell degraded at 50 °C under 1 Sun illumination in humid air (SA 50 °C 30 min), and a degraded cell which was afterwards annealed for 10 minutes in inert atmosphere at 150 °C (SA 50 °C 30 min A)

In Figure A 12, after the short annealing step in inert atmosphere, the performance was only partially restored. This implies that both, photooxidation and reversible processes involving oxygen and humidity, as described in Section d), are taking place.

Overview Table A 1 summarizes the performances of the three polymer:fullerene devices tested under different stresses, for the same period of exposure. Although each polymer device decayed at a different rate, the harshest decay took place under exposure to light in combination with oxygen, humidity and temperature. Exposure to heat in absence of oxygen lead to the least pronounced reduction of device performance. In all of the tested conditions, P3HT devices exhibit a higher stability than the other two devices. The stabilities of AnE-PV and PCDTBT differ under different stresses, AnE-PV is slightly more stabile under thermal degradation in air, PCDTBT is slightly more stabile under exposure to UV radiation in inert atmosphere, and they are equally morphologically stabile.

The degradation in air increased with the increase in temperature in case of all three polymers, as expected according to the Arrhenius law for the temperature dependence of reaction rates. Under typical operating conditions, however, the temperature is not expected to exceed 50 °C, and such a moderate temperature did not cause additional degradation in any of the three polymers.

Exposure to the most severe fraction of illumination, UV radiation, in absence of air, affected the performance to a much lesser extent than when illuminated by UV-Vis radiation in air. This emphasizes the need to develop new polymers which, additional to high efficiency, exhibit high air stability [315, 319, 320, 472-477]. A

96

relatively larger difference was observed between the polymers in their stability under UV radiation. While the performance of P3HT devices suffers only slightly, in the same time, PCDTBT loses over half of its performance and the AnE-PV devices seizes to function almost completely. This confirms the necessity of embedding UV filters into the encapsulation of the cells [311] in order to avoid destruction of polymer in form of chain scission [173, 174, 180].

Condition Degradative stress Decay in performance [%]

Light Air Temperature P3HT AnE-PV PCDTBT Vis UV O2 H2O Mid High

Dark Glove box 150 °C 2 h x 10 50 50

UV Irradiation Glove box 8 min x 40 85 60

Dark Air 150 °C 2 h x x 50 85 95

Dark Air 50 °C 2 h x x x 50 60 75

Simulator Air 50 °C 2 h x x x x x 75 100 100 Table A 1: Performance comparison of the three polymers under tested conditions. The decay represents the percentage loss of the efficiency after 2 h of exposure to degradation (8 min in case of degradation with UV radiation)

Short Summary

Selective exposure of the layer stacks to different stress combinations in absence of cathode, starting from the thermal stress, followed by iterative addition of new stresses to the combination, enabled distinguishing the effect of each stress by itself on the solar cell parameters. Although differences in the stabilities of the three polymers were observed upon exposure to different stresses, the harshest decay for all three devices was observed under simultaneous exposure to UV-Vis radiation in air. Contrary to expected from the energy levels, the two polymers with the lower lying HOMO levels, exhibited inferior stability in air than the P3HT devices, further confirming the importance of structural features on the stability.

I also report on an interesting effect that was observed during degradation studies, namely the increase in Voc of the P3HT pre-cathode annealed devices upon exposure to UV light in inert atmosphere. This effect was investigated using UPS measurements, and assigned to degradation of the face-on P3HT layer grown on top of the active layer, characteristic for the pre-cathode annealed devices. Upon degradation, the interface to the aluminum cathode becomes more electron selective compared to the non-degraded device, which in turn reduces the recombination of holes and electrons at the metal interface and thus increases the Voc. The selectivity is mainly increased by two effects, first the decrease of the HOMO level of the polymer, and second the reduction in the hole mobility as the intramolecular charge transport gets disturbed due to the chain scission of the backbone.

97

Appendix B

Supporting information for Chapter 5.2

Hydrogen Donors

Additive # Ebandgap [eV] 1 3.7 2 4.2 3 4.2 4 4.3 5 4.3 6 3.9 7 2.8

Table B 1: Optical band gaps of hindered phenols, as determined via UV-Vis spectroscopy

Additive # Conc. [%] Jsc [mAcm-2] Voc [mV] FF [%] PCE [%] 1 20 6.6 350 54 1.3 2 20 6.5 360 56 1.3 3 10 5.0 330 53 0.9 4 20 8.5 370 50 1.6 5 20 6.5 460 52 1.6 6 20 6.4 370 52 1.2 7 0.03 7.0 350 51 1.3 Ref 7.0 390 54 1.5

Table B 2: Solar cell parameters of the optimized devices containing hindered phenols

PCE [%] Conc. [%] 0.03 0.3 3 10 20 Additive # 1 1.43 1.35 1.42 1.34 1.26 2 1.47 0.74 1.18 1.34 1.30 3 1.41 1.28 1.25 0.87 0.47 4 1.23 1.40 1.30 1.21 1.55 5 1.36 1.33 1.48 1.60 1.57 6 1.38 1.27 1.32 1.42 1.23 7 1.25 0.02 0.00 0.00 0.00 Ref 1.48

Table B 3: Optimization of hindered phenols concentration with regard to efficiency. Optimum concentration is defined as the maximum additive concentration yielding efficiency not smaller than 2/3 of the reference value

98

UV Absorbers

Additive # Ebandgap (eV) 1 3.2 2 3.1 3 3.4 4 3.2 5 3.5 6 3.1

Table B 4: Band gaps of the tested UV absorbers

600 700 800 900 10000.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

Nor

mal

ized

PL

(a.u

.)

Wavelength (nm)

Reference #1 #2 #3 #4 #5 #6

Figure B 1: Photoluminescence signal of P3HT:[60]PCBM non-degraded films containing each of the UV absorbers (concentration as noted in Table 5.4 ), as well as the reference without any additive. The PL excitation wavelength was 445 nm (~2.78 eV), which is below the optical gap of each of the additives

Additive # Conc. [%] Jsc [mAcm-2] Voc [mV] FF [%] PCE [%] 1 20 6.0 350 57 1.2 2 10 8.5 380 56 1.8 3 20 6.5 380 56 1.4 4 20 7.0 390 56 1.5 5 10 7.1 360 52 1.4 6 10 6.3 530 45 1.5 Ref 7.0 390 54 1.5

Table B 5: Solar cell parameters of the optimized devices containing UV absorbers

99

PCE [%] Conc. [%] 0.03 0.3 3 10 20 Additive # 1 1.32 1.36 1.38 1.34 1.21 2 1.34 1.33 1.34 1.82 1.60 3 1.44 1.44 1.42 1.50 1.36 4 1.34 1.42 1.40 1.64 1.54 5 1.36 1.34 1.29 1.35 1.25 6 1.42 1.55 2.07 1.50 0.92 Ref 1.48

Table B 6: Optimization of UV absorbers concentration with regard to efficiency. Optimum concentration is defined as the maximum additive concentration yielding a PCE not smaller than 2/3 of the reference PCE

Hydroperoxide Decomposers

PCE [%] Conc. [%] 0.03 0.3 3 10 20 Additive # 1 1.38 1.54 2.32 1.42 0.29 2 1.35 1.34 1.36 1.09 1.10 3 1.37 1.32 1.36 1.08 0.71 4 1.33 1.34 0.78 0.22 0.05 5 1.31 1.28 1.31 1.15 1.06 Ref 1.48

Table B 7: Optimization of hydroperoxide decomposers concentration with regard to efficiency. Optimum concentration is defined as the maximum additive concentration yielding a PCE not smaller than 2/3 of the reference PCE

Additive # Conc. [%] Jsc [mAcm-2] Voc [mV] FF [%] PCE [%] 1 3 7.0 540 62 2.3 2 20 6.0 350 52 1.1 3 10 6.9 310 50 1.1 4 3 6.1 390 33 0.8 5 20 5.1 380 55 1.1 Ref 7.0 390 54 1.5

Table B 8: Solar cell parameters of the optimized devices containing hydroperoxide decomposers

100

600 700 800 900 10000.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

Nor

mal

ized

PL

(a.u

.)

Wavelength (nm)

Reference #1 #2 #3 #4 #5

Figure B 2: (top) Photoluminescence signal of P3HT:[60]PCBM non-degraded films containing each of the hydroperoxide decomposers (concentration as noted in Table B 8), as well as the reference without any additive. The PL excitation wavelength was 445 nm (~2.78 eV), which is below the optical gap of each of the additives; (bottom) Optical microscopy images (100x zoom) of (left) P3HT:[60]PCBM reference film, (middle) containing Additive #1 and (right) Additive #5

200 300 400 500 600 700 8000.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Nor

mal

ized

Abs

orba

nce

(a.u

.)

Wavelength (nm)

Reference #1 #2 #3 #4 #5

200 300 400 500 600 700 800

-0.6

-0.4

-0.2

0.0

0.2

#1 #2 #3 #4 #5

( Abs

orba

nce(

120h

) -

Abs

orba

nce

(0h)

)/

Abs

orba

nce

(0h,

500

nm)

(a.u

.)

Wavelength (nm)

Reference

Figure B 3: (left) Normalized UV-Vis absorption spectra of the non-degraded films with additives and one reference with no additive on CaF substrates; (right) Relative change in absorption of the films after 120 h of continuous 1000 W/m² illumination in ambient. Note that all spectra were normalized to the absorption peak at 500nm of the corresponding non-degraded film

101

Additive # Ebandgap (eV) 1 5.6 2 4.3 3 - 4 2.8 5 -

Table B 9: Optical band gaps of hydroperoxide decomposing additives, as determined via UV-Vis spectroscopy. Note that the band gaps of Additive #1 and #5 could not be determined as it was impossible to form films for optical measurements

0 20 40 60 80 100 120 140 1600.00.20.40.60.81.01.21.41.61.82.02.22.42.62.8

eta

(%)

Time (h)

Reference Reference fit

Additive #1 Additive #1 fit Additive #5 Additive #5 fit

0 20 40 60 80 100 120 140 1600

1

2

3

4

5

6

7

8

Jsc

(mA

/cm

²)

Time (h)

Reference Reference fit

Additive #1 Additive #1 fit Additive #5 Additive #5 fit

0 20 40 60 80 100 120 140 1600

10

20

30

40

50

60

Additive #1 Additive #1 fit Additive #5 Additive #5 fit

FF

(%

)

Time (h)

Reference Reference fit

0 20 40 60 80 100 120 140 1600.00.20.40.60.81.01.21.41.61.82.02.22.42.62.8

eta

(%)

Time (h)

Reference Reference fit

Additive #1 Additive #1 fit Additive #5 Additive #5 fit

Figure B 4: Long-term PCE, Jsc, Voc and FF of solar cells with no additives and containing Additive #1 and #5. The solar cells were measured continuously during 1000 W/m² illumination of a metal halide lamp. The time dependence of the parameters was fitted using a bi-exponential decay curve

102

List of Own Publications

First Author Publications

Turkovic V., Engmann S., Gobsch G., Hoppe H., “Methods in Determination of Morphological Degradation of Polymer:Fullerene Solar Cells”, Synthetic Metals 161, p 2534 (2012)

Turkovic V., Engmann S., Egbe D.A.M., Himmerlich M., Krischok S., Hoppe H., Gobsch G., “Multiple Stress Degradation Analysis of Active Layers in Organic Photovoltaics”, Solar Energy Materials and Solar Cells, 120 B, p 654 (2014)

Turkovic V., Engmann S., Tsierkezos N., Hoppe H., Ritter U., Gobsch G., “Long-Term Stabilization of Organic Solar Cells using Hindered Phenols”, ACS Applied Materials & Interfaces, DOI: 10.1021/am5024989

Turkovic V., Engmann S., Tsierkezos N., Hoppe H., Ritter U., Gobsch G., “Long-Term Stabilization of Organic Solar Cells using UV Absorbers”, in submission

Co-Author Publications

Engmann S., Turkovic V., Hoppe H., Gobsch G., “Ellipsometric Investigation of the Shape of Nanodomains in Polymer/Fullerene Films”, Advanced Energy Materials 1, p 684 (2011)

Engmann S., Turkovic V., Hoppe H., Gobsch G., “Aging of Polymer/Fullerene Films: Temporal Development of Composition Profiles”, Synthetic Metals 161, p 2540 (2012)

Engmann S., Turkovic V., Denner P., Hoppe H., Gobsch G., “Optical Order of the Polymer Phase within Polymer/Fullerene Blend Films”, Journal of Polymer Science B 50, p 1363 (2012)

Engmann S., Machalett M., Turkovic V., Roesch R., Raedlein E., Gobsch G., Hoppe H., “Photon Recycling Across a Ultraviolet-Blocking Layer by Luminescence in Polymer Solar Cells”, Journal of Applied Physics 112, 034517 (2012)

Engmann S., Turkovic V., Hoppe H., Gobsch G., “Direct Correlation of the Organic Solar Cell Device Performance to the In-Depth Distribution of Highly Ordered Polymer Domains in Polymer/Fullerene Films”, Advanced Energy Materials 3, p 1463 (2013)

Engmann S., Turkovic V., Hoppe H., Gobsch G., “Revealing the Active Layer Morphology within Complete Solar Cell Devices via Spectroscopic Ellipsometry”, Journal of Physical Chemistry C 117, p 25205 (2013)

103

Bibliography

1. Key World Energy Statistics; International Energy Agency, Soregraph: Nanterre, France, 2013.

2. Kruger, P. Alternative Energy Resources: The Quest for Sustainable Energy; Wiley: Hoboken, NJ, USA, 2006; pp 1-272.

3. Speth, J. G. Red Sky at Morning: America and the Crisis of the Global Environment; Yale University Press: New Haven, CT, USA, 2004; pp 1-307.

4. Balzani, V.; Armaroli, N. Energy for a Sustainable World: From the Oil Age to a Sun-Powered Future; Wiley: Weinheim, Germany, 2010; pp 1-376.

5. Statistical Review of World Energy 2013; British Petroleum, Pureprint Group: London, UK, 2013.

6. Kerr, R. A. Oil Resources Splitting the Difference between Oil Pessimists and Optimists. Science. 2009, 326, 1048.

7. Cleveland, C. J. Net Energy from the Extraction of Oil and Gas in the United States. Energy. 2005, 30, 769-782.

8. I.P.C.C. Fourth Assesment Report: Mitigation of Climate Change; The Intergovernmental Panel on Climate Change, Cambridge University Press: Cambridge, UK, 2013.

9. Wilson, E. O. The Creation: An Appeal to Save Life on Earth; W. W. Norton & Company: New York, NY, USA, 2010; pp 1-173.

10. Kirilenko, A. P.; Sedjo, R. A. Climate Change Impacts on Forestry. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 19697-19702.

11. Brown, L. R. Could Food Shortages Bring Down Civilization? Sci. Am. 2009, 300, 50-57.

12. Special Issue on Food Security. Science. 2010, 327, 747-906. 13. Godfray, H. C. J.; Beddington, J. R.; Crute, I. R.; Haddad, L.; Lawrence,

D.; Muir, J. F.; Pretty, J.; Robinson, S.; Thomas, S. M.; Toulmin, C. Food Security: The Challenge of Feeding 9 Billion People. Science. 2010, 327, 812-818.

14. Barnett, T. P.; Adam, J. C.; Lettenmaier, D. P. Potential Impacts of a Warming Climate on Water Availability in Snow-Dominated Regions. Nature. 2005, 438, 303-309.

15. Millero, F. J. The Marine Inorganic Carbon Cycle. Chem. Rev. 2007, 107, 308-341.

16. Jorgenson, M. T.; Shur, Y. L.; Pullman, E. R. Abrupt Increase in Permafrost Degradation in Arctic Alaska. Geophys. Res. Lett. 2006, 33, L02503.

17. Jacobson, M. Z. Review of Solutions to Global Warming, Air Pollution, and Energy Security. Energ. Environ. Sci. 2009, 2, 148-173.

18. Resch, G.; Held, A.; Faber, T.; Panzer, C.; Toro, F.; Haas, R. Potentials and Prospects for Renewable Energies at Global Scale. Energ. Policy. 2008, 36, 4048-4056.

19. Fthenakis, V.; Mason, J. E.; Zweibel, K. The Technical, Geographical, and Economic Feasibility for Solar Energy to Supply the Energy Needs of the US. Energ. Policy. 2009, 37, 387-399.

104

20. Jacobson, M. Z.; Delucchi, M. A. A Path to Sustainable Energy by 2030. Sci. Am. 2009, 301, 58-65.

21. Kohl, H. The Development of Renewable Energy Carriers: Renewable Energy Sources on the Rise. In Renewable Energy: Sustainable Energy Concepts for the Future; R. Wengenmayr and T. Buhrke, Eds.; Wiley-VCH: Weinheim, Germany, 2008; pp 4-13.

22. Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Polymer Photovoltaic Cells - Enhanced Efficiencies Via a Network of Internal Donor-Acceptor Heterojunctions. Science. 1995, 270, 1789-1791.

23. Brabec, C. J.; Gowrisanker, S.; Halls, J. J. M.; Laird, D.; Jia, S.; Williams, S. P. Polymer-Fullerene Bulk-Heterojunction Solar Cells. Adv. Mater. 2010, 22, 3839-3856.

24. Zhan, X.; Zhu, D. Conjugated Polymers for High-Efficiency Organic Photovoltaics. Polym. Chem. 2010, 1, 409-419.

25. Facchetti, A. Pi-Conjugated Polymers for Organic Electronics and Photovoltaic Cell Applications. Chem. Mater. 2011, 23, 733-758.

26. Brunetti, F. G.; Kumar, R.; Wudl, F. Organic Electronics from Perylene to Organic Photovoltaics: Painting a Brief History with a Broad Brush. J. Mater. Chem. 2010, 20, 2934-2948.

27. Heeger, A. J. Semiconducting Polymers: The Third Generation. Chem. Soc. Rev. 2010, 39, 2354-2371.

28. Heliatek Consolidates Its Technology Leadership by Establishing a New World Record for Organic Solar Technology with a Cell Efficiency of 12%. http://www.heliatek.com/newscenter/latest_news/neuer-weltrekord-fur-organische-solarzellen-heliatek-behauptet-sich-mit-12-zelleffizienz-als-technologiefuhrer/ (accessed Jan 16, 2013).

29. Tromholt, T.; Manceau, M.; Helgesen, M.; Carle, J. E.; Krebs, F. C. Degradation of Semiconducting Polymers by Concentrated Sunlight. Sol. Energ. Mat. Sol. Cells. 2011, 95, 1308-1314.

30. Rosch, R.; Tanenbaum, D. M.; Jorgensen, M.; Seeland, M.; Barenklau, M.; Hermenau, M.; Voroshazi, E.; Lloyd, M. T.; Galagan, Y.; Zimmermann, B.; Wurfel, U.; Hosel, M.; Dam, H. F.; Gevorgyan, S. A.; Kudret, S.; Maes, W.; Lutsen, L.; Vanderzande, D.; Andriessen, R.; Teran-Escobar, G.; Lira-Cantu, M.; Rivaton, A.; Uzunoglu, G. Y.; Germack, D.; Andreasen, B.; Madsen, M. V.; Norrman, K.; Hoppe, H.; Krebs, F. C. Investigation of the Degradation Mechanisms of a Variety of Organic Photovoltaic Devices by Combination of Imaging Techniques-the ISOS-3 Inter-Laboratory Collaboration. Energ. Environ. Sci. 2012, 5, 6521-6540.

31. Reese, M. O.; Gevorgyan, S. A.; Jorgensen, M.; Bundgaard, E.; Kurtz, S. R.; Ginley, D. S.; Olson, D. C.; Lloyd, M. T.; Moryillo, P.; Katz, E. A.; Elschner, A.; Haillant, O.; Currier, T. R.; Shrotriya, V.; Hermenau, M.; Riede, M.; Kirov, K. R.; Trimmel, G.; Rath, T.; Inganas, O.; Zhang, F.; Andersson, M.; Tvingstedt, K.; Lira-Cantu, M.; Laird, D.; McGuiness, C.; Gowrisanker, S.; Pannone, M.; Xiao, M.; Hauch, J.; Steim, R.; DeLongchamp, D. M.; Roesch, R.; Hoppe, H.; Espinosa, N.; Urbina, A.; Yaman-Uzunoglu, G.; Bonekamp, J.-B.; van Breemen, A. J. J. M.; Girotto, C.; Voroshazi, E.; Krebs, F. C. Consensus Stability Testing Protocols for

105

Organic Photovoltaic Materials and Devices. Sol. Energ. Mat. Sol. Cells. 2011, 95, 1253-1267.

32. Norrman, K.; Larsen, N. B.; Krebs, F. C. Lifetimes of Organic Photovoltaics: Combining Chemical and Physical Characterisation Techniques to Study Degradation Mechanisms. Sol. Energ. Mat. Sol. Cells. 2006, 90, 2793-2814.

33. Neugebauer, H.; Brabec, C. J.; Hummelen, J. C.; Janssen, R. A. J.; Sariciftci, N. S. Stability Studies and Degradation Analysis of Plastic Solar Cell Materials by FTIR Spectroscopy. Synt. Met. 1999, 102, 1002-1003.

34. Neugebauer, H.; Brabec, C.; Hummelen, J. C.; Sariciftci, N. S. Stability and Photodegradation Mechanisms of Conjugated Polymer/Fullerene Plastic Solar Cells. Sol. Energ. Mat. Sol. Cells. 2000, 61, 35-42.

35. Manceau, M.; Rivaton, A.; Gardette, J.-L.; Guillerez, S.; Lemaitre, N. Light-Induced Degradation of the P3HT-Based Solar Cells Active Layer. Sol. Energy Mater. Sol. Cells. 2011, 95, 1315-1325.

36. Manceau, M.; Gaume, J.; Rivaton, A.; Gardette, J.-L.; Monier, G.; Bideux, L. Further Insights into the Photodegradation of Poly(3-Hexylthiophene) by Means of X-Ray Photoelectron Spectroscopy. Thin Solid Films. 2010, 518, 7113-7118.

37. Krebs, F. C.; Carlé, J. E.; Cruys-Bagger, N.; Andersen, M.; Lilliedal, M. R.; Hammond, M. A.; Hvidt, S. Lifetimes of Organic Photovoltaics: Photochemistry, Atmosphere Effects and Barrier Layers in ITO-MEHPPV:PCBM-Aluminium Devices. Sol. Energ. Mat. Sol. Cells. 2005, 86, 499-516.

38. Hintz, H.; Egelhaaf, H. J.; Luer, L.; Hauch, J.; Peisert, H.; Chasse, T. Photodegradation of P3HT-a Systematic Study of Environmental Factors. Chem. Mater. 2011, 23, 145-154.

39. Gevorgyan, S. A.; Jørgensen, M.; Krebs, F. C. A Setup for Studying Stability and Degradation of Polymer Solar Cells. Sol. Energ. Mat. Sol. Cells. 2008, 92, 736-745.

40. Chambon, S.; Rivaton, A.; Gardette, J.-L.; Firon, M. Durability of MDMO-PPV and MDMO-PPV: PCBM Blends under Illumination in the Absence of Oxygen. Sol. Energ. Mat. Sol. Cells. 2008, 92, 785-792.

41. Chambon, S.; Rivaton, A.; Gardette, J.-L.; Firon, M. Photo- and Thermal Degradation of MDMO-PPV:PCBM Blends. Sol. Energ. Mat. Sol. Cells. 2007, 91, 394-398.

42. Chambon, S.; Manceau, M.; Firon, M.; Cros, S.; Rivaton, A.; Gardette, J.-L. Photo-Oxidation in an 18O2 Atmosphere: A Powerful Tool to Elucidate the Mechanism of UV-Visible Light Oxidation of Polymers - Application to the Photodegradation of MDMO-PPV. Polymer. 2008, 49, 3288-3294.

43. Jorgensen, M.; Norrman, K.; Krebs, F. C. Stability/Degradation of Polymer Solar Cells. Sol. Energ. Mat. Sol. Cells. 2008, 92, 686-714.

44. Reese, M. O.; Nardes, A. M.; Rupert, B. L.; Larsen, R. E.; Olson, D. C.; Lloyd, M. T.; Shaheen, S. E.; Ginley, D. S.; Rumbles, G.; Kopidakis, N. Photoinduced Degradation of Polymer and Polymer-Fullerene Active Layers: Experiment and Theory. Adv. Funct. Mater. 2010, 20, 3476-3483.

45. Turkovic, V.; Engmann, S.; Egbe, D. A. M.; Himmerlich, M.; Krischok, S.; Gobsch, G.; Hoppe, H. Multiple Stress Degradation Analysis of the Active

106

Layer in Organic Photovoltaics. Sol. Energ. Mat. Sol. Cells. 2014, 120, Part B, 654-668.

46. Jorgensen, M.; Norrman, K.; Gevorgyan, S. A.; Tromholt, T.; Andreasen, B.; Krebs, F. C. Stability of Polymer Solar Cells. Adv. Mater. 2012, 24, 580-612.

47. Grossiord, N.; Kroon, J. M.; Andriessen, R.; Blom, P. W. M. Degradation Mechanisms in Organic Photovoltaic Devices. Org. Electron. 2012, 13, 432-456.

48. Lee, J. U.; Jung, J. W.; Jo, J. W.; Jo, W. H. Degradation and Stability of Polymer-Based Solar Cells. J. Mater. Chem. 2012, 22, 24265-24283.

49. Brabec, C. J.; Hauch, J. A.; Schilinsky, P.; Waldauf, C. Production Aspects of Organic Photovoltaics and Their Impact on the Commercialization of Devices. MRS Bull. 2005, 30, 50-52.

50. Brabec, C. J. Organic Photovoltaics: Technology and Market. Sol. Energ. Mat. Sol. Cells. 2004, 83, 273-292.

51. McNeill, R.; Weiss, D. E. A Xanthene Polymer with Semiconducting Properties. Aust. J. Chem. 1959, 12, 643-656.

52. Bolto, B. A.; McNeill, R.; Weiss, D. E. Electronic Conduction in Polymers. III. Electronic Properties of Polypyrrole. Aust. J. Chem. 1963, 16, 1090-1103.

53. Bolto, B. A.; Weiss, D. E. Electronic Conduction in Polymers. II. Electrochemical Reduction of Polypyrrole at Controlled Potential. Aust. J. Chem. 1963, 16, 1076-1089.

54. McNeill, R.; Weiss, D. E.; Wardlaw, J. H.; Siudak, R. Electronic Conduction in Polymers. I. Chemical Structure of Polypyrrole. Aust. J. Chem. 1963, 16, 1056-1075.

55. Macpherson, A. S.; Siudak, R.; Weiss, D. E.; Willis, D. Electronic Conduction in Polymers. VI. Chemical Consequences of Space Charge Effects in Conducting Polymers. Aust. J. Chem. 1965, 18, 493-505.

56. McNeill, R.; Weiss, D. E.; Willis, D. Electronic Conduction in Polymers. IV. Polymers from Imidazole and Pyridine. Aust. J. Chem. 1965, 18, 477-486.

57. Jozefowicz, M.; Buvet, R. Preparation De Poly-P-Phenylenes Oligomeres. Compt. Rend. Hebd. Seances Acad. Sci. 1961, 253, 1801.

58. Jozefovicz, M.; Buvet, R. Physicochimie Macromoleculaire - Caracterisation Spectrophotometrique Des Melanges De P-Oligophenylenes. Compt. Rend. Hebd. Seances Acad. Sci. 1962, 254, 284.

59. Jozefowicz, M.; Belorgey, G.; Yu, L. T.; Buvet, R. Oxydation Et Reduction De Polyanilines Oligomeres. Compt. Rend. Hebd. Seances Acad. Sci. 1965, 260, 6367.

60. Combarel, M. F.; Belorgey, G.; Jozefowi.M; Yu, L. T.; Buvet, R. Conductivite En Courant Continu Des Polyanilines Oligomeres - Influence De Letat Acide-Base Sur La Conductivite Electronique. Compt. Rend. Hebd. Seances Acad. Sci. C. 1966, 262, 459.

61. Desurvil, R.; Jozefowicz, M.; Yu, L. T.; Perichon, J.; Buvet, R. Electrochemical Chains Using Protolytic Organic Semiconductors. Electrochim. Acta. 1968, 13, 1451.

107

62. Cristofi, F.; Desurvil, R.; Jozefowicz, M.; Yu, L. T.; Buvet, R. Electrochemical Properties of Polyaniline Sulfates. Compt. Rend. Hebd. Seances Acad. Sci. C. 1969, 268, 1346.

63. Jozefowicz, M. Preparation Et Proprietes Physiques Des Parapolyphenylenes. Bull. Soc. Chim. Fr. 1963, 2036.

64. Chiang, C. K.; Fincher, C. R.; Park, Y. W.; Heeger, A. J.; Shirakawa, H.; Louis, E. J.; Gau, S. C.; MacDiarmid, A. G. Electrical-Conductivity in Doped Polyacetylene. Phys. Rev. Lett. 1977, 39, 1098-1101.

65. Shirakawa, H.; Louis, E. J.; Macdiarmid, A. G.; Chiang, C. K.; Heeger, A. J. Synthesis of Electrically Conducting Organic Polymers - Halogen Derivatives of Polyacetylene, (CH)X. J. Chem. Soc., Chem. Commun. 1977, 578-580.

66. Chiang, C. K.; Druy, M. A.; Gau, S. C.; Heeger, A. J.; Louis, E. J.; Macdiarmid, A. G.; Park, Y. W.; Shirakawa, H. Synthesis of Highly Conducting Films of Derivatives of Polyacetylene, (CH)X. J. Am. Chem. Soc. 1978, 100, 1013-1015.

67. Chiang, C. K.; Park, Y. W.; Heeger, A. J.; Shirakawa, H.; Louis, E. J.; Macdiarmid, A. G. Conducting Polymers - Halogen Doped Polyacetylene. J. Chem. Phys. 1978, 69, 5098-5104.

68. Kallmann, H.; Pope, M. Photovoltaic Effect in Organic Crystals. J. Chem. Phys. 1959, 30, 585-586.

69. Deibel, C.; Mack, D.; Gorenflot, J.; Scholl, A.; Krause, S.; Reinert, F.; Rauh, D.; Dyakonov, V. Energetics of Excited States in the Conjugated Polymer Poly(3-Hexylthiophene). Phys. Rev. B. 2010, 81.

70. Theander, M.; Yartsev, A.; Zigmantas, D.; Sundstroem, V.; Mammo, W.; Andersson, M. R.; Inganaes, O. Photoluminescence Quenching at a Polythiophene/C60 Heterojunction. Phys. Rev. B. 2000, 61, 12957-12963.

71. Tang, C. W. 2-Layer Organic Photovoltaic Cell. Appl. Phys. Lett. 1986, 48, 183-185.

72. Sariciftci, N. S.; Smilowitz, L.; Heeger, A. J.; Wudl, F. Photoinduced Electron Transfer from a Conducting Polymer to Buckminsterfullerene. Science. 1992, 258, 1474-1476.

73. Brabec, C. J.; Heeney, M.; McCulloch, I.; Nelson, J. Influence of Blend Microstructure on Bulk Heterojunction Organic Photovoltaic Performance. Chem. Soc. Rev. 2011, 40, 1185-1199.

74. Brady, M. A.; Su, G. M.; Chabinyc, M. L. Recent Progress in the Morphology of Bulk Heterojunction Photovoltaics. Soft Matter. 2011, 7, 11065-11077.

75. Brinkmann, M. Structure and Morphology Control in Thin Films of Regioregular Poly(3-Hexylthiophene). Journal of Polymer Science Part B-Polymer Physics. 2011, 49, 1218-1233.

76. Ruderer, M. A.; Mueller-Buschbaum, P. Morphology of Polymer-Based Bulk Heterojunction Films for Organic Photovoltaics. Soft Matter. 2011, 7, 5482-5493.

77. Liu, F.; Gu, Y.; Jung, J. W.; Jo, W. H.; Russell, T. P. On the Morphology of Polymer-Based Photovoltaics. J. Polym. Sci., Part B: Polym. Phys. 2012, 50, 1018-1044.

108

78. Nagarjuna, G.; Venkataraman, D. Strategies for Controlling the Active Layer Morphologies in OPVs. Journal of Polymer Science Part B-Polymer Physics. 2012, 50, 1045-1056.

79. Flory, P. J. Thermodynamics of High Polymer Solutions. The Journal of Chemical Physics. 1942, 10, 51-61.

80. Huggins, M. L. Solutions of Long Chain Compounds. The Journal of Chemical Physics. 1941, 9, 440-440.

81. Ma, W. L.; Yang, C. Y.; Gong, X.; Lee, K.; Heeger, A. J. Thermally Stable, Efficient Polymer Solar Cells with Nanoscale Control of the Interpenetrating Network Morphology. Adv. Funct. Mater. 2005, 15, 1617-1622.

82. Koningsveld, R.; Stockmayer, W. H.; Nies, E. Polymer Phase Diagrams; Oxford University Press, Inc.: New Yok, NY, USA, 2001; pp 1-341.

83. Shaheen, S. E.; Brabec, C. J.; Sariciftci, N. S.; Padinger, F.; Fromherz, T.; Hummelen, J. C. 2.5% Efficient Organic Plastic Solar Cells. Appl. Phys. Lett. 2001, 78, 841-843.

84. Campbell, A. R.; Hodgkiss, J. M.; Westenhoff, S.; Howard, I. A.; Marsh, R. A.; McNeill, C. R.; Friend, R. H.; Greenham, N. C. Low-Temperature Control of Nanoscale Morphology for High Performance Polymer Photovoltaics. Nano Lett. 2008, 8, 3942-3947.

85. Zhang, F. L.; Jespersen, K. G.; Bjorstrom, C.; Svensson, M.; Andersson, M. R.; Sundstrom, V.; Magnusson, K.; Moons, E.; Yartsev, A.; Inganas, O. Influence of Solvent Mixing on the Morphology and Performance of Solar Cells Based on Polyfluorene Copolymer/Fullerene Blends. Adv. Funct. Mater. 2006, 16, 667-674.

86. Park, S. H.; Roy, A.; Beaupre, S.; Cho, S.; Coates, N.; Moon, J. S.; Moses, D.; Leclerc, M.; Lee, K.; Heeger, A. J. Bulk Heterojunction Solar Cells with Internal Quantum Efficiency Approaching 100%. Nature Photonics. 2009, 3, 297-302.

87. Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.; Moses, D.; Heeger, A. J.; Bazan, G. C. Efficiency Enhancement in Low-Bandgap Polymer Solar Cells by Processing with Alkane Dithiols. Nat. Mater. 2007, 6, 497-500.

88. Yao, Y.; Hou, J.; Xu, Z.; Li, G.; Yang, Y. Effect of Solvent Mixture on the Nanoscale Phase Separation in Polymer Solar Cells. Adv. Funct. Mater. 2008, 18, 1783-1789.

89. Lee, J. K.; Ma, W. L.; Brabec, C. J.; Yuen, J.; Moon, J. S.; Kim, J. Y.; Lee, K.; Bazan, G. C.; Heeger, A. J. Processing Additives for Improved Efficiency from Bulk Heterojunction Solar Cells. J. Am. Chem. Soc. 2008, 130, 3619-3623.

90. Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.-T.; Wu, Y.; Li, G.; Ray, C.; Yu, L. For the Bright Future-Bulk Heterojunction Polymer Solar Cells with Power Conversion Efficiency of 7.4%. Adv. Mater. 2010, 22, E135-E138.

91. Chen, F.-C.; Tseng, H.-C.; Ko, C.-J. Solvent Mixtures for Improving Device Efficiency of Polymer Photovoltaic Devices. Appl. Phys. Lett. 2008, 92.

92. Zhao, Y.; Xie, Z. Y.; Qu, Y.; Geng, Y. H.; Wang, L. X. Solvent-Vapor Treatment Induced Performance Enhancement of Poly(3-Hexylthiophene): Methanofullerene Bulk-Heterojunction Photovoltaic Cells. Appl. Phys. Lett. 2007, 90.

109

93. Park, J. H.; Kim, J. S.; Lee, J. H.; Lee, W. H.; Cho, K. Effect of Annealing Solvent Solubility on the Performance of Poly(3-Hexylthiophene)/Methanofullerene Solar Cells. J. Phys. Chem. C. 2009, 113, 17579-17584.

94. Jo, J.; Na, S.-I.; Kim, S.-S.; Lee, T.-W.; Chung, Y.; Kang, S.-J.; Vak, D.; Kim, D.-Y. Three-Dimensional Bulk Heterojunction Morphology for Achieving High Internal Quantum Efficiency in Polymer Solar Cells. Adv. Funct. Mater. 2009, 19, 2398-2406.

95. van Bavel, S. S.; Sourty, E.; de With, G.; Loos, J. Three-Dimensional Nanoscale Organization of Bulk Heterojunction Polymer Solar Cells. Nano Lett. 2009, 9, 507-513.

96. Li, G.; Shrotriya, V.; Huang, J. S.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. High-Efficiency Solution Processable Polymer Photovoltaic Cells by Self-Organization of Polymer Blends. Nat. Mater. 2005, 4, 864-868.

97. Li, G.; Yao, Y.; Yang, H.; Shrotriya, V.; Yang, G.; Yang, Y. "Solvent Annealing" Effect in Polymer Solar Cells Based on Poly(3-Hexylthiophene) and Methanofullerenes. Adv. Funct. Mater. 2007, 17, 1636-1644.

98. Fried, J. R. Polymer Science and Technology; Prentice Hall PTR: Englewood Cliffs, NJ, USA, 1995; pp 1-509.

99. Jones, R. A. L. Soft Condensed Matter; Oxford University Press: New York, NY, USA, 2002; pp 1-193.

100. Jones, R. A. L. Interfacial Aspects of Semiconducting Polymer Devices. In Semiconducting Polymers - Chemistry, Physics and Engineering, Volume 1; G. Hadziioannou and G.G. Malliaras, Eds.; Wiley-VCH Verlag: Weinheim, Germany, 2007; pp 121-140.

101. Arias, A. C.; Corcoran, N.; Banach, M.; Friend, R. H.; MacKenzie, J. D.; Huck, W. T. S. Vertically Segregated Polymer-Blend Photovoltaic Thin-Film Structures through Surface-Mediated Solution Processing. Appl. Phys. Lett. 2002, 80, 1695-1697.

102. Campoy-Quiles, M.; Ferenczi, T.; Agostinelli, T.; Etchegoin, P. G.; Kim, Y.; Anthopoulos, T. D.; Stavrinou, P. N.; Bradley, D. D. C.; Nelson, J. Morphology Evolution Via Self-Organization and Lateral and Vertical Diffusion in Polymer: Fullerene Solar Cell Blends. Nat. Mater. 2008, 7, 158-164.

103. Chabinyc, M. L.; Salleo, A.; Wu, Y. L.; Liu, P.; Ong, B. S.; Heeney, M.; McCulloch, I. Lamination Method for the Study of Interfaces in Polymeric Thin Film Transistors. J. Am. Chem. Soc. 2004, 126, 13928-13929.

104. Germack, D. S.; Chan, C. K.; Hamadani, B. H.; Richter, L. J.; Fischer, D. A.; Gundlach, D. J.; DeLongchamp, D. M. Substrate-Dependent Interface Composition and Charge Transport in Films for Organic Photovoltaics. Appl. Phys. Lett. 2009, 94.

105. Germack, D. S.; Chan, C. K.; Kline, R. J.; Fischer, D. A.; Gundlach, D. J.; Toney, M. F.; Richter, L. J.; DeLongchamp, D. M. Interfacial Segregation in Polymer/Fullerene Blend Films for Photovoltaic Devices. Macromolecules. 2010, 43, 3828-3836.

106. Zweifel, H. Stabilization of Polymeric Materials; Springer Verlag: Berlin, Germany, 1998; pp 1-213.

110

107. Billingham, N. C. Physical Phenomena in the Oxidation and Stabilization of Polymers. In Oxidation Inhibition in Organic Materials, Volume 2; J. Pospisil and P. Klemchuk, Eds.; CRC Press: Boca Raton, FL, USA, 1990; pp 249-297.

108. Rabek, J. F. Polymer Photodegradation - Mechanisms and Experimental Methods; Chapman & Hall: London, UK, 1995; pp 1-654.

109. Billingham, N. C.; Prentice, P.; Walker, T. J. Some Effects of Morphology on Oxidation and Stabilization of Polyolefins. Journal of Polymer Science: Polymer Symposia. 1976, 57, 287-297.

110. Horng, P. L.; Klemchuk, P. P. Investigation of Thermal-Oxidation and Stabilization of High-Density Polyethylene. ACS Symp. Ser. 1985, 280, 235-246.

111. Allen, N. S. Fundamentals of Polymer Degradation and Stabilization; Springer: Barking, Essex, UK, 1992; pp 1-201.

112. Abdou, M. S. A. Photochemical Processes of Conductive Polymers. In Handbook of Organic Conductive Molecules and Polymers, Volume 4; H. Singh Nalwa, Ed.; John Wiley & Sons Ltd.: West Sussex, UK, 1997; pp 173-232.

113. Jaffe, H. H.; Miller, A. L. Fates of Electronic Excitation Energy. J. Chem. Educ. 1966, 43, 469.

114. Priestley, E. B.; Haug, A. Phosphorescence Spectrum of Pure Crystalline Naphthalene. J. Chem. Phys. 1968, 49, 622-629.

115. Jablonski Diagram. http://chemwiki.ucdavis.edu/Physical_Chemistry/Spectroscopy/Electronic_Spectroscopy/Jablonski_diagram (accessed Dec 6, 2013).

116. Jablonski Energy Diagram. http://www.olympusmicro.com/primer/java/jablonski/jabintro/ (accessed Dec 6, 2013).

117. Guillet, J. Polymer Photophysics and Photochemistry: An Introduction to the Study of Photoprocesses in Macromolecules; Cambridge University Press: Cambridge, UK, 1987; pp 1-391.

118. Phillips, D. Polymer Photophysics; Springer: New York, NY, USA, 1985; pp 1-437.

119. van Krevelen, D. W.; te Nijenhuis, K. Properties of Polymers; Elsevier: Amsterdam, The Netherlands, 2009; pp 1-953.

120. Hofmann, A. W. VIII.-Miscellaneous Observations. Quarterly Journal of the Chemical Society of London. 1861, 13, 51-90.

121. Bolland, J. L.; Gee, G. Kinetic Studies in the Chemistry of Rubber and Related Materials. II. The Kinetics of Oxidation of Unconjugated Olefins. Trans. Faraday Soc. 1946, 42, 236-243.

122. Bolland, J. L. Kinetics of Olefin Oxidation. Quarterly Reviews, Chemical Society. 1949, 3, 1-21.

123. Bolland, J. L. Kinetic Studies in the Chemistry of Rubber and Related Materials. VI. The Benzoyl Peroxide-Catalysed Oxidation of Ethyl Linoleate. Trans. Faraday Soc. 1948, 44, 669-677.

124. Pospisil, J.; Klemchuk, P. P. Oxidative Deterioration Process in Organic Materials. In Oxidation Inhibition in Organic Materials, Volume 1; J.

111

Pospisil and P.P. Klemchuk, Eds.; CRC Press: Boca Raton, FL, USA, 1990; pp 1-9.

125. Schnabel, W. Polymers and Light; Wiley-VCH: Weinheim, Germany, 2007; pp 1-365.

126. Chirinos-Padron, A. J.; Allen, S. A. Aspects of Polymer Stabilization. In Handbook of Polymer Degradation; H.S. Hamid, M.B. Amin, and A.G. Maadhah, Eds.; Marcel Dekker, Inc.: New York, NY, USA, 1992; pp 261-303.

127. Rabek, J. F. Photodegradation of Polymers; Springer Verlag: Berlin, Germany, 1996; pp 1-208.

128. Allen, N. S.; Chirinospadron, A.; Henman, T. J. Photoinitiated Oxidation of Polypropylene - a Review. Prog. Org. Coat. 1985, 13, 97-122.

129. Chirinos Padrón, A. J. Mechanistic Aspects of Polymer Photostabilization. Journal of Photochemistry and Photobiology A: Chemistry. 1989, 49, 1-39.

130. Carlsson, D. J.; Wiles, D. M. Photooxidative Degradation of Polypropylene. I. Photooxidation and Photoinitiation Processes. Journal of Macromolecular Science-Reviews in Macromolecular Chemistry and Physics. 1976, C 14, 65-106.

131. Carlsson, D. J.; Bazan, G.; Chmela, S.; Wiles, D. M.; Russell, K. E. Oxidation of Solid Polyethylene Films - Effects of Backbone Branching. Polym. Degrad. Stab. 1987, 19, 195-206.

132. Allen, N. S. Photo-Degradation and Photo-Oxidation of Polyolefins: Importance of Oxygen-Polymer Charge Transfer Complexes. Polym. Degrad. Stab. 1980, 2, 155-161.

133. MacCallum, J. R.; Ramsay, D. A. The Photo-Oxidation of Styrene-Type Copolymers—I. Eur. Polym. J. 1977, 13, 945-948.

134. Rabek, J. F.; Ranby, B. Studies on Photooxidation Mechanism of Polymers. I. Photolysis and Photooxidation of Polystyrene. J. Polym. Sci., Part A: Polym. Chem. 1974, 12, 273-294.

135. Chien, J. C. W. On Possible Initiation of Photooxidation by Charge-Transfer Excitation. J. Phys. Chem. 1965, 69, 4317-4325.

136. Tsuji, K.; Seiki, T. Observation of Absorption Spectra Due to Charge Transfer Complexes of Polymers with Oxygen and Its Possible Contribution to Radical Formical Formation in Polymers by Ultraviolet Irradiation. Journal of Polymer Science Part B-Polymer Letters. 1970, 8, 817-821.

137. Benson, S. W. Kinetics of Pyrolysis of Alkyl Hydroperoxides + Their O-O Bond Dissociation Energies. J. Chem. Phys. 1964, 40, 1007-1013.

138. Benson, S. W. Bond Energies. J. Chem. Educ. 1965, 42, 502. 139. Lloyd, R. A. Low Level Chemiluminescence from Hydrobarbon Autoxidation

Reactions. II. Thermal Decomposition of Benzoyl Peroxide Cumene Hydroperoxide and U.V. Irradiated Solvents. Trans. Faraday Soc. 1965, 61, 2182-2193.

140. Reich, L.; Stivala, S. S. Some Remarks on Origin of Weak Chemiluminescence During Polymer Oxidation. Makromolekulare Chemie. 1967, 103, 74-82.

141. Vassil'ev, R. F. On the Mechanism of Chemiluminescence in Oxidation of Organic Substances and Polymers. Makromolekulare Chemie. 1969, 126, 231-238.

112

142. Gijsman, P.; Hennekens, J.; Vincent, J. The Mechanism of the Low-Temperature Oxidation of Polypropylene. Polym. Degrad. Stab. 1993, 42, 95-105.

143. Allen, N. S.; Fatinikun, K. O.; Henman, T. J. Thermal and Photochemical Oxidation of Polypropylene. Influence of Residual Catalyst Levels in Unstabilised Diluent and Gas Phase Polymers. Eur. Polym. J. 1983, 19, 551-554.

144. Allen, N. S. Recent Advances in the Photooxidation and Stabilization of Polymers. Chem. Soc. Rev. 1986, 15, 373-404.

145. McCallum, J. R. Kinetics and Mechanisms of Photo-Oxidation Processes. In Developments in Polymer Degradation, Volume 6; N. Grassie, Ed.; Elsevier Applied Science Publishers: Essex, UK, 1985; pp 191-205.

146. Gugumus, F. Stabilization of Plastics against Thermal Oxidation. In Oxidation Inhibition in Organic Materials, Volume 1; J. Pospisil and P.P. Klemchuk, Eds.; CRC Press, Inc.: Boca Raton, FL, USA, 1990; pp 61-173.

147. Chirinos-Padrón, A. J.; Hernández, P. H.; Chávez, E.; Allen, N. S.; Vasiliou, C.; DePoortere, M. Influences of Unsaturation and Metal Impurities on the Oxidative Degradation of High Density Polyethylene. Eur. Polym. J. 1987, 23, 935-940.

148. Wong, W. K.; Schupp, H.; Schnabel, W. Kinetic-Mechanistic Studies on the Photorearrangement of Ortho-Nitrobenzyl Ester Groups in Polymer Matrices. Macromolecules. 1989, 22, 2176-2181.

149. Feldmann, G.; Winsauer, A.; Pfleger, J.; Schnabel, W. Kinetic-Studies Concerning the Decay of Nitronic Acids in Polymer Matrices Using a Data Evaluation Method Based on Dispersive Kinetics. Macromolecules. 1994, 27, 4391-4396.

150. Turro, N. J. Modern Molecular Photochemistry; University Science Books: Sausalito, CA, USA, 1991; pp 1-615.

151. Rånby, B. G.; Rabek, J. F. Singlet Oxygen Reactions with Organic Compounds and Polymers; Wiley-Interscience: New York, NY, USA, 1978; pp 1-342.

152. Sheldrick, G. E.; Vogl, O. Induced Photodegradation of Styrene Polymers - Survey. Polym. Eng. Sci. 1976, 16, 65-73.

153. Golub, M. A. Photosensitized Oxidation of Unsaturated Polymers. Pure Appl. Chem. 1980, 52, 305-323.

154. Rabek, J. F.; Lucki, J.; Ranby, B. Comparative Studies of Reactions of Commercial Polymers with Molecular-Oxygen, Singlet Oxygen, Atomic Oxygen and Ozone. I. Reactions with Cis-1,4-Polybutadiene. Eur. Polym. J. 1979, 15, 1089-1100.

155. Rabek, J. F.; Ranby, B. Studies on Photooxidation Mechanism of Polymers. V. Oxidation of Polybutadienes by Singlet Oxygen from Microwave-Discharge and in Dye-Photosensitized Reactions. J. Polym. Sci., Part A: Polym. Chem. 1976, 14, 1463-1473.

156. Rabek, J. F.; Ranby, B.; Arct, J. Stabilization of Cis-1,4-Polybutadiene against Photo-Oxidation and Singlet Oxygen Oxidation by Hindered Phenols. Polym. Degrad. Stab. 1983, 5, 65-77.

157. Wilkinson, F.; Helman, W. P.; Ross, A. B. Quantum Yields for the Photosensitized Formation of the Lowest Electronically Excited Singlet-

113

State of Molecular-Oxygen in Solution. J. Phys. Chem. Ref. Data. 1993, 22, 113-262.

158. Ogilby, P. R.; Kristiansen, M.; Martire, D. O.; Scurlock, R. D.; Taylor, V. L.; Clough, R. L. Formation and Removal of Singlet Oxygen in Bulk Polymers: Events That May Influence Photodegradation. In Polymer Durability; R.L. Clough, N.C. Billingham, and K.T. Gillen, Eds.; American Chemical Society: Washington, DC, USA, 1993; pp 113-126.

159. Scurlock, R. D.; Ogilby, P. R. Singlet Molecular-Oxygen Formation Upon Irradiation of an Oxygen Organic-Molecule Charge-Transfer Absortion-Band. J. Phys. Chem. 1989, 93, 5493-5500.

160. Ogilby, P. R.; Kristiansen, M.; Clough, R. L. Singlet Oxygen Formation in a Solid Organic Polymer Upon Irradiation of the Oxygen Polymer Charge-Transfer Band. Macromolecules. 1990, 23, 2698-2704.

161. Kristiansen, M.; Scurlock, R. D.; Iu, K. K.; Ogilby, P. R. Charge-Transfer State and Singlet Oxygen Production in Photoexcited Organic-Molecule Molecular-Oxygen Complexes. J. Phys. Chem. 1991, 95, 5190-5197.

162. Singh, A. Interconversion of Singlet Oxygen and Related Species - Introduction. Photochem. Photobiol. 1978, 28, 429-433.

163. Scurlock, R. D.; Wang, B. J.; Ogilby, P. R.; Sheats, J. R.; Clough, R. L. Singlet Oxygen as a Reactive Intermediate in the Photodegradation of an Electroluminescent Polymer. J. Am. Chem. Soc. 1995, 117, 10194-10202.

164. Dam, N.; Scurlock, R. D.; Wang, B. J.; Ma, L. C.; Sundahl, M.; Ogilby, P. R. Singlet Oxygen as a Reactive Intermediate in the Photodegradation of Phenylenevinylene Oligomers. Chem. Mater. 1999, 11, 1302-1305.

165. Cumpston, B. H.; Jensen, K. F. Photo-Oxidation of Polymers Used in Electroluminescent Devices. Synt. Met. 1995, 73, 195-199.

166. Henari, F. Z.; Manaa, H.; Kretsch, K. P.; Blau, W. J.; Rost, H.; Pfeiffer, S.; Teuschel, A.; Tillmann, H.; Hörhold, H. H. Effective Stimulated Emission and Excited State Absorption Measurements in the Phenylene–Vinylene Oligomer (1,4-Bis-(Alpha-Cyanostyryl)-2,5-Dimethoxybenzene)). Chem. Phys. Lett. 1999, 307, 163-166.

167. Colaneri, N. F.; Bradley, D. D. C.; Friend, R. H.; Burn, P. L.; Holmes, A. B.; Spangler, C. W. Photoexcited States in Poly(Para-Phenylene Vinylene) - Comparison with Trans,Trans-Distyrylbenzene, a Model Oligomer. Phys. Rev. B. 1990, 42, 11670-11681.

168. Partee, J.; Frankevich, E. L.; Uhlhorn, B.; Shinar, J.; Ding, Y.; Barton, T. J. Delayed Fluorescence and Triplet-Triplet Annihilation in Pi-Conjugated Polymers. Phys. Rev. Lett. 1999, 82, 3673-3676.

169. Ma, L. C.; Wang, X. S.; Wang, B. J.; Chen, J. R.; Wang, J. H.; Huang, K.; Zhang, B. W.; Cao, Y.; Han, Z. H.; Qian, S. P.; Yao, S. D. Photooxidative Degradation Mechanism of Model Compounds of Poly(P-Phenylenevinylenes) PPVs. Chem. Phys. 2002, 285, 85-94.

170. Chambon, S.; Rivaton, A.; Gardette, J.-L.; Firon, M.; Lutsen, L. Aging of a Donor Conjugated Polymer: Photochemical Studies of the Degradation of Poly 2-Methoxy-5-(3 ',7 '-Dimethyloctyloxy)-1,4-Phenylenevinylene. J. Polym. Sci., Part A: Polym. Chem. 2007, 45, 317-331.

114

171. Chambon, S.; Rivaton, A.; Gardette, J.-L.; Firon, M. Reactive Intermediates in the Initiation Step of the Photo-Oxidation of MDMO-PPV. J. Polym. Sci., Part A: Polym. Chem. 2009, 47, 6044-6052.

172. Abdou, M. S. A.; Holdcroft, S. Mechanisms of Photodegradation of Poly(3-Alkylthiophenes) in Solution. Macromolecules. 1993, 26, 2954-2962.

173. Abdou, M. S. A.; Holdcroft, S. Solid-State Photochemistry of Pi-Conjugated Poly(3-Alkylthiophenes). Can. J. Chem. 1995, 73, 1893-1901.

174. Holdcroft, S. A Photochemical Study of Poly(3-Hexylthiophene). Macromolecules. 1991, 24, 4834-4838.

175. Koch, M.; Nicolaescu, R.; Kamat, P. V. Photodegradation of Polythiophene-Based Polymers: Excited State Properties and Radical Intermediates. Journal of Physical Chemistry C. 2009, 113, 11507-11513.

176. Song, X.; Fanelli, M. G.; Cook, J. M.; Bai, F.; Parish, C. A. Mechanisms for the Reaction of Thiophene and Methylthiophene with Singlet and Triplet Molecular Oxygen. J. Phys. Chem. A. 2012, 116, 4934-4946.

177. de Melo, J. S.; Burrows, H. D.; Svensson, M.; Andersson, M. R.; Monkman, A. P. Photophysics of Thiophene Based Polymers in Solution: The Role of Nonradiative Decay Processes. J. Chem. Phys. 2003, 118, 1550-1556.

178. Osterbacka, R.; An, C. P.; Jiang, X. M.; Vardeny, Z. V. Two-Dimensional Electronic Excitations in Self-Assembled Conjugated Polymer Nanocrystals. Science. 2000, 287, 839-842.

179. Oelkrug, D.; Egelhaaf, H. J.; Worrall, D. R.; Wilkinson, F. Fluorescence Yields and Molecular Orientation of Thin Organic Films: Vapor-Deposited Oligothiophenes Αlpha3t–Αlpha8t. J Fluoresc. 1995, 5, 165-170.

180. Manceau, M.; Rivaton, A.; Gardette, J.-L.; Guillerez, S.; Lemaitre, N. The Mechanism of Photo- and Thermooxidation of Poly(3-Hexylthiophene) (P3HT) Reconsidered. Polym. Degrad. Stab. 2009, 94, 898-907.

181. Manceau, M.; Rivaton, A.; Gardette, J.-L. Involvement of Singlet Oxygen in the Solid-State Photochemistry of P3HT. Macromol. Rapid Commun. 2008, 29, 1823-1827.

182. Hintz, H.; Egelhaaf, H. J.; Peisert, H.; Chasse, T. Photo-Oxidation and Ozonization of Poly(3-Hexylthiophene) Thin Films as Studied by UV/Vis and Photoelectron Spectroscopy. Polym. Degrad. Stab. 2010, 95, 818-825.

183. Hintz, H.; Sessler, C.; Peisert, H.; Egelhaaf, H. J.; Chasse, T. Wavelength-Dependent Pathways of Poly-3-Hexylthiophene Photo-Oxidation. Chem. Mater. 2012, 24, 2739-2743.

184. Saunders, B. B.; Kaufman, P. C.; Matheson, M. S. Reactions of Thiophene with Radiolytically Produced Radicals. I. Hydroxyl Radical. J. Phys. Chem. 1978, 82, 142-150.

185. Ljungqvist, N.; Hjertberg, T. Oxidative-Degradation of Poly(3-Octylthiophene). Macromolecules. 1995, 28, 5993-5999.

186. Abdou, M. S. A.; Orfino, F. P.; Xie, Z. W.; Deen, M. J.; Holdcroft, S. Reversible Charge-Transfer Complexes between Molecular-Oxygen and Poly(3-Alkylthiophene)S. Adv. Mater. 1994, 6, 838-841.

187. Seemann, A.; Sauermann, T.; Lungenschmied, C.; Armbruster, O.; Bauer, S.; Egelhaaf, H. J.; Hauch, J. Reversible and Irreversible Degradation of

115

Organic Solar Cell Performance by Oxygen. Solar Energy. 2011, 85, 1238-1249.

188. Hintz, H.; Peisert, H.; Egelhaaf, H. J.; Chasse, T. Reversible and Irreversible Light-Induced P-Doping of P3HT by Oxygen Studied by Photoelectron Spectroscopy (XPS/UPS). Journal of Physical Chemistry C. 2011, 115, 13373-13376.

189. Liao, H.-H.; Yang, C.-M.; Liu, C.-C.; Horng, S.-F.; Meng, H.-F.; Shy, J.-T. Dynamics and Reversibility of Oxygen Doping and De-Doping for Conjugated Polymer. J. Appl. Phys. 2008, 103.

190. Schafferhans, J.; Baumann, A.; Deibel, C.; Dyakonov, V. Trap Distribution and the Impact of Oxygen-Induced Traps on the Charge Transport in Poly(3-Hexylthiophene). Appl. Phys. Lett. 2008, 93, 093303.

191. Schafferhans, J.; Baumann, A.; Wagenpfahl, A.; Deibel, C.; Dyakonov, V. Oxygen Doping of P3HT:PCBM Blends: Influence on Trap States, Charge Carrier Mobility and Solar Cell Performance. Org. Electron. 2010, 11, 1693-1700.

192. Sperlich, A.; Kraus, H.; Deibel, C.; Blok, H.; Schmidt, J.; Dyakonov, V. Reversible and Irreversible Interactions of Poly(3-Hexylthiophene) with Oxygen Studied by Spin-Sensitive Methods. J. Phys. Chem. B. 2011, 115, 13513-13518.

193. Abdou, M. S. A.; Orfino, F. P.; Son, Y.; Holdcroft, S. Interaction of Oxygen with Conjugated Polymers: Charge Transfer Complex Formation with Poly(3-Alkylthiophenes). J. Am. Chem. Soc. 1997, 119, 4518-4524.

194. Weller, A. Singlet- and Triplet-State Exciplexes. In The Exciplex; M. Gordon and W.R. Ware, Eds.; Academic Press: New York, NY, USA, 1975; pp 23-36.

195. Rehm, D.; Weller, A. Bond Status and Fluorescent Spectra of Hetero-Excimers. Z. Phys. Chem. 1970, 69, 183.

196. Rehm, D. Spectroscopic Study of Heteroexcimers with Heterocyclic Aromatic Hydrocarbons as Electron Acceptors. Z. Naturforsch., A: Phys. Sci. 1970, A 25, 1442.

197. Liu, C. C.; Yang, C. M.; Liu, W. H.; Liao, H. H.; Horng, S. F.; Meng, H. F. Interface Effect of Oxygen Doping in Polythiophene. Synt. Met. 2009, 159, 1131-1134.

198. Zhuo, J.-M.; Zhao, L.-H.; Png, R.-Q.; Wong, L.-Y.; Chia, P.-J.; Tang, J.-C.; Sivaramakrishnan, S.; Zhou, M.; Ou, E. C. W.; Chua, S.-J.; Sim, W.-S.; Chua, L.-L.; Ho, P. K. H. Direct Spectroscopic Evidence for a Photodoping Mechanism in Polythiophene and Poly(Bithiophene-Alt-Thienothiophene) Organic Semiconductor Thin Films Involving Oxygen and Sorbed Moisture. Adv. Mater. 2009, 21, 4747-4752.

199. Lu, C.-K.; Meng, H.-F. Hole Doping by Molecular Oxygen in Organic Semiconductors: Band-Structure Calculations. Phys. Rev. B. 2007, 75.

200. Luer, L.; Egelhaaf, H. J.; Oelkrug, D.; Cerullo, G.; Lanzani, G.; Huisman, B. H.; de Leeuw, D. Oxygen-Induced Quenching of Photoexcited States in Polythiophene Films. Org. Electron. 2004, 5, 83-89.

201. Kawano, K.; Adachi, C. Evaluating Carrier Accumulation in Degraded Bulk Heterojunction Organic Solar Cells by a Thermally Stimulated Current Technique. Adv. Funct. Mater. 2009, 19, 3934-3940.

116

202. Aygul, U.; Hintz, H.; Egelhaaf, H. J.; Distler, A.; Abb, S.; Peisert, H.; Chasse, T. Energy Level Alignment of a P3HT/Fullerene Blend During the Initial Steps of Degradation. Journal of Physical Chemistry C. 2013, 117, 4992-4998.

203. Tapponnier, A.; Biaggio, I.; Gunter, P. Ultrapure C-60 Field-Effect Transistors and the Effects of Oxygen Exposure. Appl. Phys. Lett. 2005, 86.

204. Hamed, A.; Sun, Y. Y.; Tao, Y. K.; Meng, R. L.; Hor, P. H. Effects of Oxygen and Illumination on the Insitu Conductivity of C-60 Thin-Films. Phys. Rev. B. 1993, 47, 10873-10880.

205. Katz, E. A.; Faiman, D.; Mishori, B.; Shapira, Y.; Shames, A. I.; Shtutina, S.; Goren, S. Changes in the Photoelectrical Properties and Generation of Photoinduced Defects under Light/Air Exposure of C-60 Thin Films. J. Appl. Phys. 1998, 84, 3333-3337.

206. Habuchi, H.; Nitta, S.; Han, D. X.; Nonomura, S. Localized Electronic States Related to O-2 Intercalation and Photoirradiation on C-60 Films and C-70 Films. J. Appl. Phys. 2000, 87, 8580-8588.

207. Eklund, P. C.; Rao, A. M.; Zhou, P.; Wang, Y. Photochemical Transformation of C-60 and C-70 Films. Thin Solid Films. 1995, 257, 185-203.

208. Creegan, K. M.; Robbins, J. L.; Robbins, W. K.; Millar, J. M.; Sherwood, R. D.; Tindall, P. J.; Cox, D. M.; Smith, A. B.; McCauley, J. P.; Jones, D. R.; Gallagher, R. T. Synthesis and Characterization of (C60)O, the 1st Fullerene Epoxide. J. Am. Chem. Soc. 1992, 114, 1103-1105.

209. Lessmann, R.; Hong, Z.; Scholz, S.; Maennig, B.; Riede, M. K.; Leo, K. Aging of Flat Heterojunction Zinc Phthalocyanine/Fullerene C-60 Organic Solar Cells. Org. Electron. 2010, 11, 539-543.

210. Wohlers, M.; Werner, H.; Herein, D.; SchedelNiedrig, T.; Bauer, A. Reaction of C-60 and C-70 with Molecular Oxygen. Synt. Met. 1996, 77, 299-302.

211. Matsuo, Y.; Ozu, A.; Obata, N.; Fukuda, N.; Tanaka, H.; Nakamura, E. Deterioration of Bulk Heterojunction Organic Photovoltaic Devices by a Minute Amount of Oxidized Fullerene. Chem. Commun. 2012, 48, 3878-3880.

212. Konenkamp, R.; Priebe, G.; Pietzak, B. Carrier Mobilities and Influence of Oxygen in C-60 Films. Phys. Rev. B. 1999, 60, 11804-11808.

213. Rivaton, A.; Chambon, S.; Manceau, M.; Gardette, J.-L.; Lamaitre, N.; Guillerez, S. Light-Induced Degradation of the Active Layer of Polymer-Based Solar Cells. Polym. Degrad. Stab. 2010, 95, 278-284.

214. Manceau, M.; Chambon, S.; Rivaton, A.; Gardette, J.-L.; Guillerez, S.; Lemaitre, N. Effects of Long-Term UV-Visible Light Irradiation in the Absence of Oxygen on P3HT and P3HT:PCBM Blend. Sol. Energ. Mat. Sol. Cells. 2010, 94, 1572-1577.

215. Sarkas, H. W.; Kwan, W.; Flom, S. R.; Merritt, C. D.; Kafafi, Z. H. Enhanced Photooxidative Stability of Conjugated Polymers Via C-60 Doping. J. Phys. Chem. 1996, 100, 5169-5171.

216. Brabec, C. J.; Zerza, G.; Cerullo, G.; De Silvestri, S.; Luzzati, S.; Hummelen, J. C.; Sariciftci, S. Tracing Photoinduced Electron Transfer

117

Process in Conjugated Polymer/Fullerene Bulk Heterojunctions in Real Time. Chem. Phys. Lett. 2001, 340, 232-236.

217. Tromholt, T.; Madsen, M. V.; Carle, J. E.; Helgesen, M.; Krebs, F. C. Photochemical Stability of Conjugated Polymers, Electron Acceptors and Blends for Polymer Solar Cells Resolved in Terms of Film Thickness and Absorbance. J. Mater. Chem. 2012, 22, 7592-7601.

218. Distler, A.; Kutka, P.; Sauermann, T.; Egelhaaf, H. J.; Guldi, D. M.; Di Nuzzo, D.; Meskers, S. C. J.; Janssen, R. A. J. Effect of PCBM on the Photodegradation Kinetics of Polymers for Organic Photovoltaics. Chem. Mater. 2012, 24, 4397-4405.

219. Soon, Y. W.; Cho, H.; Low, J.; Bronstein, H.; McCulloch, I.; Durrant, J. R. Correlating Triplet Yield, Singlet Oxygen Generation and Photochemical Stability in Polymer/Fullerene Blend Films. Chem. Commun. 2013, 49, 1291-1293.

220. Morton, J. R.; Negri, F.; Preston, K. F. Addition of Free Radicals to C60. Acc. Chem. Res. 1998, 31, 63-69.

221. Kahnt, A.; Guldi, D. M.; Brettreich, M.; Hartnagel, U.; Hirsch, A. Reaction of Water-Soluble Fullerenes with O2 Radical- and Other Reactive Radical Species. J. Mater. Chem. 2010, 20, 83-89.

222. Yin, J.-J.; Lao, F.; Fu, P. P.; Wamer, W. G.; Zhao, Y.; Wang, P. C.; Qiu, Y.; Sun, B.; Xing, G.; Dong, J.; Liang, X.-J.; Chen, C. The Scavenging of Reactive Oxygen Species and the Potential for Cell Protection by Functionalized Fullerene Materials. Biomaterials. 2009, 30, 611-621.

223. Chen, Y.; Lin, K. C. Radical Polymerization of Styrene in the Presence of C-60. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 2969-2975.

224. Cook, S.; Ohkita, H.; Kim, Y.; Benson-Smith, J. J.; Bradley, D. D. C.; Durrant, J. R. A Photophysical Study of PCBM Thin Films. Chem. Phys. Lett. 2007, 445, 276-280.

225. Prat, F.; Stackow, R.; Bernstein, R.; Qian, W. Y.; Rubin, Y.; Foote, C. S. Triplet-State Properties and Singlet Oxygen Generation in a Homologous Series of Functionalized Fullerene Derivatives. J. Phys. Chem. A. 1999, 103, 7230-7235.

226. Salazar, F. A.; Fedorov, A.; BerberanSantos, M. N. A Study of Thermally Activated Delayed Fluorescence in C-60. Chem. Phys. Lett. 1997, 271, 361-366.

227. Wang, H.; He, Y.; Li, Y.; Su, H. Photophysical and Electronic Properties of Five PCBM-Like C-60 Derivatives: Spectral and Quantum Chemical View. J. Phys. Chem. A. 2012, 116, 255-262.

228. Arbogast, J. W.; Darmanyan, A. P.; Foote, C. S.; Rubin, Y.; Diederich, F. N.; Alvarez, M. M.; Anz, S. J.; Whetten, R. L. Photophysical Properties of C60. J. Phys. Chem. 1991, 95, 11-12.

229. Yamakoshi, Y.; Umezawa, N.; Ryu, A.; Arakane, K.; Miyata, N.; Goda, Y.; Masumizu, T.; Nagano, T. Active Oxygen Species Generated from Photoexcited Fullerene (C-60) as Potential Medicines: O-2 Radical Versus O-1(2). J. Am. Chem. Soc. 2003, 125, 12803-12809.

230. Hoke, E. T.; Sachs-Quintana, I. T.; Lloyd, M. T.; Kauvar, I.; Mateker, W. R.; Nardes, A. M.; Peters, C. H.; Kopidakis, N.; McGehee, M. D. The Role of Electron Affinity in Determining Whether Fullerenes Catalyze or Inhibit

118

Photooxidation of Polymers for Solar Cells. Adv. Energy Mater. 2012, 2, 1351-1357.

231. Koehler, A.; Baessler, H. Triplet States in Organic Semiconductors. Materials Science & Engineering R-Reports. 2009, 66, 71-109.

232. Veldman, D.; Ipek, O.; Meskers, S. C. J.; Sweelssen, J.; Koetse, M. M.; Veenstra, S. C.; Kroon, J. M.; van Bavel, S. S.; Loos, J.; Janssen, R. A. J. Compositional and Electric Field Dependence of the Dissociation of Charge Transfer Excitons in Alternating Polyfluorene Copolymer/Fullerene Blends. J. Am. Chem. Soc. 2008, 130, 7721-7735.

233. Veldman, D.; Meskers, S. C. J.; Janssen, R. A. J. The Energy of Charge-Transfer States in Electron Donor-Acceptor Blends: Insight into the Energy Losses in Organic Solar Cells. Adv. Funct. Mater. 2009, 19, 1939-1948.

234. Liedtke, M.; Sperlich, A.; Kraus, H.; Baumann, A.; Deibel, C.; Wirix, M. J. M.; Loos, J.; Cardona, C. M.; Dyakonov, V. Triplet Exciton Generation in Bulk-Heterojunction Solar Cells Based on Endohedral Fullerenes. J. Am. Chem. Soc. 2011, 133, 9088-9094.

235. Di Nuzzo, D.; Aguirre, A.; Shahid, M.; Gevaerts, V. S.; Meskers, S. C. J.; Janssen, R. A. J. Improved Film Morphology Reduces Charge Carrier Recombination into the Triplet Excited State in a Small Bandgap Polymer-Fullerene Photovoltaic Cell. Adv. Mater. 2010, 22, 4321-4324.

236. Ford, T. A.; Avilov, I.; Beljonne, D.; Greenham, N. C. Enhanced Triplet Exciton Generation in Polyfluorene Blends. Phys. Rev. B. 2005, 71.

237. Offermans, T.; van Hal, P. A.; Meskers, S. C. J.; Koetse, M. M.; Janssen, R. A. J. Exciplex Dynamics in a Blend of Pi-Conjugated Polymers with Electron Donating and Accepting Properties: MDMO-PPV and PCNEPV. Phys. Rev. B. 2005, 72.

238. Kawaoka, K.; Khan, A. U.; Kearns, D. R. Role of Singlet Excited States of Molecular Oxygen in Quenching of Organic Triplet States. J. Chem. Phys. 1967, 46, 1842-1853.

239. Yu, G.; Zhang, C.; Heeger, A. J. Dual-Function Semiconducting Polymer Devices - Light-Emitting and Photodetecting Diodes. Appl. Phys. Lett. 1994, 64, 1540-1542.

240. Wuerfel, P. Physics of Solar Cells: From Principles to New Concepts; Wiley: Weinheim, Germany, 2008; pp 1-241.

241. Riedel, I.; Parisi, J.; Dyakonov, V.; Lutsen, L.; Vanderzande, D.; Hummelen, J. C. Effect of Temperature and Illumination on the Electrical Characteristics of Polymer-Fullerene Bulk-Heterojunction Solar Cells. Adv. Funct. Mater. 2004, 14, 38-44.

242. Deibel, C.; Dyakonov, V. Polymer-Fullerene Bulk Heterojunction Solar Cells. Rep. Prog. Phys. 2010, 73.

243. Blom, P. W. M.; Mihailetchi, V. D.; Koster, L. J. A.; Markov, D. E. Device Physics of Polymer : Fullerene Bulk Heterojunction Solar Cells. Adv. Mater. 2007, 19, 1551-1566.

244. ISOS-2. http://isos-2.wikispaces.com/ (accessed Jun 3, 2013). 245. ISOS-1. http://isos-1.wikispaces.com/ (accessed Jun 3, 2013). 246. Griffiths, P.; De Haseth, J. A. Fourier Transform Infrared Spectrometry;

John Wiley & Sons: Hoboken, NJ, USA, 1986; pp 1-509.

119

247. Banwell, C. N.; McCash, E. M. Fundamentals of Molecular Spectroscopy; McGraw-Hill: Berkshire, UK, 1994; pp 1-308.

248. Crews, P.; Rodríguez, J.; Jaspars, M. Organic Structure Analysis; Oxford University Press, Inc.: New York, NY, US, 1998; pp 1-552.

249. Foster, G. N. Analytical Methods Applied to the Testing of Oxidation Inhibition. In Oxidation Inhibition in Organic Materials, Volume 2; J. Pospisil and P.P. Klemchuk, Eds.; CRC Press: Boca Raton, FL, USA, 1990; pp 299-346.

250. Koenig, J. L. Raman Scattering of Synthetic Polymers‐a Review. Applied Spectroscopy Reviews. 1971, 4, 233-305.

251. Siesler, H. W.; Holland-Moritz, K. Infrared and Raman Spectroscopy of Polymers; Marcel Dekker: New York, NY, USA, 1980; pp 1-389.

252. Haslam, J.; Willis, H. A.; Squirrell, D. C. M. Identification and Analysis of Plastics; Wiley: Hoboken, NJ, USA, 1972; pp 1-756.

253. Bellamy, L. J. The Infrared Spectra of Complex Molecules; Methuen: London, UK, 1966; pp 1-425.

254. Gosser, D. K. Cyclic Voltammetry: Simulation and Analysis of Reaction Mechanisms; VCH: New York, NY, USA, 1993; pp 1-154.

255. Kaifer, A. E.; Gomez-Kaifer, M. Supramolecular Electrochemistry; Wiley: Toronto, Canada, 1999; pp 1-233.

256. Cyclic Voltammetry. http://chemwiki.ucdavis.edu/Analytical_Chemistry/Instrumental_Analysis/Cyclic_Voltammetry (accessed Dec 8, 2013).

257. Kissinger, P. T.; Heineman, W. R. Cyclic Voltammetry. J. Chem. Educ. 1983, 60, 702.

258. Heinze, J. Cyclic Voltammetry - Electrochemical Spectroscopy. Angewandte Chemie-International Edition. 1984, 23, 831-847.

259. Kucur, E.; Buecking, W.; Nann, T. Electrochemical Determination of Mesoscopic Phenomena, Defect States in Cdse Nanocrystals and Charge Carrier Manipulability. Microchimica Acta. 2008, 160, 299-308.

260. Janietz, S.; Bradley, D. D. C.; Grell, M.; Giebeler, C.; Inbasekaran, M.; Woo, E. P. Electrochemical Determination of the Ionization Potential and Electron Affinity of Poly(9,9-Dioctylfluorene). Appl. Phys. Lett. 1998, 73, 2453-2455.

261. Johansson, T.; Mammo, W.; Svensson, M.; Andersson, M. R.; Inganas, O. Electrochemical Bandgaps of Substituted Polythiophenes. J. Mater. Chem. 2003, 13, 1316-1323.

262. Yang, X.; Loos, J. Toward High-Performance Polymer Solar Cells: The Importance of Morphology Control. Macromolecules. 2007, 40, 1353-1362.

263. Zhao, J.; Swinnen, A.; Van Assche, G.; Manca, J.; Vanderzande, D.; Van Mele, B. Phase Diagram of P3HT/PCBM Blends and Its Implication for the Stability of Morphology. J. Phys. Chem. B. 2009, 113, 1587-1591.

264. Bertho, S.; Janssen, G.; Cleij, T. J.; Conings, B.; Moons, W.; Gadisa, A.; D'Haen, J.; Goovaerts, E.; Lutsen, L.; Manca, J.; Vanderzande, D. Effect of Temperature on the Morphological and Photovoltaic Stability of Bulk Heterojunction Polymer:Fullerene Solar Cells. Sol. Energ. Mat. Sol. Cells. 2008, 92, 753-760.

120

265. Nicholson, P. G.; Castro, F. A. Organic Photovoltaics: Principles and Techniques for Nanometre Scale Characterization. Nanotechnology. 2010, 21.

266. Hoppe, H.; Sariciftci, N. S. Polymer Solar Cells. In Photoresponsive Polymers II; S.R. Marder and K.S. Lee, Eds., 2008; pp 1-86.

267. Burke, K. B.; Belcher, W. J.; Thomsen, L.; Watts, B.; McNeill, C. R.; Ade, H.; Dastoor, P. C. Role of Solvent Trapping Effects in Determining the Structure and Morphology of Ternary Blend Organic Devices. Macromolecules. 2009, 42, 3098-3103.

268. Ade, H.; Zhang, X.; Cameron, S.; Costello, C.; Kirz, J.; Williams, S. Chemical Contrast in X-Ray Microscopy and Spatially Resolved XANES Spectroscopy of Organic Specimens. Science. 1992, 258, 972-975.

269. McNeill, C. R.; Watts, B.; Thomsen, L.; Belcher, W. J.; Kilcoyne, A. L. D.; Greenham, N. C.; Dastoor, P. C. X-Ray Spectromicroscopy of Polymer/Fullerene Composites: Quantitative Chemical Mapping. Small. 2006, 2, 1432-1435.

270. Hemraj-Benny, T.; Banerjee, S.; Sambasivan, S.; Balasubramanian, M.; Fischer, D. A.; Eres, G.; Puretzky, A. A.; Geohegan, D. B.; Lowndes, D. H.; Han, W. Q.; Misewich, J. A.; Wong, S. S. Near-Edge X-Ray Absorption Fine Structure Spectroscopy as a Tool for Investigating Nanomaterials. Small. 2006, 2, 26-35.

271. McNeill, C. R.; Watts, B.; Thomsen, L.; Belcher, W. J.; Greenham, N. C.; Dastoor, P. C. Nanoscale Quantitative Chemical Mapping of Conjugated Polymer Blends. Nano Lett. 2006, 6, 1202-1206.

272. Erb, T.; Zhokhavets, U.; Gobsch, G.; Raleva, S.; Stuhn, B.; Schilinsky, P.; Waldauf, C.; Brabec, C. J. Correlation between Structural and Optical Properties of Composite Polymer/Fullerene Films for Organic Solar Cells. Adv. Funct. Mater. 2005, 15, 1193-1196.

273. Chiu, M.-Y.; Jeng, U. S.; Su, C.-H.; Liang, K. S.; Wei, K.-H. Simultaneous Use of Small- and Wide-Angle X-Ray Techniques to Analyze Nanometerscale Phase Separation in Polymer Heterojunction Solar Cells. Adv. Mater. 2008, 20, 2573-2578.

274. Li, C.-P.; Wu, C.-H.; Wei, K.-H.; Sheu, J.-T.; Huang, J. Y.; Jeng, U. S.; Liang, K. S. The Effect of Nanoscale Confinement on the Collective Electron Transport Behavior in an Nanoparticles Self-Assembled in a Nanostructured Polystyrene-Block-Poly(4-Vinylpyridine) Diblock Copolymer Ultra-Thin Film. Adv. Funct. Mater. 2007, 17, 2283-2290.

275. Binnig, G.; Quate, C. F.; Gerber, C. Atomic Force Microscope. Phys. Rev. Lett. 1986, 56, 930-933.

276. Giessibl, F. J. Advances in Atomic Force Microscopy. Reviews of Modern Physics. 2003, 75, 949-983.

277. Pingree, L. S. C.; Reid, O. G.; Ginger, D. S. Electrical Scanning Probe Microscopy on Active Organic Electronic Devices. Adv. Mater. 2009, 21, 19-28.

278. Flores, S. M.; Toca-Herrera, J. L. The New Future of Scanning Probe Microscopy: Combining Atomic Force Microscopy with Other Surface-Sensitive Techniques, Optical and Microscopy Fluorescence Techniques. Nanoscale. 2009, 1, 40-49.

121

279. Dante, M.; Garcia, A.; Nguyen, T.-Q. Three-Dimensional Nanoscale Organization of Highly Efficient Low Band-Gap Conjugated Polymer Bulk Heterojunction Solar Cells. Journal of Physical Chemistry C. 2009, 113, 1596-1600.

280. Moon, J. S.; Lee, J. K.; Cho, S.; Byun, J.; Heeger, A. J. "Columnlike" Structure of the Cross-Sectional Morphology of Bulk Heterojunction Materials. Nano Lett. 2009, 9, 230-234.

281. Engmann, S.; Turkovic, V.; Gobsch, G.; Hoppe, H. Ellipsometric Investigation of the Shape of Nanodomains in Polymer/Fullerene Films. Adv. Energy Mater. 2011, 1, 684-689.

282. Engmann, S.; Turkovic, V.; Hoppe, H.; Gobsch, G. Aging of Polymer/Fullerene Films: Temporal Development of Composition Profiles. Synt. Met. 2012, 161, 2540-2543.

283. Engmann, S.; Turkovic, V.; Denner, P.; Hoppe, H.; Gobsch, G. Optical Order of the Polymer Phase within Polymer/Fullerene Blend Films. Journal of Polymer Science Part B-Polymer Physics. 2012, 50, 1363-1373.

284. Engmann, S.; Singh, C. R.; Turkovic, V.; Hoppe, H.; Gobsch, G. Direct Correlation of the Organic Solar Cell Device Performance to the in-Depth Distribution of Highly Ordered Polymer Domains in Polymer/Fullerene Films. Adv. Energy Mater. 2013, 3, 1463-1472.

285. Engmann, S.; Turkovic, V.; Hoppe, H.; Gobsch, G. Revealing the Active Layer Morphology within Complete Solar Cell Devices Via Spectroscopic Ellipsometry. Journal of Physical Chemistry C. 2013, 117, 25205-25210.

286. Singh, C. R.; Gupta, G.; Lohwasser, R.; Engmann, S.; Balko, J.; Thelakkat, M.; Thurn-Albrecht, T.; Hoppe, H. Correlation of Charge Transport with Structural Order in Highly Ordered Melt-Crystallized Poly(3-Hexylthiophene) Thin Films. Journal of Polymer Science Part B-Polymer Physics. 2013, 51, 943-951.

287. Jinnai, H.; Nishikawa, Y.; Ikehara, T.; Nishi, T. Emerging Technologies for the 3D Analysis of Polymer Structures. In NMR - 3D Analysis - Photopolymerization; A. Abe, A.-C. Albertsson, R. Duncan, K. Dusek, W.H. de Jeu, J.F. Joanny, H.-H. Kausch, S. Kobayashi, K.-S. Lee, L. Leibler, T.E. Long, I. Manners, M. Moller, O. Nuyken, E.M. Terentjev, B. Voit, and G. Wegner, Eds.; Springer: Berlin, Germany, 2004; pp 115-167.

288. Weyland, M.; Midgley, P. A. Electron Tomography. Mater. Today. 2004, 7, 32-40.

289. Gotzen, N.-A.; Huth, H.; Schick, C.; Van Assche, G.; Neus, C.; Van Mele, B. Phase Separation in Polymer Blend Thin Films Studied by Differential AC Chip Calorimetry. Polymer. 2010, 51, 647-654.

290. Turkovic, V.; Engmann, S.; Gobsch, G.; Hoppe, H. Methods in Determination of Morphological Degradation of Polymer:Fullerene Solar Cells. Synth. Met. 2012, 161, 2534-2539.

291. Swinnen, A.; Haeldermans, I.; vande Ven, M.; D'Haen, J.; Vanhoyland, G.; Aresu, S.; D'Olieslaeger, M.; Manca, J. Tuning the Dimensions of C-60-Based Needlelike Crystals in Blended Thin Films. Adv. Funct. Mater. 2006, 16, 760-765.

122

292. Jo, J.; Kim, S.-S.; Na, S.-I.; Yu, B.-K.; Kim, D.-Y. Time-Dependent Morphology Evolution by Annealing Processes on Polymer:Fullerene Blend Solar Cells. Adv. Funct. Mater. 2009, 19, 866-874.

293. Watts, B.; Belcher, W. J.; Thomsen, L.; Ade, H.; Dastoor, P. C. A Quantitative Study of PCBM Diffusion During Annealing of P3HT: PCBM Blend Films. Macromolecules. 2009, 42, 8392-8397.

294. Karagiannidis, P. G.; Kassavetis, S.; Pitsalidis, C.; Logothetidis, S. Thermal Annealing Effect on the Nanomechanical Properties and Structure of P3HT: PCBM Thin Films. Thin Solid Films. 2011, 519, 4105-4109.

295. Bertho, S.; Haeldermans, I.; Swinnen, A.; Moons, W.; Martens, T.; Lutsen, L.; Vanderzande, D.; Manca, J.; Senes, A.; Bonfiglio, A. Influence of Thermal Ageing on the Stability of Polymer Bulk Heterojunction Solar Cells. Sol. Energ. Mat. Sol. Cells. 2007, 91, 385-389.

296. Conings, B.; Bertho, S.; Vandewal, K.; Senes, A.; D'Haen, J.; Manca, J.; Janssen, R. A. J. Modeling the Temperature Induced Degradation Kinetics of the Short Circuit Current in Organic Bulk Heterojunction Solar Cells. Appl. Phys. Lett. 2010, 96, 163301.

297. Nguyen, L. H.; Hoppe, H.; Erb, T.; Guenes, S.; Gobsch, G.; Sariciftci, N. S. Effects of Annealing on the Nanomorphology and Performance of Poly(Alkylthiophene): Fullerene Bulk-Heterojunction Solar Cells. Adv. Funct. Mater. 2007, 17, 1071-1078.

298. McNeill, C. R.; Westenhoff, S.; Groves, C.; Friend, R. H.; Greenham, N. C. Influence of Nanoscale Phase Separation on the Charge Generation Dynamics and Photovoltaic Performance of Conjugated Polymer Blends: Balancing Charge Generation and Separation. Journal of Physical Chemistry C. 2007, 111, 19153-19160.

299. Prosa, T. J.; Winokur, M. J.; Moulton, J.; Smith, P.; Heeger, A. J. X-Ray Structural Studies of Poly(3-Alkylthiophenes) - an Example of an Inverse Comb. Macromolecules. 1992, 25, 4364-4372.

300. Bulle-Lieuwma, C. W. T.; van Gennip, W. J. H.; van Duren, J. K. J.; Jonkheijm, P.; Janssen, R. A. J.; Niemantsverdriet, J. W. Characterization of Polymer Solar Cells by Tof-Sims Depth Profiling. Appl. Surf. Sci. 2003, 203-204, 547-550.

301. Vaynzof, Y.; Kabra, D.; Zhao, L. H.; Chua, L. L.; Steiner, U.; Friend, R. H. Surface-Directed Spinodal Decomposition in Poly[3-Hexylthiophene] and C-61-Butyric Acid Methyl Ester Blends. ACS Nano. 2011, 5, 329-336.

302. Tanenbaum, D. M.; Dam, H. F.; Rosch, R.; Jorgensen, M.; Hoppe, H.; Krebs, F. C. Edge Sealing for Low Cost Stability Enhancement of Roll-to-Roll Processed Flexible Polymer Solar Cell Modules. Sol. Energ. Mat. Sol. Cells. 2012, 97, 157-163.

303. Madakasira, P.; Inoue, K.; Ulbricht, R.; Lee, S. B.; Zhou, M.; Ferraris, J. P.; Zakhidov, A. A. Multilayer Encapsulation of Plastic Photovoltaic Devices. Synt. Met. 2005, 155, 332-335.

304. Krebs, F. C. Encapsulation of Polymer Photovoltaic Prototypes. Sol. Energ. Mat. Sol. Cells. 2006, 90, 3633-3643.

305. Dennler, G.; Lungenschmied, C.; Neugebauer, H.; Sariciftci, N. S.; Latrèche, M.; Czeremuszkin, G.; Wertheimer, M. R. A New Encapsulation

123

Solution for Flexible Organic Solar Cells. Thin Solid Films. 2006, 511-512, 349-353.

306. Amberg-Schwab, S.; Weber, U.; Burger, A.; Nique, S.; Xalter, R. Development of Passive and Active Barrier Coatings on the Basis of Inorganic-Organic Polymers. Monatsh. Chem. 2006, 137, 657-666.

307. Kuwabara, T.; Nakayama, T.; Uozumi, K.; Yamaguchi, T.; Takahashi, K. Highly Durable Inverted-Type Organic Solar Cell Using Amorphous Titanium Oxide as Electron Collection Electrode Inserted between ITO and Organic Layer. Sol. Energ. Mat. Sol. Cells. 2008, 92, 1476-1482.

308. Jongwoon, P.; Hyogyun, H.; Youngmo, K.; Jongho, L.; Junghwan, P. Barrier Property of Liquid Getter-Filled Encapsulation for Transparent Organic Solar Cells. In 2011 37th IEEE Photovoltaic Specialists Conference (PVSC 2011); IEEE: Seattle, OR, USA, 2011; pp 001192-001194.

309. Van Mol, A. M. B.; Van De Weijer, P.; Kotzev, D. L. Opto-Electric Device and Method of Manufacturing an Opto-Electric Device. Patent Nos. EP2445028-A1; WO2012057615-A2; WO2012057615-A3; US 20140049825 A1, Sep 27, 2011.

310. Kato, T.; Ohnishi, T. Organic Photoelectric Conversion Element. Patent Nos. WO2011052572-A1; JP2011119684-A; PCT/JP2010/068946; JP2010000241245; H01L51/42, Oct 27, 2010.

311. Engmann, S.; Machalett, M.; Turkovic, V.; Rosch, R.; Radlein, E.; Gobsch, G.; Hoppe, H. Photon Recycling across a Ultraviolet-Blocking Layer by Luminescence in Polymer Solar Cells. J. Appl. Phys. 2012, 112, 034517-034514.

312. Sun, H.; Weickert, J.; Hesse, H. C.; Schmidt-Mende, L. UV Light Protection through TiO2 Blocking Layers for Inverted Organic Solar Cells. Sol. Energ. Mat. Sol. Cells. 2011, 95, 3450-3454.

313. Ryu, M. S.; Cha, H. J.; Jang, J. Improvement of Operation Lifetime for Conjugated Polymer: Fullerene Organic Solar Cells by Introducing a UV Absorbing Film. Sol. Energ. Mat. Sol. Cells. 2010, 94, 152-156.

314. Sahin, Y.; Alem, S.; de Bettignies, R.; Nunzi, J.-M. Development of Air Stable Polymer Solar Cells Using an Inverted Gold on Top Anode Structure. Thin Solid Films. 2005, 476, 340-343.

315. Krebs, F. C. Air Stable Polymer Photovoltaics Based on a Process Free from Vacuum Steps and Fullerenes. Sol. Energ. Mat. Sol. Cells. 2008, 92, 715-726.

316. Alexander, M. D.; Specker, C. E.; Dudis, D. S. Designing Air and Water Stable N-Dopable Model Compounds and Polymers II. In Reactive Intermediates in Organic and Biological Electrochemistry; D.G. Peters, H.J. Schafer, M.S. Workentin, and J. Yoshida, Eds.; The Electrochemical Society: Pennington, NJ, USA, 2001; pp 164-167.

317. Mishra, S. P.; Palai, A. K.; Kumar, A.; Srivastava, R.; Kamalasanan, M. N.; Patri, M. Highly Air-Stable Thieno 3,2-B Thiophene-Thiophene-Thiazolo 5,4-D Thiazole-Based Polymers for Light-Emitting Diodes. Macromol. Chem. Phys. 2010, 211, 1890-1899.

318. Manceau, M.; Bundgaard, E.; Carle, J. E.; Hagemann, O.; Helgesen, M.; Sondergaard, R.; Jorgensen, M.; Krebs, F. C. Photochemical Stability of Pi-

124

Conjugated Polymers for Polymer Solar Cells: A Rule of Thumb. J. Mater. Chem. 2011, 21, 4132-4141.

319. Kim, S. H.; Hwang, I.-W.; Jin, Y.; Song, S.; Moon, J.; Suh, H.; Lee, K. Long-Lived Bulk Heterojunction Solar Cells Fabricated with Photo-Oxidation Resistant Polymer. Sol. Energ. Mat. Sol. Cells. 2011, 95, 361-364.

320. Xia, Y.; Tong, J.; Li, B.; Wang, C.; Liu, H. Alternating Narrow Band Gap Copolymers Derived from Indeno 1,2-B Fluorene and Thiophene-Cored-Thieno 3,4-B Pyrazine Derivatives-Synthesis, Characterization and Comparative Studies of Photochemical Stability. J. Polym. Sci., Part A: Polym. Chem. 2011, 49, 2969-2979.

321. Gevorgyan, S. A.; Krebs, F. C. Bulk Heterojunctions Based on Native Polythiophene. Chem. Mater. 2008, 20, 4386-4390.

322. Grisorio, R.; Allegretta, G.; Mastrorilli, P.; Suranna, G. P. On the Degradation Process Involving Polyfluorenes and the Factors Governing Their Spectral Stability. Macromolecules. 2011, 44, 7977-7986.

323. Turkovic, V.; Engmann, S.; Tsierkezos, N.; Hoppe, H.; Ritter, U.; Gobsch, G. Long-Term Stabilization of Organic Solar Cells Using Hindered Phenols as Additives. ACS Appl. Mater. Interfaces. 2014, DOI: 10.1021/am5024989.

324. Turkovic, V.; Engmann, S.; Tsierkezos, N. G.; Hoppe, H.; Ritter, U.; Gobsch, G. Long-Term Stabilization of Organic Solar Cells Using UV Absorbers as Additives. submitted.

325. Pospíšil, J. Mechanistic Action of Phenolic Antioxidants in Polymers—a Review. Polym. Degrad. Stab. 1988, 20, 181-202.

326. Pospíšil, J. Chemical and Photochemical Behaviour of Phenolic Antioxidants in Polymer Stabilization—a State of the Art Report, Part I. Polym. Degrad. Stab. 1993, 40, 217-232.

327. Pospíšil, J. Chemical and Photochemical Behaviour of Phenolic Antioxidants in Polymer Stabilization: A State of the Art Report, Part II. Polym. Degrad. Stab. 1993, 39, 103-115.

328. Pospíšil, J.; Nešpůrek, S. Chain-Breaking Stabilizers in Polymers: The Current Status. Polym. Degrad. Stab. 1995, 49, 99-110.

329. Onoda, M.; Tada, K.; Zakhidov, A. A.; Yoshino, K. Photoinduced Charge Separation in Photovoltaic Cell with Heterojunction of P- and N-Type Conjugated Polymers. Thin Solid Films. 1998, 331, 76-81.

330. Chirvase, D.; Chiguvare, Z.; Knipper, M.; Parisi, J.; Dyakonov, V.; Hummelen, J. C. Electrical and Optical Design and Characterisation of Regioregular Poly(3-Hexylthiophene-2,5diyl)/Fullerene-Based Heterojunction Polymer Solar Cells. Synt. Met. 2003, 138, 299-304.

331. Min, J.; Ameri, T.; Gresser, R.; Lorenz-Rothe, M.; Baran, D.; Troeger, A.; Sgobba, V.; Leo, K.; Riede, M.; Guldi, D. M.; Brabec, C. J. Two Similar near-Infrared (IR) Absorbing Benzannulated Aza-BODIPY Dyes as near-IR Sensitizers for Ternary Solar Cells. ACS Appl. Mater. Interfaces. 2013, 5, 5609-5616.

332. Cooling, N. A.; Zhou, X. J.; Sales, T. A.; Sauer, S. E.; Lind, S. J.; Gordon, K. C.; Jones, T. W.; Burke, K. B.; Dastoor, P. C.; Belcher, W. J. A Study of the Factors Influencing the Performance of Ternary MEH-PPV:Porphyrin:PCBM Heterojunction Devices: Electronic Effects in

125

Porphyrinoid Ternary Blend Bulk Heterojunction Photovoltaic Devices. Sol. Energ. Mat. Sol. Cells. 2012, 98, 308-316.

333. Troshin, P. A.; Susarova, D. K.; Moskvin, Y. L.; Kuznetsov, I. E.; Ponomarenko, S. A.; Myshkovskaya, E. N.; Zakharcheva, K. A.; Balakai, A. A.; Babenko, S. D.; Razumov, V. F. Impedance Measurements as a Simple Tool to Control the Quality of Conjugated Polymers Designed for Photovoltaic Applications. Adv. Funct. Mater. 2010, 20, 4351-4357.

334. Leong, W. L.; Welch, G. C.; Kaake, L. G.; Takacs, C. J.; Sun, Y.; Bazan, G. C.; Heeger, A. J. Role of Trace Impurities in the Photovoltaic Performance of Solution Processed Small-Molecule Bulk Heterojunction Solar Cells. Chemical Science. 2012, 3, 2103-2109.

335. Kaake, L.; Dang, X.-D.; Leong, W. L.; Zhang, Y.; Heeger, A.; Nguyen, T.-Q. Effects of Impurities on Operational Mechanism of Organic Bulk Heterojunction Solar Cells. Adv. Mater. 2013, 25, 1706-1712.

336. Cowan, S. R.; Leong, W. L.; Banerji, N.; Dennler, G.; Heeger, A. J. Identifying a Threshold Impurity Level for Organic Solar Cells: Enhanced First-Order Recombination Via Well-Defined PC84BM Traps in Organic Bulk Heterojunction Solar Cells. Adv. Funct. Mater. 2011, 21, 3083-3092.

337. Mandoc, M.; Kooistra, F.; Hummelen, J.; De Boer, B.; Blom, P. Effect of Traps on the Performance of Bulk Heterojunction Organic Solar Cells. Appl. Phys. Lett. 2007, 91, 263505.

338. Cowan, S. R.; Roy, A.; Heeger, A. J. Recombination in Polymer-Fullerene Bulk Heterojunction Solar Cells. Phys. Rev. B. 2010, 82, 245207.

339. Kuik, M.; Koster, L. J. A.; Wetzelaer, G. A. H.; Blom, P. W. M. Trap-Assisted Recombination in Disordered Organic Semiconductors. Phys. Rev. Lett. 2011, 107, 256805.

340. Hall, R. N. Electron-Hole Recombination in Germanium. Physical Review. 1952, 87, 387-387.

341. Shockley, W.; Read, W. T., Jr. Statistics of the Recombinations of Holes and Electrons. Physical Review. 1952, 87, 835-842.

342. Mutoro, E.; Luerßen, B.; Günther, S.; Janek, J. The Electrode Model System Pt(O2)|YSZ: Influence of Impurities and Electrode Morphology on Cyclic Voltammograms. Solid State Ionics. 2009, 180, 1019-1033.

343. Fosnacht, D. R.; O'Keefe, T. J. Evaluation of Zinc Sulphate Electrolytes Containing Certain Impurities and Additives by Cyclic Voltammetry. J. Appl. Electrochem. 1980, 10, 495-504.

344. Endres, F.; El Abedin, S. Z.; Borissenko, N. Probing Lithium and Alumina Impurities in Air- and Water Stable Ionic Liquids by Cyclic Voltammetry and in Situ Scanning Tunneling Microscopy. Z. Phys. Chem. 2006, 220, 1377-1394.

345. Camaioni, N.; Tinti, F.; Franco, L.; Fabris, M.; Toffoletti, A.; Ruzzi, M.; Montanari, L.; Bonoldi, L.; Pellegrino, A.; Calabrese, A.; Po, R. Effect of Residual Catalyst on Solar Cells Made of a Fluorene-Thiophene-Benzothiadiazole Copolymer as Electron-Donor: A Combined Electrical and Photophysical Study. Org. Electron. 2012, 13, 550-559.

346. Nikiforov, M. P.; Lai, B.; Chen, W.; Chen, S.; Schaller, R. D.; Strzalka, J.; Maser, J.; Darling, S. B. Detection and Role of Trace Impurities in High-Performance Organic Solar Cells. Energ. Environ. Sci. 2013, 6, 1513-1520.

126

347. Dupuis, A.; Wong-Wah-Chung, P.; Rivaton, A.; Gardette, J.-L. Influence of the Microstructure on the Photooxidative Degradation of Poly(3-Hexylthiophene). Polym. Degrad. Stab. 2012, 97, 366-374.

348. Renaud, C.; Huang, C. H.; Lee, C. W.; Le Rendu, P.; Nguyen, T. P. Study of Trap States in Polyfluorene Based Devices by Using TSC Technique. Thin Solid Films. 2008, 516, 7209-7213.

349. Cho, S.; Lee, J. K.; Moon, J. S.; Yuen, J.; Lee, K.; Heeger, A. J. Bulk Heterojunction Bipolar Field-Effect Transistors Processed with Alkane Dithiol. Org. Electron. 2008, 9, 1107-1111.

350. Coates, N. E.; In-Wook, H.; Peet, J.; Bazan, G. C.; Moses, D.; Heeger, A. J. 1,8-Octanedithiol as a Processing Additive for Bulk Heterojunction Materials: Enhanced Photoconductive Response. Appl. Phys. Lett. 2008, 93, 072105.

351. Lee, J. K.; Coates, N. E.; Cho, S.; Cho, N. S.; Moses, D.; Bazan, G. C.; Lee, K.; Heeger, A. J. Efficacy of TiOx Optical Spacer in Bulk-Heterojunction Solar Cells Processed with 1,8-Octanedithiol. Appl. Phys. Lett. 2008, 92.

352. Morana, M.; Wegscheider, M.; Bonanni, A.; Kopidakis, N.; Shaheen, S.; Scharber, M.; Zhu, Z.; Waller, D.; Gaudiana, R.; Brabec, C. Bipolar Charge Transport in PCPDTBT-PCBM Bulk-Heterojunctions for Photovoltaic Applications. Adv. Funct. Mater. 2008, 18, 1757-1766.

353. Hwang, I. W.; Cho, S.; Kim, J. Y.; Lee, K.; Coates, N. E.; Moses, D.; Heeger, A. J. Carrier Generation and Transport in Bulk Heterojunction Films Processed with 1,8-Octanedithiol as a Processing Additive. J. Appl. Phys. 2008, 104.

354. Qin, D. S.; Quan, W.; Liu, J. S.; Li, G. F.; Chen, L.; Zhang, J. D.; Yan, D. H. The Manageable Formation of the Gradient-Layered Polymer/Fullerene Blend Films Via the Addition of Bathocuproine and Their Applications in Organic Solar Cells. Phys. Status Solidi A. 2012, 209, 1150-1156.

355. Shin, M.; Kim, H.; Nam, S.; Park, J.; Kim, Y. Influence of Hole-Transporting Material Addition on the Performance of Polymer Solar Cells. Energ. Environ. Sci. 2010, 3, 1538-1543.

356. Lobez, J. M.; Andrew, T. L.; Bulovic, V.; Swager, T. M. Improving the Performance of P3HT-Fullerene Solar Cells with Side-Chain-Functionalized Poly(Thiophene) Additives: A New Paradigm for Polymer Design. ACS Nano. 2012, 6, 3044-3056.

357. Wu, Y.-C.; Chao, Y.-H.; Wang, C.-L.; Wu, C.-T.; Hsu, C.-S.; Zeng, Y.-L.; Lin, C.-Y. Porphyrin–Diindenothieno[2,3-B]Thiophene Alternating Copolymer—a Blue-Light Harvester in Ternary-Blend Polymer Solar Cells. J. Polym. Sci., Part A: Polym. Chem. 2012, 50, 5032-5040.

358. Ray, B.; Alam, M. A. Random vs Regularized OPV: Limits of Performance Gain of Organic Bulk Heterojunction Solar Cells by Morphology Engineering. Sol. Energ. Mat. Sol. Cells. 2012, 99, 204-212.

359. Dibb, G. F. A.; Jamieson, F. C.; Maurano, A.; Nelson, J.; Durrant, J. R. Limits on the Fill Factor in Organic Photovoltaics: Distinguishing Nongeminate and Geminate Recombination Mechanisms. J. Phys. Chem. Lett. 2013, 4, 803-808.

127

360. Kazukauskas, V.; Pranaitis, M.; Kajzar, F.; Glatthaar, M.; Hinsch, A. Charge Carrier Trapping in Organic Solar Cell Structures P3HT:PCBM. Mol. Cryst. Liq. Cryst. 2007, 484, 739-747.

361. Aguirre, A.; Meskers, S. C. J.; Janssen, R. A. J.; Egelhaaf, H. J. Formation of Metastable Charges as a First Step in Photoinduced Degradation in Pi-Conjugated Polymer: Fullerene Blends for Photovoltaic Applications. Org. Electron. 2011, 12, 1657-1662.

362. Distler, A.; Sauermann, T.; Egelhaaf, H.-J.; Rodman, S.; Waller, D.; Cheon, K.-S.; Lee, M.; Guldi, D. M. The Effect of PCBM Dimerization on the Performance of Bulk Heterojunction Solar Cells. Adv. Energy Mater. 2014, 4.

363. Drori, T.; Holt, J.; Vardeny, Z. V. Optical Studies of the Charge Transfer Complex in Polythiophene/Fullerene Blends for Organic Photovoltaic Applications. Phys. Rev. B. 2010, 82, 075207.

364. Cook, S.; Furube, A.; Katoh, R. Matter of Minutes Degradation of Poly(3-Hexylthiophene) under Illumination in Air. J. Mater. Chem. 2012, 22, 4282-4289.

365. Pospıšil, J.; Habicher, W. D.; Pilař, J.; Nešpůrek, S.; Kuthan, J.; Piringer, G. O.; Zweifel, H. Discoloration of Polymers by Phenolic Antioxidants. Polym. Degrad. Stab. 2002, 77, 531-538.

366. Pospıšil, J.; Nešpůrek, S.; Zweifel, H.; Kuthan, J. Photo-Bleaching of Polymer Discoloration Caused by Quinone Methides. Polym. Degrad. Stab. 2002, 78, 251-255.

367. Pospíšil, J.; Nešpůrek, S.; Zweifel, H. The Role of Quinone Methides in Thermostabilization of Hydrocarbon Polymers—I. Formation and Reactivity of Quinone Methides. Polym. Degrad. Stab. 1996, 54, 7-14.

368. Vulic, I.; Vitarelli, G.; Zenner, J. M. Structure-Property Relationships: Phenolic Antioxidants with High Efficacy and Low Color Contribution. Macromolecular Symposia. 2001, 176, 1-16.

369. Klemchuk, P. P.; Horng, P.-L. Transformation Products of Hindered Phenolic Antioxidants and Colour Development in Polyolefins. Polym. Degrad. Stab. 1991, 34, 333-346.

370. Pacios, R.; Chatten, A. J.; Kawano, K.; Durrant, J. R.; Bradley, D. D. C.; Nelson, J. Effects of Photo-Oxidation on the Performance of Poly 2-Methoxy-5-(3 ',7 '-Dimethyloctyloxy)-1,4-Phenylene Vinylene : 6,6 -Phenyl C-61-Butyric Acid Methyl Ester Solar Cells. Adv. Funct. Mater. 2006, 16, 2117-2126.

371. Bauld, R.; Fleury, L. M.; Van Walsh, M.; Fanchini, G. Correlation between Density of Paramagnetic Centers and Photovoltaic Degradation in Polythiophene-Fullerene Bulk Heterojunction Solar Cells. Appl. Phys. Lett. 2012, 101, 103306-103304.

372. Peters, C. H.; Sachs-Quintana, I. T.; Kastrop, J. P.; Beaupré, S.; Leclerc, M.; McGehee, M. D. High Efficiency Polymer Solar Cells with Long Operating Lifetimes. Adv. Energy Mater. 2011, 1, 491-494.

373. Pospisil, J.; Nespurek, S.; Zweifel, H. Formation and Role of Conjugated Cyclic Dienones in Polymer Stabilization. In Chemistry and Technology of Polymer Additives; S. Al-Malaika, A. Golovoy, and C.A. Wilkie, Eds.; Blackwell Science Ltd.: Oxford, UK, 1999; pp 36-61.

128

374. Pospisil, J. Chemical and Photochemical Behavior of Phenolic Antioxidants in Polymer Stabilization. I. A State-of-the-Art Report. Polym. Degrad. Stab. 1993, 40, 217-232.

375. Pospisil, J.; Nespurek, S. Highlights in the Inherent Chemical Activity of Polymer Stabilizers. In Handbook of Polymer Degradation; H.S. Hamid, M.B. Amin, and A.G. Maadhah, Eds.; Marcel Dekker, Inc.: New York, NY, USA, 1992; pp 191-277.

376. Pospíšil, J.; Nešpůrek, S. Highlights in Chemistry and Physics of Polymer Stabilization. Macromolecular Symposia. 1997, 115, 143-163.

377. Zweifel, H. Effect of Stabilization of Polypropylene During Processing and Its Influence on Long-Term Behavior under Thermal Stress. In Polymer Durability - Degradation, Stabilization, and Lifetime Prediction; R.L. Clough, N.C. Billingham, and K.T. Gillen, Eds.; American Chemical Society: Washington, DC, USA, 1996; pp 375-397.

378. Pospíšil, J.; Nešpůrek, S.; Zweifel, H. The Role of Quinone Methides in Thermostabilization of Hydrocarbon Polymers —II. Properties and Activity Mechanisms. Polym. Degrad. Stab. 1996, 54, 15-21.

379. Seeland, M.; Roesch, R.; Hoppe, H. Luminescence Imaging of Polymer Solar Cells: Visualization of Progressing Degradation. J. Appl. Phys. 2011, 109, 064513.

380. Gevorgyan, S. A.; Medford, A. J.; Bundgaard, E.; Sapkota, S. B.; Schleiermacher, H.-F.; Zimmermann, B.; Wuerfel, U.; Chafiq, A.; Lira-Cantu, M.; Swonke, T.; Wagner, M.; Brabec, C. J.; Haillant, O. O.; Voroshazi, E.; Aernouts, T.; Steim, R.; Hauch, J. A.; Elschner, A.; Pannone, M.; Xiao, M.; Langzettel, A.; Laird, D.; Lloyd, M. T.; Rath, T.; Maier, E.; Trimmel, G.; Hermenau, M.; Menke, T.; Leo, K.; Roesch, R.; Seeland, M.; Hoppe, H.; Nagle, T. J.; Burke, K. B.; Fell, C. J.; Vak, D.; Singh, T. B.; Watkins, S. E.; Galagan, Y.; Manor, A.; Katz, E. A.; Kim, T.; Kim, K.; Sommeling, P. M.; Verhees, W. J. H.; Veenstra, S. C.; Riede, M.; Christoforo, M. G.; Currier, T.; Shrotriya, V.; Schwartz, G.; Krebs, F. C. An Inter-Laboratory Stability Study of Roll-to-Roll Coated Flexible Polymer Solar Modules. Sol. Energy Mater. Sol. Cells. 2011, 95, 1398-1416.

381. Dong Hwan, W.; Jung Kyu, K.; Jung Hwa, S.; Park, O. O.; Jong Hyeok, P. Stability Comparison: A PCDTBT/PC71BM Bulk-Heterojunction Versus a P3HT/PC71BM Bulk-Heterojunction. Sol. Energ. Mat. Sol. Cells. 2012, 101, 249-255.

382. Krebs, F. C.; Norrman, K. Analysis of the Failure Mechanism for a Stable Organic Photovoltaic During 10000 h of Testing. Prog. Photovoltaics. 2007, 15, 697-712.

383. Andreasen, B.; Tanenbaum, D. M.; Hermenau, M.; Voroshazi, E.; Lloyd, M. T.; Galagan, Y.; Zimmernann, B.; Kudret, S.; Maes, W.; Lutsen, L.; Vanderzande, D.; Wurfel, U.; Andriessen, R.; Rosch, R.; Hoppe, H.; Teran-Escobar, G.; Lira-Cantu, M.; Rivaton, A.; Uzunoglu, G. Y.; Germack, D. S.; Hosel, M.; Dam, H. F.; Jorgensen, M.; Gevorgyan, S. A.; Madsen, M. V.; Bundgaard, E.; Krebs, F. C.; Norrman, K. Tof-Sims Investigation of Degradation Pathways Occurring in a Variety of Organic Photovoltaic Devices - the ISOS-3 Inter-Laboratory Collaboration. Phys. Chem. Chem. Phys. 2012, 14, 11780-11799.

129

384. Socrates, G. Infrared and Raman Characteristic Group Frequencies; John Wiley and Sons Ltd.: Chichester, England, 2001; pp 1-343.

385. Chen, T.-A.; Wu, X.; Rieke, R. D. Regiocontrolled Synthesis of Poly(3-Alkylthiophenes) Mediated by Rieke Zinc: Their Characterization and Solid-State Properties. J. Am. Chem. Soc. 1995, 117, 233-244.

386. Gugumus, F. Light Stabilizers. In Plastics Additives Handbook; R. Gachter, M. H., and P.P. Klemchuk, Eds.; Hanser/Gardner Publications Inc.: Cincinnati, OH, US, 1993; pp 129-270.

387. Pospisil, J.; Nespurek, S. Photostabilization of Coatings. Mechanisms and Performance. Prog. Polym. Sci. 2000, 25, 1261-1335.

388. Formosinho, S. J.; Arnaut, L. G. Excited-State Proton-Transfer Reactions. II. Intramolecular Reactions. Journal of Photochemistry and Photobiology a-Chemistry. 1993, 75, 21-48.

389. Ormson, S. M.; Brown, R. G. Excited-State Intramolecular Proton-Transfer. I. ESIPT to Nitrogen. Prog. React. Kinet. 1994, 19, 45-91.

390. Arnaut, L. G.; Formosinho, S. J. Excited-State Proton-Transfer Reactions. I. Fundamentals and Intermolecular Reactions. Journal of Photochemistry and Photobiology a-Chemistry. 1993, 75, 1-20.

391. Legourrierec, D.; Ormson, S. M.; Brown, R. G. Excited-State Intramolecular Proton-Transfer. II. ESIPT to Oxygen. Prog. React. Kinet. 1994, 19, 211-275.

392. Klöpffer, W. Intramolecular Proton Transfer in Ortho Hydroxybenzophenones. Journal of Polymer Science: Polymer Symposia. 1976, 57, 205-212.

393. Rieker, J.; Lemmert-Schmitt, E.; Goeller, G.; Roessler, M.; Stueber, G. J.; Schettler, H.; Kramer, H. E. A.; Stezowski, J. J.; Hoier, H. Ultraviolet Stabilizers of the 2-(Hydroxyphenyl)Benzotriazole Class: Influence of Substituents on Structure and Spectra. The Journal of Physical Chemistry. 1992, 96, 10225-10234.

394. Ghiggino, K. P. Photostabilization of Polymeric Materials. Journal of Macromolecular Science, Part A. 1996, 33, 1541-1553.

395. Goeller, G.; Rieker, J.; Maier, A.; Stezowski, J. J.; Daltrozzo, E.; Neureiter, M.; Port, H.; Wiechmann, M.; Kramer, H. E. A. Deactivation Processes of Ultraviolet Stabilizers of the 2-(Hydroxyphenyl)Benzotriazole Class with Intramolecular Hydrogen Bonds. The Journal of Physical Chemistry. 1988, 92, 1452-1458.

396. Heller, H. J. Protection of Polymers against Light Irradiation. Eur. Polym. J. 1969, 5, Supplement, 105-132.

397. Rabek, J. F. Photostabilization of Polymers: Principles and Applications; Elsevier Applied Science: London, UK, 1990; pp 1-595.

398. Stueber, G. J.; Kieninger, M.; Schettler, H.; Busch, W.; Goeller, B.; Franke, J.; Kramer, H. E. A.; Hoier, H.; Henkel, S.; Fischer, P.; Port, H.; Hirsch, T.; Rytz, G.; Birbaum, J. L. Ultraviolet Stabilizers of the 2-(2'-Hydroxyphenyl)-1,3,5-Triazine Class - Structural and Spectroscopic Characterization. J. Phys. Chem. 1995, 99, 10097-10109.

399. Ghiggino Kenneth, P.; Scully, A. D.; Bigger, S. W. Photophysics of Hydroxyphenylbenzotriazole Polymer Photostabilizers. In The Effects of

130

Radiation on High-Technology Polymers; E. Reichmanis and J.H. O'Donnell, Eds.; American Chemical Society: Washington, DC, USA, 1989; pp 57-79.

400. Gugumus, F. Photooxidation of Polymers and Its Inhibition. In Oxidation Inhibition in Organic Materials, Volume 2; P.P. Klemchuk and J. Pospisil, Eds.; CRC Press: Boca Raton, FL, USA, 1990; pp 29-162.

401. Sivokhin, V. S. On the Mechanism of the Photostabilizing Action of Aromatic Carboxylic Acids. Polymer Science U.S.S.R. 1979, 21, 1207-1213.

402. Bauer, D. R. Chemical-Criteria for Durable Automotive Topcoats. Journal of Coatings Technology. 1994, 66, 57-65.

403. Vink, P. Loss of UV Stabilisers from Polyolefins During Photo-Oxidation. In Developments in Polymer Stabilisation, Volume 3; G. Scott, Ed.; Applied Science Publishers Ltd.: London, UK, 1980; pp 117-138.

404. Pospíšil, J. Transformations of Phenolic Antioxidants and the Role of Their Products in the Long-Term Properties of Polyolefins. In Properties of Polymers; Springer Berlin Heidelberg: Berlin, Germany, 1980; pp 69-133.

405. Gerlock, J. L.; Smith, C. A.; Cooper, V. A.; Dusbiber, T. G.; Weber, W. H. On the Use of Fourier Transform Infrared Spectroscopy and Ultraviolet Spectroscopy to Assess the Weathering Performance of Isolated Clearcoats from Different Chemical Families. Polym. Degrad. Stab. 1998, 62, 225-234.

406. Gerlock, J. L.; Tang, W.; Dearth, M. A.; Korniski, T. J. Reaction of Benzotriazole Ultraviolet-Light Absorbers with Free-Radicals. Polym. Degrad. Stab. 1995, 48, 121-130.

407. Dearth, M. A.; Korniski, T. J.; Gerlock, J. L. The LC/MS/MS Characterization of Photolysis Products of Benzotriazole-Based Ultraviolet Absorbers. Polym. Degrad. Stab. 1995, 48, 111-120.

408. Weller, A. Innermolekularer Protonenübergang Im Angeregten Zustand. Zeitschrift für Elektrochemie, Berichte der Bunsengesellschaft für physikalische Chemie. 1956, 60, 1144-1147.

409. Keck, J.; Stuber, G. J.; Kramer, H. E. A. Deactivation Processes of 2-Hydroxyphenyl-1,3,5-Triazines - Polymeric and Monomeric UV Absorbers of the Benzotriazole and Triazine Class. Angew. Makromol. Chem. 1997, 252, 119-138.

410. Woessner, G.; Goeller, G.; Kollat, P.; Stezowski, J. J.; Hauser, M.; Klein, U. K. A.; Kramer, H. E. A. Photophysical and Photochemical Deactivation Processes of Ultraviolet Stabilizers of the (2-Hydroxyphenyl)Benzotriazole Class. The Journal of Physical Chemistry. 1984, 88, 5544-5550.

411. Shizuka, H.; Machii, M.; Higaki, Y.; Tanaka, M.; Tanaka, I. Excited-State and Ground-State Intramolecular Proton-Transfer Reactions of 6-(2-Hydroxy-5-Methylphenyl)-S-Triazines in Poly(Methyl Methacrylate). The Journal of Physical Chemistry. 1985, 89, 320-326.

412. Briggs, P. J.; McKellar, J. F. Effect of Some Nickel Chelate Ultraviolet Stabilisers on Triplet Anthracene. Chemistry & Industry. 1967, 622.

413. Adamczyk, A.; Wilkinson, F. Quenching of Triplet-States by Schiff-Base Nickel (II) Complexes. Journal of the Chemical Society-Faraday Transactions II. 1972, 68, 2031-2041.

131

414. Adamczyk, A.; Wilkinson, F. Triplet-State Quenching by Photostabilizers of Polyolefins. J. Appl. Polym. Sci. 1974, 18, 1225-1232.

415. Harper, D. J.; McKellar, J. F. Photostabilizing Effect of Ni(II) Chelates in Polymers. I. Polystyrene. J. Appl. Polym. Sci. 1974, 18, 1233-1238.

416. Wiles, D. M. Photostabilization of Macromolecules by Excited-State Quenching. Pure Appl. Chem. 1978, 50, 291-297.

417. Allen, N. S.; Chirinospadron, A.; Appleyard, J. H. Photo-Stabilizing Action of Metal-Chelates in Polypropylene. I. Excited-State Quenching Versus UV Antioxidant Action under Polychromatic Irradiation. Polym. Degrad. Stab. 1982, 4, 223-237.

418. Allen, N. S.; Chirinospadron, A.; Appleyard, J. H. Photostabilizing Action of Ni(II) Bis(1-Phenyl-3-Methyl-4-Decanoyl-5-Pyrazolate), Sanduvor NPU, in a Polypropylene Matrix. Polymer Photochemistry. 1984, 5, 333-349.

419. Allen, N. S.; Chirinos-Padron, A.; Appleyard, J. H. Photo-Stabilising Action of Metal Chelate Stabilisers in Polypropylene: Part V—Light Stability as a Function of Concentration and Further Studies in Metal Stearate-Stabiliser Systems. Polym. Degrad. Stab. 1983, 5, 323-338.

420. Cowell, G. K. Additives for Plastics—Antioxidants. Plast. Eng. 1976, 51-57. 421. Shelton, J. R. Stabilization against Thermal Oxidation. In Polymer

Stabilization; W.L. Hawkins, Ed.; Wiley-Interscience: New York, NY, USA, 1972; pp 29-116.

422. Gugumus, F. Antioxidants. In Plastics Additives Handbook; R. Gachter, H. Muller, and P.P. Klemchuk, Eds.; Hanser/Gardner Publications, Inc.: Cincinnati, OH, USA, 1993; pp 1-104.

423. Armstrong, C.; Husbands, M. J.; Scott, G. Mechanisms of Antioxidant Action - Antioxidant-Active Products Formed from the Dialkyl Thiodipropionate Esters. Eur. Polym. J. 1979, 15, 241-248.

424. Advapak Neo-1120 Stabilizer One-Pack MSDS; Rohm and Haas Chemicals Llc: Philadelphia, PA, USA, 22 May, 2012.

425. Xiang, X.; Chen, S.; Zhang, J.; Huang, X. Effect of Relatively Nontoxic Thermal Stabilizers on Photodegradation of Poly(Vinyl Chloride). Polym. Eng. Sci. 2010, 50, 1095-1104.

426. Arkis, E.; Balkose, D. Thermal Stabilisation of Poly(Vinyl Chloride) by Organotin Compounds. Polym. Degrad. Stab. 2005, 88, 46-51.

427. Chen, X. D.; Wang, J. S.; Shen, J. R. Effect of UV-Irradiation on Poly(Vinyl Chloride) Modified by Methyl Methacrylate-Butadiene-Styrene Copolymer. Polym. Degrad. Stab. 2005, 87, 527-533.

428. Anton-Prinet, C.; Dubois, J.; Mur, G.; Gay, M.; Audouin, L.; Verdu, J. Photoageing of Rigid PVC - II. Degradation Thickness Profiles. Polym. Degrad. Stab. 1998, 60, 275-281.

429. Gumargalieva, K. Z.; Ivanov, V. B.; Zaikov, G. E.; Moiseev, J. V.; Pokholok, T. V. Problems of Ageing and Stabilization of Poly(Vinyl Chloride). Polym. Degrad. Stab. 1996, 52, 73-79.

430. Xiang, X.; Chen, S.; Zhang, J.; Chai, R. Photodegradation of Plasticized Poly(Vinyl Chloride) Stabilized by Different Types of Thermal Stabilizers. Polym. Eng. Sci. 2011, 51, 624-631.

132

431. Dupuis, A.; Tournebize, A.; Bussiere, P. O.; Rivaton, A.; Gardette, J. L. Morphology and Photochemical Stability of P3HT:PCBM Active Layers of Organic Solar Cells. European Physical Journal-Applied Physics. 2011, 56.

432. Kim, H. J.; Park, J. H.; Lee, H. H.; Lee, D. R.; Kim, J.-J. The Effect of Al Electrodes on the Nanostructure of Poly(3-Hexylthiophene): Fullerene Solar Cell Blends During Thermal Annealing. Org. Electron. 2009, 10, 1505-1510.

433. Orimo, A.; Masuda, K.; Honda, S.; Benten, H.; Ito, S.; Ohkita, H.; Tsuji, H. Surface Segregation at the Aluminum Interface of Poly(3-Hexylthiophene)/Fullerene Solar Cells. Appl. Phys. Lett. 2010, 96, 043305.

434. Singh, C. R.; Sommer, M.; Himmerlich, M.; Wicklein, A.; Krischok, S.; Thelakkat, M.; Hoppe, H. Morphology Controlled Open Circuit Voltage in Polymer Solar Cells. Physica Status Solidi-Rapid Research Letters. 2011, 5, 247-249.

435. Peumans, P.; Uchida, S.; Forrest, S. R. Efficient Bulk Heterojunction Photovoltaic Cells Using Small-Molecular-Weight Organic Thin Films. Nature. 2003, 425, 158-162.

436. Kim, H. J.; Lee, H. H.; Kim, J.-J. Real Time Investigation of the Interface between a P3HT:PCBM Layer and an Al Electrode During Thermal Annealing. Macromol. Rapid Commun. 2009, 30, 1269-1273.

437. Chen, D.; Nakahara, A.; Wei, D.; Nordlund, D.; Russell, T. P. P3HT/PCBM Bulk Heterojunction Organic Photovoltaics: Correlating Efficiency and Morphology. Nano Lett. 2011, 11, 561-567.

438. Hayakawa, A.; Yoshikawa, O.; Fujieda, T.; Uehara, K.; Yoshikawa, S. High Performance Polythiophene/Fullerene Bulk-Heterojunction Solar Cell with a TiOx Hole Blocking Layer. Appl. Phys. Lett. 2007, 90.

439. Haruna, T. Aspects of Stabilization with Phosphorus Antioxidants in Polymers. Angew. Makromol. Chem. 1995, 232, 119-131.

440. Pearson, L. T.; Sousa, M. Optimization of Stabilizer Formulations for Polyolefin Processing. Polym. Eng. Sci. 1992, 32, 475-480.

441. Habicher, W. D.; Bauer, I.; Scheim, K.; Rautenberg, C.; Lossack, A.; Yamaguchi, K. Interactions of Stabilizers During Oxidation Processes. Macromolecular Symposia. 1997, 115, 93-125.

442. Luengo, C.; Allen, N. S.; Wilkinson, A.; Edge, M.; Parellada, M. D.; Barrio, J. A.; Santa, V. R. Photostabilization of Styrene-Ethylene-Butylene-Styrene Block Copolymer by Hindered Phenol and Phosphite Antioxidants. Journal of Vinyl & Additive Technology. 2006, 12, 2-7.

443. Verdu, J.; Rychly, J.; Audouin, L. Synergism between Polymer Antioxidants - Kinetic Modelling. Polym. Degrad. Stab. 2003, 79, 503-509.

444. Egbe, D. A. M.; Tu rk, S.; Rathgeber, S.; Kuhnlenz, F.; Jadhav, R.; Wild, A.; Birckner, E.; Adam, G.; Pivrikas, A.; Cimrova, V.; Knor, G. n.; Sariciftci, N. S.; Hoppe, H. Anthracene Based Conjugated Polymers: Correlation between Π−Π-Stacking Ability, Photophysical Properties, Charge Carrier Mobility, and Photovoltaic Performance. Macromolecules. 2010, 43, 1261-1269.

445. Jadhav, R.; Türk, S.; Kühnlenz, F.; Cimrova, V.; Rathgeber, S.; Egbe, D. A. M.; Hoppe, H. Anthracene-Containing PPE-PPV Copolymers: Effect of

133

Side-Chain Nature and Length on Photophysical and Photovoltaic Properties. Phys. Status Solidi A. 2009, 206, 2695-2699.

446. Kastner, C.; Susarova, D. K.; Jadhav, R.; Ulbricht, C.; Egbe, D. A. M.; Rathgeber, S.; Troshin, P. A.; Hoppe, H. Morphology Evaluation of a Polymer-Fullerene Bulk Heterojunction Ensemble Generated by the Fullerene Derivatization. J. Mater. Chem. 2012, 22, 15987-15997.

447. Kastner, C.; Ulbricht, C.; Egbe, D. A. M.; Hoppe, H. Polymer BHJ Solar Cell Performance Tuning by C60 Fullerene Derivative Alkyl Side-Chain Length. Journal of Polymer Science Part B-polymer Physics. 2012, 50, 1562-1566.

448. Usluer, Ö.; Kästner, C.; Abbas, M.; Ulbricht, C.; Cimrova, V.; Wild, A.; Birckner, E.; Tekin, N.; Sariciftci, N. S.; Hoppe, H.; Rathgeber, S.; Egbe, D. A. M. Charge Carrier Mobility, Photovoltaic, and Electroluminescent Properties of Anthracene-Based Conjugated Polymers Bearing Randomly Distributed Side Chains. Journal of Polymer Science Part A: Polymer Chemistry. 2012, 50, 3425-3436.

449. Muhsin, B. Reproduzierbar Effiziente Prozessierung Von Flexiblen Polymersolarmodulen. Ph. D. Thesis, Ilmenau University of Technology, March 2013.

450. Hotta, S.; Rughooputh, S. D. D. V.; Heeger, A. J.; Wudl, F. Spectroscopic Studies of Soluble Poly(3-Alkylthienylenes). Macromolecules. 1987, 20, 212-215.

451. Verploegen, E.; Mondal, R.; Bettinger, C. J.; Sok, S.; Toney, M. F.; Bao, Z. Effects of Thermal Annealing Upon the Morphology of Polymer-Fullerene Blends. Adv. Funct. Mater. 2010, 20, 3519-3529.

452. Beiley, Z. M.; Hoke, E. T.; Noriega, R.; Dacuña, J.; Burkhard, G. F.; Bartelt, J. A.; Salleo, A.; Toney, M. F.; McGehee, M. D. Morphology-Dependent Trap Formation in High Performance Polymer Bulk Heterojunction Solar Cells. Adv. Energy Mater. 2011, 1, 954-962.

453. Holdcroft, S. Photochain Scission of the Soluble Electronically Conducting Polymer: Poly(3-Hexylthiophene). Macromolecules. 1991, 24, 2119-2121.

454. D'Andrade, B. W.; Datta, S.; Forrest, S. R.; Djurovich, P.; Polikarpov, E.; Thompson, M. E. Relationship between the Ionization and Oxidation Potentials of Molecular Organic Semiconductors. Org. Electron. 2005, 6, 11-20.

455. Kumar, A.; Sista, S.; Yang, Y. Dipole Induced Anomalous S-Shape I-V Curves in Polymer Solar Cells. J. Appl. Phys. 2009, 105, 094512-094516.

456. Wagenpfahl, A.; Rauh, D.; Binder, M.; Deibel, C.; Dyakonov, V. S-Shaped Current-Voltage Characteristics of Organic Solar Devices. Phys. Rev. B. 2010, 82, 115306.

457. Motaung, D. E.; Malgas, G. F.; Arendse, C. J. Insights into the Stability and Thermal Degradation of P3HT:C-60 Blended Films for Solar Cell Applications. Journal of Materials Science. 2011, 46, 4942-4952.

458. Haillant, O.; Dumbleton, D.; Zielnik, A. An Arrhenius Approach to Estimating Organic Photovoltaic Module Weathering Acceleration Factors. Sol. Energ. Mat. Sol. Cells. 2011, 95, 1889-1895.

459. De Bettignies, R.; Leroy, J.; Firon, M.; Sentein, C. Accelerated Lifetime Measurements of P3HT : PCBM Solar Cells. Synt. Met. 2006, 156, 510-513.

134

460. Billingham, N. C. Localization of Oxidation in Polypropylene. Makromolekulare Chemie. Macromolecular Symposia. 1989, 28, 145-163.

461. Grassie, N.; Scott, G. Polymer Degradation and Stabilisation; Cambridge University Press: Cambridge, UK, 1985; pp 1-222.

462. Norrman, K.; Madsen, M. V.; Gevorgyan, S. A.; Krebs, F. C. Degradation Patterns in Water and Oxygen of an Inverted Polymer Solar Cell. J. Am. Chem. Soc. 2010, 132, 16883-16892.

463. Voroshazi, E.; Verreet, B.; Buri, A.; Mueller, R.; Di Nuzzo, D.; Heremans, P. Influence of Cathode Oxidation Via the Hole Extraction Layer in Polymer:Fullerene Solar Cells. Org. Electron. 2011, 12, 736-744.

464. Kawano, K.; Pacios, R.; Poplavskyy, D.; Nelson, J.; Bradley, D. D. C.; Durrant, J. R. Degradation of Organic Solar Cells Due to Air Exposure. Sol. Energ. Mat. Sol. Cells. 2006, 90, 3520-3530.

465. de Jong, M. P.; van Ijzendoorn, L. J.; de Voigt, M. J. A. Stability of the Interface between Indium-Tin-Oxide and Poly(3,4-Ethylenedioxythiophene)/Poly(Styrenesulfonate) in Polymer Light-Emitting Diodes. Appl. Phys. Lett. 2000, 77, 2255-2257.

466. de Leeuw, D. M.; Simenon, M. M. J.; Brown, A. R.; Einerhand, R. E. F. Stability of N-Type Doped Conducting Polymers and Consequences for Polymeric Microelectronic Devices. Synt. Met. 1997, 87, 53-59.

467. Thompson, B. C.; Kim, Y.-G.; Reynolds, J. R. Spectral Broadening in MEH-PPV:PCBM-Based Photovoltaic Devices Via Blending with a Narrow Band Gap Cyanovinylene−Dioxythiophene Polymer. Macromolecules. 2005, 38, 5359-5362.

468. Chen, C.-H.; Hsieh, C.-H.; Dubosc, M.; Cheng, Y.-J.; Hsu, C.-S. Synthesis and Characterization of Bridged Bithiophene-Based Conjugated Polymers for Photovoltaic Applications: Acceptor Strength and Ternary Blends. Macromolecules. 2009, 43, 697-708.

469. Cumpston, B. H.; Jensen, K. F. Photooxidative Stability of Substituted Poly(Phenylene Vinylene) (PPV) and Poly(Phenylene Acetylene) (PPA). J. Appl. Polym. Sci. 1998, 69, 2451-2458.

470. Bussiere, P.-O.; Rivaton, A.; Therias, S.; Gardette, J.-L. Multiscale Investigation of the Poly(N-Vinylcarbazole) Photoageing Mechanism. J. Phys. Chem. B. 2012, 116, 802-812.

471. Rivaton, A.; Mailhot, B.; Derderian, G.; Bussiere, P. O.; Gardette, J. L. Investigation of the Photophysical Processes and Photochemical Reactions Involved in PVK Films Irradiated at Lambda > 300 Nm. Macromolecules. 2003, 36, 5815-5824.

472. Carle, J. E.; Jorgensen, M.; Manceau, M.; Helgesen, M.; Hagemann, O.; Sondergaard, R.; Krebs, F. C. Fused Thiophene/Quinoxaline Low Band Gap Polymers for Photovoltaic's with Increased Photochemical Stability. Sol. Energ. Mat. Sol. Cells. 2011, 95, 3222-3226.

473. Song, S.; Jin, Y.; Kim, S. H.; Moon, J.; Kim, K.; Kim, J. Y.; Park, S. H.; Lee, K.; Suh, H. Stabilized Polymers with Novel Indenoindene Backbone against Photodegradation for LEDs and Solar Cells. Macromolecules. 2008, 41, 7296-7305.

474. Kim, H. J.; Han, A. R.; Cho, C.-H.; Kang, H.; Cho, H.-H.; Lee, M. Y.; Frechet, J. M. J.; Oh, J. H.; Kim, B. J. Solvent-Resistant Organic

135

Transistors and Thermally Stable Organic Photovoltaics Based on Cross-Linkable Conjugated Polymers. Chem. Mater. 2012, 24, 215-221.

475. Yasuda, T.; Suzuki, T.; Takahashi, M.; Han, L. Air-Stable Triarylamine-Based Amorphous Polymer as Donor Material for Bulk-Heterojunction Organic Solar Cells. Sol. Energ. Mat. Sol. Cells. 2011, 95, 3509-3515.

476. Manceau, M.; Helgesen, M.; Krebs, F. C. Thermo-Cleavable Polymers: Materials with Enhanced Photochemical Stability. Polym. Degrad. Stab. 2010, 95, 2666-2669.

477. Petersen, M. H.; Gevorgyan, S. A.; Krebs, F. C. Thermocleavable Low Band Gap Polymers and Solar Cells Therefrom with Remarkable Stability toward Oxygen. Macromolecules. 2008, 41, 8986-8994.

136

Acknowledgments

First of all, I would like to express my profoundest gratitude to my mentor Prof. Dr. Gerhard Gobsch, for guidance and great moral support, who was the leading person through all my scientific research, helping me and encouraging me in all difficult situations, until this proud moment when I can present this doctoral thesis.

As a member of Experimental Physics I group, I would like to express my thanks for good cooperation to Dr. Harald Hoppe and all colleagues who were a part of my pre-doctoral life.

Special thanks goes to my faithful partner in science, Dr. Sebastian Engmann.

On this exciting expedition through the world of science, I have encountered many wonderful people, among which I would like to specially mention Dr. Stefan Krischok and Dr. Marcel Himmerlich for cooperation on PES investigations of OPV degradation pathways; Prof. Dr. Uwe Ritter, Dr. Nikos Tsierkezos, Ms. Katrin Risch and Ms. Doreen Schneider for their expertise in chemical issues and measurements; Dr. Wichard Beenken for being a walking encyclopedia on all issues concerning physical and non-physical world and being a great friend; Dr. Pavel Troshin for valuable discussions and friendly advices; Prof. Dr. Thomas Hannappel and Prof. Dr. Klaus Heinemann for holding inspiring lectures; Dr. Alexander Groß for always lending a sympathetic ear a questioning electrical engineer stuck in the world of physical chemistry; all of the project partners and especially Prof. Dr. Jürgen Parisi and Prof. Dr. Elizabeth von Hauff from University of Oldenburg, and Dr. Frank Meyer and Dr. Miguel Carrasco-Orozco from Merck for great collaboration on our BMBF project EOS; Mr. Ingo Werner from Reagens Deutschland, Ms. Christiane Orset from Rohm and Haas Europe and Dr. Daniel Ayuk Mbi Egbe for providing material samples; German Federal Ministry of Education and Research (BMBF) and State of Thuringia for funding me; the great staff of TU Ilmenau- Ms. Katjana Kuhnt, Mr. Tilo Nicolai, Ms. Annette Mörstedt, Ms. Kirsti Legien, Ms. Yvonne Büttner, Ms. Sabrina Schneider and Ms. Jana Spindler and each and everyone at ZMN and APZ research facilities.

Finally, I thank my Mom, Papa and Uncle for being my loving Family.

137

Erklärung (gemäß Anlage 1 der Promotionsordnung der TU Ilmenau)

Ich versichere, dass ich die vorliegende Arbeit ohne unzulässige Hilfe Dritter und ohne Benutzung anderer als der angegebenen Hilfsmittel angefertigt habe. Die aus anderen Quellen direkt oder indirekt übernommenen Daten und Konzepte sind unter Angabe der Quelle gekennzeichnet.

Bei der Auswahl und Auswertung folgenden Materials haben mir die nachstehend aufgeführten Personen in der jeweils beschriebenen Weise entgeltlich/unentgeltlich geholfen: Autoren bzw. Co-Autoren der gemeinsam veröffentlichen Publikationen (siehe Seite 102)

Weitere Personen waren an der inhaltlich-materiellen Erstellung der vorliegenden Arbeit nicht beteiligt. Insbesondere habe ich hierfür nicht die entgeltliche Hilfe von Vermittlungs- bzw. Beratungsdiensten (Promotionsberater oder anderer Personen) in Anspruch genommen. Niemand hat von mir unmittelbar oder mittelbar geldwerte Leistungen für Arbeiten erhalten, die im Zusammenhang mit dem Inhalt der vorgelegten Dissertation stehen.

Die Arbeit wurde bisher weder im In- noch im Ausland in gleicher oder ähnlicher Form einer Prüfungsbehörde vorgelegt.

Ich bin darauf hingewiesen worden, dass die Unrichtigkeit der vorstehenden Erklärung als Täuschungsversuch angesehen wird und den erfolglosen Abbruch des Promotionsverfahrens zu Folge hat.

Ilmenau, 12 April 2014

Dipl. Ing. Vida Turkovic