Semiconducting Polymer Composites E-Book
187,99 €
Mehr erfahren.
- Herausgeber: Wiley-VCH
- Kategorie: Wissenschaft und neue Technologien
- Sprache: Englisch
The first part of Semiconducting Polymer Composites describes the principles and concepts of semiconducting polymer composites in general, addressing electrical conductivity, energy alignment at interfaces, morphology, energy transfer, percolation theory and processing techniques. In later chapters, different types of polymer composites are discussed: mixtures of semiconducting and insulating or semiconducting and semiconducting components, respectively. These composites are suitable for a variety of applications that are presented in detail, including transistors and solar cells, sensors and detectors, diodes and lasers as well as anti-corrosive and anti-static surface coatings.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 1053
Veröffentlichungsjahr: 2012
Ähnliche
Contents
Cover
Related Titles
Title Page
Copyright
List of Contributors
Preface
Chapter 1: Solubility, Miscibility, and the Impact on Solid-State Morphology
1.1 Introduction
1.2 General Aspects
1.3 Solubility, Solvents, and Solution Formulations
1.4 Miscibility
1.5 Conclusions
References
Chapter 2: Nanoscale Morphological Characterization for Semiconductive Polymer Blends
2.1 Introduction
2.2 The Importance of Morphology Control
2.3 The Classic Blend: MDMO-PPV/PCBM as a Model for an Amorphous Donor System
2.4 Intermezzo: Morphology Imaging with Scanning Transmission Electron Microscopy
2.5 Volume Characterization of the Photoactive Layer: Electron Tomography
2.6 Measuring Nanoscale Electrical Properties: Conductive AFM
2.7 Current Progress and Outlook
Acknowledgments
References
Chapter 3: Energy Level Alignment at Semiconductive Polymer Interfaces: Correlating Electronic Energy Levels and Electrical Conductivity
3.1 Introduction
3.2 General View of Electronic Structure of Organic Solids
3.3 Experimental Methods
3.4 Valence Electronic Structure of Organic Semiconductors: Small Molecules
3.5 Valence Electronic Structure of Polymers
3.6 Role of the Interface Dipole Layer: Its Impact on the Energy Level Alignment
3.7 Future Prospects
Acknowledgments
References
Chapter 4: Energy and Charge Transfer
4.1 Introduction
4.2 Energy Transfer
4.3 Charge Transfer in Polymer/Fullerene Composites
References
Chapter 5: Percolation Theory and Its Application in Electrically Conducting Materials
5.1 Introduction
5.2 Lattice Percolation
5.3 Continuum Percolation
5.4 Percolation Behavior When the Interparticle Conduction Is by Tunneling
5.5 The Structure of Composite Materials
5.6 The Observations and Interpretations of the σ(x) Dependence in Composite Materials
5.7 Summary and Conclusions
Acknowledgments
References
Chapter 6: Processing Technologies of Semiconducting Polymer Composite Thin Films for Photovoltaic Cell Applications
6.1 Introduction
6.2 Optimization of Bulk Heterojunction Composite Nanostructures
6.3 Fabrication of Sub-20 nm Scale Semiconducting Polymer Nanostructure
6.4 Conclusions
References
Chapter 7: Thin-Film Transistors Based on Polythiophene/Insulating Polymer Composites with Enhanced Charge Transport
7.1 Introduction
7.2 Fundamental Principle and Operating Mode of OTFTs
7.3 Strategies for Preparing High-Performance OTFTs Based on Semiconducting/Insulating Blends
7.4 Blend Films with Vertical Stratified Structure
7.5 Blened Films with Embedded P3HT Nanowires
7.6 Conclusions and Outlook
References
Chapter 8: Semiconducting Organic Molecule/Polymer Composites for Thin-Film Transistors
8.1 Introduction
8.2 Unipolar Films for OFETs
8.3 Polymer/Fullerene Ambipolar OFETs
8.4 Conclusions
References
Chapter 9: Enhanced Electrical Conductivity of Polythiophene/Insulating Polymer Composite and Its Morphological Requirement
9.1 Introduction
9.2 Phase Evolution and Morphology
9.3 Enhanced Conductivity of Conjugated Polymer/Insulating Polymer Composites at Low Doping Level: Interpenetrated Three-Dimensional Interfaces
9.4 Conductivity of Semiconducting Polymer/Insulating Polymer Composites Doped by Molecular Dopant
9.5 Mechanisms for the Enhanced Conductivity/Mobility
9.6 Perspective
Acknowledgments
References
Chapter 10: Intrinsically Conducting Polymers and Their Composites for Anticorrosion and Antistatic Applications
10.1 ICPs and Their Composites for Anticorrosion Application
10.2 Antistatic Coating
10.3 Summary
References
Chapter 11: Conjugated–Insulating Block Copolymers: Synthesis, Morphology, and Electronic Properties
11.1 Introduction
11.2 Oligo- and Polythiophene Rod–Coil Block Copolymers
11.3 Poly(p-phenylene vinylene) Block Copolymers
11.4 Polyfluorenes
11.5 Other Semiconducting Rod–Coil Systems
11.6 Conjugated–Insulating Rod–Rod Block Copolymers
11.7 Conclusions and Outlook
References
Chapter 12: Fullerene/Conjugated Polymer Composite for the State-of-the-Art Polymer Solar Cells
12.1 Introduction
12.2 Working Mechanism
12.3 Optimization of Fullerene/Polymer Solar Cells
12.4 Outlook
References
Chapter 13: Semiconducting Nanocrystal/Conjugated Polymer Composites for Applications in Hybrid Polymer Solar Cells
13.1 Introduction
13.2 Composite Materials
13.3 Device Structure
13.4 State of the Art of Hybrid Solar Cells
13.5 Novel Approaches in Hybrid Solar Cell Development
13.6 Outlook and Perspectives
Acknowledgments
References
Chapter 14: Conjugated Polymer Blends: Toward All-Polymer Solar Cells
14.1 Introduction
14.2 Review of Polymer Photophysics and Device Operation
14.3 Material Considerations
14.4 Device Achievements to Date
14.5 Key Issues Affecting All-Polymer Solar Cells
14.6 Summary and Outlook
References
Chapter 15: Conjugated Polymer Composites and Copolymers for Light-Emitting Diodes and Laser
15.1 Introduction
15.2 Properties of Organic Semiconductors
15.3 Polymer-Based Composites
15.4 Use of Polymer Composites in Photonic Applications
15.5 Conclusions
References
Chapter 16: Semiconducting Polymer Composite Based Bipolar Transistors
16.1 Introduction
16.2 Basics of Organic Field-Effect Transistors
16.3 Bipolar Field-Effect Transistors
16.4 Perspectives
References
Chapter 17: Nanostructured Conducting Polymers for Sensor Development
17.1 Introduction
17.2 Conducting Polymers and Their Nanostructures
17.3 Synthetical Methods for Conducting Polymer Nanostructures
17.4 Typical Conducting Polymer Nanostructures
17.5 Multifunctionality of Conducting Polymer Nanostructures
17.6 Conducting Polymer-Based Sensors
17.7 Summary and Outlook
Acknowledgments
References
Index
Related Titles
Klauk, H. (ed.)
Organic Electronics II
More Materials and Applications
2012
ISBN: 978-3-527-32647-1
Mittal, V. (ed.)
In-situ Synthesis of Polymer Nanocomposites
2012
ISBN: 978-3-527-32879-6
Knoll, W., Advincula, R. C. (eds.)
Functional Polymer Films
2 Volume Set
2011
ISBN: 978-3-527-32190-2
Mittal, V. (ed.)
Surface Modification of Nanotube Fillers
2011
ISBN: 978-3-527-32878-9
Chujo, Y. (ed.)
Conjugated Polymer Synthesis
Methods and Reactions
2011
ISBN: 978-3-527-32267-1
Kumar, C. S. S. R. (ed.)
Nanocomposites
2010
ISBN: 978-3-527-32168-1
Cosnier, S., Karyakin, A. (eds.)
Electropolymerization
Concepts, Materials and Applications
2010
ISBN: 978-3-527-32414-9
Leclerc, M., Morin, J.-F. (eds.)
Design and Synthesis of Conjugated Polymers
2010
ISBN: 978-3-527-32474-3
Guldi, D. M., Martín, N. (eds.)
Carbon Nanotubes and Related Structures
Synthesis, Characterization, Functionalization, and Applications
2010
ISBN: 978-3-527-32406-4
Mittal, V. (ed.)
Optimization of Polymer Nanocomposite Properties
2010
ISBN: 978-3-527-32521-4
All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.
Library of Congress Card No.: applied for
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.
Bibliographic information published by the Deutsche Nationalbibliothek
The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.d-nb.de.
©2012 Wiley-VCH Verlag & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany
All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.
Print ISBN: 978-3-527-33030-0
ePDF ISBN: 978-3-527-64871-9
ePub ISBN: 978-3-527-64870-2
mobi ISBN: 978-3-527-64869-6
oBook ISBN: 978-3-527-64868-9
Cover Design Adam-Design, Weinheim; Germany
Typesetting Thomson Digital, Noida, India
List of Contributors
Thomas D. Anthopoulos
Imperial College London
Department of Physics
South Kensington Campus
London SW7 2AZ
UK
Isaac Balberg
The Hebrew University
The Racah Institute of Physics
Jerusalem 91904
Israel
Christoph J. Brabec
Friedrich-Alexander University
Department of Material Science and Engineering
Institute of Materials for Electronics and Energy Technology
Martensstrasse 7
91058 Erlangen
Germany
and
Bavarian Center for Applied Energy Research (ZAE Bayern)
Am Weichselgarten 7
91058 Erlangen
Germany
Tang-Kuei Chang
National Cheng Kung University
Department of Chemical Engineering
Taiwan 70101
Taiwan
Kilwon Cho
Pohang University of Science and Technology
Department of Chemical Engineering
San 31 Hyojia-dong, Namgu
Pohang 790-784
Korea
Michael Eck
FMF - Freiburger
Materialforschungszentrum
Institute for Microsystems Technology
Stefan-Meier-Str. 21
79104 Freiburg
Germany
L. Jay Guo
The University of Michigan
Macromolecular Science and Engineering
and
Department of Electrical Engineering and Computer Science
Ann Arbor, MI 48109
USA
Dahlia Haynes
Carnegie Mellon University
The McCullough Group
5000 Forbes Avenue Warner Hall 608
Pittsburgh, PA 15213
USA
Ian A. Howard
Max Planck Institute for Polymer Research
Ackermannweg 10
55128 Mainz
Germany
Pascale Jolinat
Université de Toulouse
Laboratoire Plasma et Conversion d'Energie
118, Route de Narbonne
31062 Toulouse cedex 9
France
Michael Krueger
FMF - Freiburger Materialforschungszentrum
Institute for Microsystems Technology
Stefan-Meier-Str. 21
79104 Freiburg
Germany
John G. Labram
Imperial College London
Department of Physics
London SW7 2AZ
UK
Frédéric Laquai
Max Planck Institute for Polymer Research
Ackermannweg 10
55128 Mainz
Germany
Yingping Li
Graduate School of Chinese Academy of Sciences
Beijing 100039
China
Maria Antonietta Loi
University of Groningen
Zernike Institute for Advanced Materials
Nijenborgh 4
Groningen 9747 AG
The Netherlands
Joachim Loos
University of Glasgow
Kelvin Nanocharacterisation Centre (KNC)
Scottish University Physics Alliance (SUPA)
and
School of Physics and Astronomy
Glasgow G12 8QQ
Scotland
UK
Guanghao Lu
Chinese Academy of Sciences
Changchun Institute of Applied Chemistry
Changchun 130022
China
Wanli Ma
Soochow University
Institute of Functional Nano and Soft Materials
No. 1, Shizi Street
Suzhou, 215123
China
Florian Machui
Friedrich-Alexander University
Department of Material Science and Engineering
Institute of Materials for Electronics and Energy Technology
Martensstrasse 7
91058 Erlangen
Germany
Ralf Mauer
Max Planck Institute for Polymer Research
Ackermannweg 10
55128 Mainz
Germany
Richard D. McCullough
Carnegie Mellon University
The McCullough Group
5000 Forbes Avenue, Warner Hall 608
Pittsburgh, PA 15213
USA
Christopher R. McNeill
Monash University
Department of Materials Engineering
Clayton, Victoria 3800
Australia
Moshe Narkis
Technion-Israel Institute of Technology
Department of Chemical Engineering
Haifa 32000
Israel
Thien Phap Nguyen
Université de Nantes
Institut des Matériaux Jean Rouxel
CNRS
2, rue de la Houssiniére
44322 Nantes Cedex 3
France
Hui Joon Park
The University of Michigan
Macromolecular Science and Engineering
Ann Arbor, MI 48109
USA
Claudia Piliego
University of Groningen
Zernike Institute for Advanced Materials
Nijenborgh 4
Groningen 9747 AG
The Netherlands
Hongxu Qi
Tsinghua University
Department of Chemistry
Beijing 100084
China
Longzhen Qiu
Hefei University of Technology
Academy of Opto-Electronic Technology
Key Lab of Special Display Technology Ministry of Education
National Engineering Lab of Special Display Technology
National Key Lab of Advanced Display Technology
193 Tunxi Road
Hefei 230009
China
Frank-Stefan Riehle
FMF - Freiburger Materialforschungszentrum
Institute for Microsystems Technology
Stefan-Meier-Str. 21
79104 Freiburg
Germany
Ester Segal
Technion-Israel Institute of Technology
Department of Chemical Engineering
Haifa 32000
Israel
Gaoquan Shi
Tsinghua University
Department of Chemistry
Beijing 100084
China
Jeremy N. Smith
Imperial College London
Department of Physics
London SW7 2AZ
UK
Mihaela C. Stefan
Carnegie Mellon University
The McCullough Group
5000 Forbes Avenue
Warner Hall 608
Pittsburgh, PA 15213
USA
Krisztina Szendrei
University of Groningen
Zernike Institute for Advanced Materials
Nijenborgh 4
Groningen 9747 AG
The Netherlands
Lei Tao
Tsinghua University
Department of Chemistry
Beijing 100084
China
Nobuo Ueno
Chiba University
Graduate School of Advanced Integration Science
Inage-ku
Chiba 263-8522
Japan
Meixiang Wan
Chinese Academy of Sciences
Institute of Chemistry
Center for Molecular Science
Organic Solid Laboratory
Beijing 100080
China
Xianhong Wang
Chinese Academy of Sciences
Changchun Institute of Applied Chemistry
Key Laboratory of Polymer Ecomaterials
Renmin Street 5625
Changchun 130022
China
Xiaohong Wang
Hefei University of Technology
Academy of Opto-Electronic Technology
Key Lab of Special Display Technology Ministry of Education
National Engineering Lab of Special Display Technology
National Key Lab of Advanced Display Technology
193 Tunxi Road
Hefei 230009
China
Yen Wei
Tsinghua University
Department of Chemistry
Beijing 100084
China
and
National Cheng Kung University
Department of Chemical Engineering
Tainan 70101
Taiwan
Ten-Chin Wen
National Cheng Kung University
Department of Chemical Engineering
Taiwan 70101
Taiwan
Xiaoniu Yang
Chinese Academy of Sciences
Changchun Institute of Applied Chemistry
Renmin Street 5625
Changchun 130022
China
Yunfei Zhou
FMF - Freiburger Materialforschungszentrum
Institute for Microsystems Technology
Stefan-Meier-Str. 21
79104 Freiburg
Germany
Preface
The research on (semi-)conducting polymers has attracted dramatically increased attention from both academic and industrial communities. The commercial products based on these new materials, for example, polymer thin-film displays and polymer solar cells, are already available on the market. Solution-based thin-film deposition technology makes it possible to carry out large-scale device fabrication with very low cost, which has been regarded as the most attractive advantage of semiconducting polymers for applications in next-generation optoelectronic devices. In most cases, a composite instead of only one polymer species is employed to realize the specific functionality of the device, which results in more scientific questions that need to be answered, for example, with respect to morphological, interfacial, and mechanical properties as well as to charge transfer mechanisms within the composite film. A book collecting the already existing knowledge on the respective topics is necessary for new researchers to become acquainted with the field as well as for giving an overview and addressing the key questions within a short time. In addition, this book aims at giving a systematic and in-depth coverage of semiconducting polymer composites from their fundamental concepts to morphology control and their applications in real devices for researchers already working in the field. Consequently, particular attention is given to the unique advantages of semiconducting polymer composites where polymers with specific functionalities are employed to form a multicomponent material with a desired morphology in order to obtain required materials properties and high-performance devices.
This book contains three parts, where the first part describes the principles and concepts of semiconducting polymer composites, including the mechanism of morphology formation, morphology characterization, energy level alignment at interfaces, energy transfer between the components, percolation theory, and processing techniques. These composites can be classified into two categories in terms of functionality of the components, mainly the matrix polymer involved, which is detailed in Parts II and III, respectively. Part II discusses the semiconducting/insulating polymer composites where a conjugated polymer or an organic semiconductor is dispersed in an insulating polymer matrix, forming a composite with exceptional properties. Part III is concerned with semiconducting/semiconducting polymer composites where conjugated polymers are used as the matrix. The ?applications of these composites in, for example, polymer solar cells, light-emitting diodes, transistors, and biosensors are presented.
I am greatly indebted to my colleagues who have been working in the respective fields for years and have agreed to contribute their expertise to this book. Their support made it possible to present the current state-of-the-art overview of semiconducting polymer composites in terms of both its academic value and potential applications.
I would also like to thank the people at Wiley-VCH who offered me this opportunity initially, helped me to overcome numerous difficulties, and made it become reality eventually.
Xiaoniu YangChangchun, ChinaApril 2012
Chapter 1
Solubility, Miscibility, and the Impact on Solid-State Morphology
Florian Machui and Christoph J. Brabec
1.1 Introduction
In recent years, organic semiconductors have been of increasing interest in academic and industrial fields. Compared to their inorganic counterparts, they offer various advantages such as ease of processing, mechanical flexibility, and potential in low-cost fabrication of large areas [1]. Furthermore, modifications of the chemical structure allow tailoring material properties and thus enhancing the applicability [2]. After the discovery of metallic conduction in polyacetylene in 1977 by Heeger, MacDiarmid, and Shirakawa, the path was paved for new material classes of electrical conductive polymers, possible due to chemical doping of conjugated polymers. This resulted in an increase of electrical conductivity by several orders of magnitude [3]. The main advantage of organic semiconductors is their processability from solution, which opens different applications such as flat panel displays and illumination, integrated circuits, and energy conversion [4–7]. Before widespread commercial application, further scientific investigations are necessary to achieve improved device performance and environmental stability.
The first organic solar cells were based on an active composite consisting of one single material between two electrodes with different work functions. Light absorption forms Coulomb-bound electron–hole pairs, so-called excitons, which have to be separated for charge generation [8]. In single-material active layers, this is possible by overcoming the exciton binding energy, either thermally or at the contacts [9]. Since both processes have rather low (<1%) efficiencies for pristine organic semiconductors, only few excitons are dissociated and recombination is very dominant. Therefore, single-layer organic solar cells exhibit device efficiencies far below 1% [10]. The first organic bilayer solar cell was presented by Tang, where copper phthalocyanine in combination with a perylene derivative is used as light absorption composite. In bilayer devices, excitons could diffuse within the donor phase toward an interface with a strongly electronegative acceptor material, which provides enough energy for exciton separation [11]. The electron gets transferred to the acceptor (i.e., lower in energy) and the hole remains on the donor. Currently, the most commonly used concept for the active layer in organic photovoltaic devices is the bulk heterojunction (BHJ), which consists of an interpenetrating network of a hole conductor and an electron acceptor, taking care of the low exciton diffusion length [12]. The main advantage of the BHJ concept is the increased interfacial surface leading to very efficient exciton dissociation within the whole active layer of the solar cell. The most commonly employed materials are conjugated polymers as donors and fullerene derivatives as acceptors [13–16]. By spontaneous phase separation, a specific nanostructure is formed that is decisive for the charge transport, since charge separation takes place at the interface. In the field of organic photovoltaics, several groups have realized devices with efficiencies over 6% [17–20]. Significant improvements have raised certified efficiencies up to 8.3% and novel concepts are under investigation to reach efficiencies beyond 10% (Konarka, http://www.konarka.com; Heliatek, http://www.heliatek.com.) [21].
For an efficient bulk heterojunction solar cell, good control of morphology is a key aspect, which is mainly influenced by the components' solubility during processing, the components' miscibility, and the formation of the resulting film. Solubility describes to what extent a substance dissolves in a particular solvent. This is the key phenomenon with regard to the design of inks and solvent systems with mutual multicomponent solubility regimes. The miscibility of several components in the film is mainly influenced by thermodynamic parameters. Film formation is additionally influenced by the surface energy differences of the substrate to the printing medium as well as by kinetic aspects.
Upscaling from lab to mass production facilities is one of the major necessities for cost optimization. In the case of organic solar cells, this is possible by large-area roll-to-roll processing, allowing throughputs of 10 000 m2 h−1. This is orders of magnitude higher compared to silicon processing capabilities [22]. Currently, the most employed deposition method for organic solar cells is spin coating, since inherent advantages such as high film uniformity and ease of production are suitable for research activities. However, spin coating is very unfavorable for production due to its limitation in size. Doctor blading as an alternative coating technique is more suitable for larger area substrates and is easily transferred to roll-to-roll processing. For all of these techniques, it is necessary to know of the ink's solubility to adjust the formulation. Accordingly, the material parameters for ink definition are viscosity, evaporation rate of the solvent systems, and the spreading behavior. These phenomena together define the quality and the functionality of an organic semiconductor layer. Due to the high technical relevance for organic photovoltaics and, more generally, for organic electronics, the impacts of these phenomena on the performance and functionality of bulk heterojunction composite formation are the major topics in this chapter.
1.2 General Aspects
In general, chlorinated solvents are commonly used for processing in laboratories, which have restricted application in industrial operation due to safety risks and processing costs. Environment-friendly inks are therefore one decisive criterion for mass production that should provide full functionality. Since solubility is one of the determining factors for processing of the active layer in organic solar cells, several approaches are under investigation to predict solubility of the materials in question.
1.2.1 Solubility
Different approaches can be utilized to determine the solubility of a material. While simulation of solubility is a helpful tool to predict material behavior, experimental verification is of utmost importance. In order to reduce the expensive and time-consuming experimental efforts as well as frequent toxicity issues, simulations are a welcome tool to accompany experiments. One possibility to predict the material solubility is the use of solubility parameters, which was first proposed by Hildebrand and Scott and diversified by Hansen [23, 24]. In this approach, the energy of mixing is related to the vaporization energies of pure components. For liquids as well as for polymers, the solubility parameter δ was defined as the square root of the cohesive energy density (CED) with ΔEv as energy of vaporization and Vm as average molar volume. Here the energy of mixing is related to the energies of vaporization of the pure components according to Eqs. (1.1)–(1.3). The contributions to in Eq. (1.2) are the difference in enthalpy of evaporation ΔH, the absolute temperature T, and the global ideal gas constant R.
(1.1)
(1.2)
Blanks, Prausnitz, and Weimer assigned the separation of vaporization energy into a nonpolar, dispersive part and a polar part [25, 26]. The polar part was further divided into dipole–dipole contribution and hydrogen bonding contribution by Hansen with δD as solubility parameter due to dispersion forces, δP as solubility parameter due to polar dipole forces, and δH as solubility parameter due to hydrogen bonding interactions according to Eq. (1.3) [27–29].
(1.3)
Hansen solubility properties are usually plotted in a three-dimensional coordinate system with the Hansen parameters as x, y, z axes. The coordinates of a solute can be determined by analyzing the solubility of a solute in a series of solvents with known Hansen parameters. By fitting a spheroid into the solubility space, the solubility volume of this solute can be identified. The solubility space of a solute is defined by the origin of a spheroid, resulting from the three coordinates, and the three radii in each dimension, with solvents inside the spheroid and nonsolvents outside. The radius of the sphere, R0, indicates the maximum difference for solubility. Generally, good solvents are within the sphere, and bad ones are outside of it. Furthermore, the solubility “distance” parameter, Ra, between one solvent and one solute reflecting their respective partial solubility parameters can be defined with Eq. (1.4), with δD2 as dispersive component for the solvent, δD1 as dispersive component of the solute, and a, b, and c as weighting factors. Setting of a = 4 and b = c = 1 was suggested by Hansen based on empirical testing. To convert the Hansen spheroid into an ellipsoid, different ratios of weighting factors are used. When the scale for the dispersion parameter is doubled, the spheroidal shaped volume is converted into a spherical body [24].
(1.4)
Further studies by Small revealed that solubility parameters of polymers could also be calculated by using group contribution methods, which was intensified by Hoy, van Krevelen, and Coleman et al. [30–33]. The properties of molecules are investigated by separating them into smaller subgroups. The basic assumption is that the free energy of a molecule transfer between two phases is the sum of its individual contributions of groups, and that these group contributions are independent of the rest of the molecule. There is an obvious trade-off in group contributions. It is possible to define several groups in different ways. The more the subgroups used, the more accurate the group contributions become, but the less likely that there is sufficient statistical data to make predictions. More examples have been employed elsewhere [34–36]. For predicting solubility parameters using the group contribution method, frequently the following approach is used with Fi as molar attraction constant of a specific group i and Vm as molar volume:
(1.5)
Another method to predict solubility of solutes in different solvents is based on the prediction of the activity coefficient using density functional theory [37]. Molecules exhibit a rigid structure, but can possess different conformations, whose physical and chemical properties depend on their ultimate three-dimensional confirmation. Jork et al. showed that different conformations have different influence on the predictions of the activity coefficient [38]. Klamt et al. introduced a conductor-like screening model for real solvents (COSMO-RS), which allows a priori calculation of chemical potentials of one component within an arbitrary environment [39–42]. Here, modeling is realized by statistical thermodynamics where interacting molecules are substituted by corresponding pairwise interacting surface segments with densely packed contact areas. Since every segment has a constant charge density σ, the characterization of a molecule is possible by knowing the distribution function of the charge density P(σ), the σ-profile. With that the properties of the molecules are solely dependent on the number of segments. The σ-profile of a pure component results directly from the density functional theory calculation. Further methods are based on molecular dynamics or Monte Carlo simulations but discussions are beyond the topic of this chapter.
1.2.2 Miscibility–Thermodynamic Relationships
A decisive criterion of organic semiconductor applications is their ability of mixing. In general, blends of two or more components can be categorized according to the miscibility of their phases in one-phase or multiphase systems. Miscibility is usually defined by thermodynamic parameters. Here the Gibbs free mixing enthalpy ΔGm is decisive for compatibility of two phases. Figure 1.1 shows the Gibbs free energy as a function of compositions. If ΔGm is positive, the components are not miscible (A). If ΔGm is negative and the second derivative is positive, both components are totally miscible (B). Independent of composition, a homogeneous blend is formed. If ΔGm is negative and the second derivative is negative as well, the components are partially miscible (C). Phases with different composition are formed, which consist of both components [43, 44].
(1.6)
ΔGm can be determined according by changes in enthalpy (ΔHm) and entropy of mixing the components (ΔSm). Compared to low molecular mass components, the entropy increase is low for mixing polymers. Mixing of two polymers results in a smaller increase of ΔSm as compared to a binary blend of two low molecular weight components. Therefore, according to Eq. (1.6), the enthalpy change is the decisive parameter for thermodynamic miscibility [43]. The relatively smaller increase of entropy for polymers versus small molecules can be explained with Figure 1.2. The two-dimensional grids in Figure 1.2 represent places for molecules or for polymer segments. The number of possible configurations W is significantly higher for the arrangement with the small molecules. With S ~ kT ln(W), the lower entropy increase for polymer blends becomes obvious.
Figure 1.1 Gibbs free energy of mixing as a function of composition. (According to Refs. [43, 44].)
Figure 1.2 Schematic depiction of blends with components of smaller molar mass (a) and higher molar mass (a). (According to Refs [43, 44].)
Since Gibbs free energy ΔGm cannot be determined directly, thermodynamic models are used for the estimation. An often used model for polymer–polymer systems is the Flory–Huggins theory [45]. The Flory–Huggins definition of the Gibbs free energy and its implication on polymer blends are discussed in Eq. (1.7). It describes the free energy of binary systems, with the first two parts of the equation representing the entropic part and the third part describing the enthalpic phenomena. Here, is the volume fraction of component i, Vi is the molar volume of component i, B12 is the interaction parameter, and R is the ideal gas constant. In the case of polymer blends, the free energy is dominated by the enthalpic changes, which need to be negative for miscible systems. ΔHm is directly proportional to the number of interactions between the two components, and becomes negative for strong interactions such as ion, acid–base, hydrogen bonds, or dipole–dipole interactions.
(1.7)
1.3 Solubility, Solvents, and Solution Formulations
1.3.1 Solubility
In this chapter, the solubility of organic semiconductors, their influence on OPV devices, and their correlation with Hansen solubility parameters (HSPs) are discussed. Before that, experimental methods to determine the absolute solubility of organic semiconductors are reviewed. High-performance liquid chromatography (HPLC) is a separation method to identify and quantify exact concentrations of nonvolatile components. Saturated filtered solutions are analyzed and compared with standard solutions with known concentrations [48]. Spectrophotometrical measurements are also a commonly used method to determine the absolute solubility. First, saturated solutions are filtered or centrifuged to obtain true solutions. Next, these solutions are further diluted and characterized by optical absorption measurements. By comparing the optical density (OD) of the investigated solutions with the OD of calibrated master solutions, the solubility of the component in the investigated media can be determined. Examples for determination of organic semiconductor solubility measurements with this method have been reported by Walker et al. and Machui et al. [46, 47].
Ruoff et al. analyzed the solubility of pure C60 in different solvents [48]. HPLC was used to measure the solubility at room temperature in 47 solvents. Categorizing the solvents according to their chemical structure helped to identify good solvents such as naphthalenes and halogenated aromatics. In the first study on conjugated polymer:fullerene bulk heterojunction solar cells, the limited solubility of pure C60 in organic solvents and their tendency to crystallize during film formation was recognized by members of the Heeger group [12, 49]. Homogeneous stable blends with more than 80 wt% fullerene content became processable by the use of soluble C60 derivatives such as [6,6]-phenyl-C61-butyric acid methyl ester (PC61BM). A rough estimation of the solubility of PC61BM in toluene and chlorobenzene (CB) was achieved via saturated solutions by Hoppe and Sariciftci and reported with 1 wt% in toluene and 4.2 wt% in CB [64]. Kronholm and Hummelen later on published solubility values for PC61BM and [6,6]-phenyl-C71-butyric acid methyl ester (PC71BM) in different aromatic solvents, that is, toluene, p-xylene, o-xylene, CB, chloroform, and 1,2-dichlorobenzene (o-DCB) [50]. Solubility was determined by HPLC analysis of the liquid phase at room temperature. For both PC61BM and PC71BM, highest solubility was found in o-DCB (30 mg ml−1 for PC61BM), followed by CB and chloroform (each 25 mg ml−1) and o-xylene, toluene, and p-xylene (<20 mg ml−1). PC71BM was in all cases better soluble than PC61BM. Troshin et al. analyzed the solubility of different fullerene derivatives and compared them to the resulting device performance, which is shown in Figure 1.3 [51]. Especially remarkable is a steep increase of short-circuit current density (JSC), fill factor (FF), and power conversion efficiency (PCE, η) for increasing fullerene solubility in CB from 0 to about 40 mg ml−1. Higher solubility values of about 60 mg ml−1 again resulted in a decrease of device performance. For open-circuit voltage (VOC), an increase until a solubility of 30 mg ml−1 was recognizable. Higher solubility values did not change VOC.
Figure 1.3 (a–d) Relationship between solar cell output parameters (ISC, VOC, FF, and η, respectively) and solubility of the fullerene derivative used as electron-acceptor material in the active layer. The lines are included as a guide for the eye. (From Ref. [51].)
Hansen and Smith introduced Hansen solubility parameters for organic semiconductors and analyzed pristine C60 in organic solvents [52]. It was concluded that C60 would be soluble in polymers with aromatic rings or atoms that are significantly larger than carbon, such as sulfur or chlorine. The temperature-dependent solubility and the mutual solubility regimes for poly(3-hexylthiophene-2,5-diyl) (P3HT), PC61BM, and small bandgap polymer-bridged bithiophene poly[2,6-(4,4-bis-(2-ethylhexyl)-4H-cyclopenta[2,1-b;3,4-b′]-dithiophene)-alt-4,7-(2,1,3-benzothiadiazole)] (PCPDTBT) have been analyzed [47]. For the dominantly amorphous polymer PCPDTBT and the fullerene, results showed a good consistency over a broad temperature regime. Due to the semicrystalline character of P3HT, an exact determination of the solubility parameters was found difficult in a temperature regime of 25–140 °C. With increasing temperature, the solubility radius of P3HT increases significantly as well, which was explained by breaking of aggregates at elevated temperatures. Mutual solubility regimes for all three components have been identified as shown in Figure 1.4. For P3HT, the HSP parameters at 60 °C were reported as δD = 18.7 MPa1/2, δP = 1.4 MPa1/2, δH = 4.5 MPa1/2, and solubility radius R0 = 4.3 MPa1/2. The δD, δP, δH, and R0 values were determined to be 17.3, 3.6, 8.7, and 8.2 MPa1/2 for PCPDTBT and 18.7, 4.0, 6.1, and 7.0 MPa1/2 for PC61BM, respectively, at 60 °C.
Figure 1.4 HSP diagram for solutes at 60 °C with 34 solvents, 2.5 g l−1 for P3HT, PCPDTBT, and PC61BM. (From Ref. [47].)
Park et al. used Hansen solubility parameters and showed that non-halogenated solvent blends with the same Hansen parameters as o-DCB can be used to reach comparable device performance [53]. They mixed mesitylene (MS) with acetophenone (AP) in different ratios to match o-DCB Hansen parameters. Different mixtures of AP and MS were used with different ratios resulting in PCEs ranging from 1.5% (pure MS) to 3.38% (20 vol.% acetophenone) for P3HT:PC61BM cells with best external quantum efficiency (EQE) match with o-DCB. This has so far been the first combination for solvent blends and Hansen solubility parameters for organic semiconductors. Walker et al. analyzed a conjugated polymer 3,6-bis(5-(benzofuran-2-yl)thiophen-2-yl)-2,5-bis(2-ethylhexyl)pyrrolo[3,4-c]pyrrole-1,4-dione (DPP(TBFu)2) and PC71BM [46]. The solvents were classified into good, intermediate, and poor solvents. For PC71BM, mostly higher solubility values were found in comparison to PC61BM. The average δD, δP, and δH parameters were 19.33 ± 0.05, 4.78 ± 0.50, and 6.26 ± 0.48 MPa1/2, respectively, for DPP(TBFu)2 and 20.16 ± 0.28, 5.37 ± 0.80, and 4.49 ± 0.57 MPa1/2, respectively, for PC71BM. Atomic force microscopy (AFM) images of films prepared with chloroform, thiophene, trichloroethylene, and carbon disulfide were compared before and after annealing at 110 °C for 10 min. As-cast devices with the different solvents showed poor efficiencies. This is in agreement with the AFM images showing little phase separation. Annealing improves the PCE with efficiencies of up to 4.2% for carbon disulfide and 4.3% for chloroform. It was concluded that good solvents for both components result in optimal phase separation after annealing and HSPs could be used as a general tool for designing and understanding of solution-processed devices.
1.3.2 Solvents
As the active layer of organic solar cells is typically processed from solution, morphology is mainly determined by interactions between the used semiconductor solutes and the solvent during film formation. In this chapter, the influence of the used solvent on the resulting morphology and thus device performance is discussed. Different approaches to manipulate the morphology by solution processing methods are introduced.
1.3.2.1 Impact of Different Solvents on the Solid-State Morphology
Generally, good device efficiencies require the use of solvents that contain halogens (e.g., chloroform (CF), CB, o-DCB, and 1,2,4-trichlorobenzene (TCB)), whose toxicity poses potential problems for manufacturing [54–60]. Dang et al. compared different publications of P3HT:PC61BM analyzing material parameters and resulting device efficiencies including a comparison of different solvents and device performance [61]. For the most popular solvents such as CB and o-DCB, most PCEs were in the range of 2.5–4%. However, reports for other solvents for device processing such as CF, toluene, xylene, and tetrahydronaphthalene with also high efficiencies were found.
The choice of solvent has a great influence on the resulting morphology and thus on the device performance. This phenomenon was observed in case of poly[2-methoxy-5-(3,7-dimethyloctyloxy)]-1,4-phenylenevinylene (MDMO-PPV) blended with PC61BM by Shaheen et al., who compared toluene and CB as processing solvents [62]. They found a threefold better device performance for CB-processed cells, mainly attributed to higher short-circuit current density and better fill factor due to the better solubility of both components in CB (Figure 1.5). AFM images showed a smaller scale of phase separation, that is, smaller PC61BM-rich domains in MDMO-PPV-rich matrix suppressing phase segregation of PC61BM molecules into clusters, and smoother surface roughness that improved the interface contacts to the cathode.
Figure 1.5 (a) AFM images showing the surface morphology of MDMO-PPV:PC61BM blend films when spin coated from a toluene solution (left) and from a CB solution (right). The images show the first derivative of the actual surface heights. The cross sections of the true surface heights for the films were taken horizontally from the points indicated by the arrow. (b) Characteristics for devices with an active layer that is spin coated from a toluene solution (dashed line) and from a CB solution (full line). (From Ref. [62].)
Hoppe et al. further investigated the influence of various solvents on the morphology [63, 64]. For toluene as processing solvent, photoluminescence (PL) measurements indicated pure PC61BM clusters with larger extent than exciton diffusion range. PL measurements also showed increased material phase separation after annealing and a photocurrent loss due to PC61BM clusters. Phase separation for toluene was dependent on blend ratio and solute concentration. The comparison of PC61BM in toluene and CB showed larger PC61BM clusters in case of using the poorer solvent toluene as shown in Figure 1.6 [63].
Figure 1.6 Tapping mode AFM topography scans of MDMO-PPV:PC61BM 1 : 4 (by weight) blended films, spin cast from CB (a) and toluene (b) solution. The toluene-cast film exhibits height variations that are one order of magnitude larger than those on CB-cast films. Features of a few hundred nanometers in width are visible in (a), while features in (b) are around 50 nm. (Reproduced from Ref. [63].)
For MDMO-PPV:PC61BM, Rispens et al. analyzed the influence of solvents on crystal structure of PC61BM as shown in Figure 1.7 [65]. A comparison of o-DCB, CB, and xylene as spin casting solvent showed that CB was the best choice as processing solvent. Single PC61BM crystals were obtained from CB resulting in significantly higher charge mobility than from other solvents resulting in amorphous confirmations.
Figure 1.7 Molecular structure of PC61BM, crystallized from (a) o-DCB and (b) CB (red = oxygen; green = chlorine). (From Ref. [65].)
The influence of different solvents on morphology was also investigated by Ruderer et al. for P3HT and PC61BM [66]. Spin-coated films with processing solvents such as CF, toluene, CB, and xylene were investigated by optical microscopy, grazing incidence wide-angle X-ray scattering (GIWAXS), AFM, X-ray reflectivity (XRR), and grazing incidence small-angle X-ray scattering (GISAXS) investigations. Using this wide range of investigation tools led to good understanding of how processing solvents can manipulate the lateral and vertical phase separation. Major influence on device performance resulted from vertical phase separation. PC61BM clusters were formed for low-solubility solvents. P3HT crystallinity was mainly influenced by annealing, and increased with higher boiling point of the solvent attributed to longer drying time during spin coating. The lattice constants were independent for the used solvents. Figure 1.8 shows the schematic vertical morphology resulting from different solvents for P3HT (white areas) and PC61BM (black areas), neglecting phases containing both components. These structures were reconstructed representing the findings with aforementioned methods, suggesting vertical nanostructures for CF-, toluene-, CB-, and xylene-processed films. For toluene-, CB-, and xylene-processed films, lateral nanostructures were found. P3HT accumulation at the bottom was found for toluene- and CB-processed films, while PC61BM accumulation at the bottom was found for chloroform and xylene. P3HT enrichment at the bottom and PC61BM accumulation at the top are considered as advantageous for the “normal” device architecture. Nevertheless, there was no great difference in device performance for all four solvents. It was concluded that lateral and vertical structures are not the only determining factors as long as the phase separation and the material distribution are in the range of the exciton diffusion length (here from 35 to 65 nm) and percolation paths are recognizable.
Figure 1.8 Black and white schematic morphology of annealed P3HT:PC61BM films made using CF, toluene, CB, and xylene solutions, as reconstructed from the results of AFM, XRR, and GISAXS investigations. Black areas correspond to pure PC61BM phases and white to pure P3HT phases. Characteristic lengths are indicated. (Reprinted from Ref. [66].)
Yu compared the influence of different solvents on device performance [67]. As processing solvents, CF, CB, o-DCB, and TCB were used. According to absorption and PL measurements, charge transport dark current density–voltage (j–V) curve, XRD pattern, and AFM images, a higher P3HT crystallinity for higher boiling point solvents was concluded, since polymer chains have longer time for self-organization. This resulted in increased absorption and charge carrier mobility leading to higher device performance for high boiling point solvent-processed devices. Kwong et al. processed P3HT:TiO2 nanocomposite solar cells using different solvents for spin coating the active layer [68]. A comparison of tetrahydrofuran (THF), CB, CF, and xylene showed that device performance can be strongly influenced by the used solvent. Best cells in this case were achieved with xylene. It was concluded that a good solvent for P3HT with a low evaporation rate may improve the mixing of the components resulting in better exciton dissociation and short-circuit current density. AFM studies showed that the roughest surface was obtained for films spin coated from xylene. Park et al. compared the influence of different solvents on solar cells made of the copolymer poly[N-9″-heptadecanyl-2,7-carbazole-alt-5,5-(4′,7′-di-2-thienyl-2′,1′,3′-benzothiadiazole)] (PCDTBT) in bulk heterojunction composites with the fullerene derivative PC71BM [69]. Transmission electron microscopy (TEM) and AFM comparison for CF-, CB-, and o-DCB-processed films showed a decreased phase separation with decreasing volatility of solvents and a higher incident photon to electron conversion efficiency (IPCE).
Jaczewska et al. presented a polymer–solvent diagram including film structures for polystyrene (PS):polythiophene blends (1 : 1, w/w) for different processing solvents [70]. Structures were observed with different microscopic techniques, and solubility parameters were used for establishing a polymer–solubility versus solvent–solubility relation. A relation between film morphology and stability of the layers showed a dependence on the surface energy. Dewetting effects could be inhibited by decreasing polythiophene content. Furthermore, a ternary phase diagram was developed for the system polymer, fullerene, and solvent [64]. At constant temperature and pressure, a schematic diagram is shown in Figure 1.9. A decreasing amount of solvent leads to higher repulsive interactions between polymer and fullerene molecules. Removing the solvent quickly enough can freeze the blend morphology of the polymer and the fullerene, because phase separation is a temperature-dependent process in time and the system is quenched in a metastable state. Thermal annealing does reactivate the molecule mobility and allows a reorientation and eventual recrystallization of the polymer chains within the composite. In case of very slow drying (i.e., for high boiling point solvents), molecules have more time to orient resulting in higher phase separation and larger domains.
Figure 1.9 Schematic ternary phase diagram of a polymer–fullerene–solvent system at constant temperature T and constant pressure p. The arrows indicate the direction of increasing concentration; CS,i, CP,i, and CF,i are the initial concentrations of solvent, polymer, and fullerene in the solution, respectively. During film formation, a more or less rapid quenching of the solution toward a solid-state blend takes place upon extraction of the solvent. (Adapted from Ref. [64].)
1.3.2.2 Non-Halogenic Solvents
Xylene is an often used non-halogenic solvent that frequently offers comparable device performance as with halogenated solvents [61]. p-Xylene was used by Berson et al. to form P3HT nanofibers [71]. P3HT was previously dissolved in p-xylene at elevated temperatures. Nanofibers are formed after cooling to room temperature without precipitations for concentrations in a range of 0.5–3 wt% as shown in Figure 1.10. The time-dependent formation of nanofibers is monitored for a solution of P3HT in p-xylene. With more concentrated p-xylene solutions, a homogeneous thick film is obtained, which is crucial for the fabrication of photovoltaic active layers. The dimensions of the nanostructures have been determined from the AFM images; the nanofibers had lengths ranging from 0.5 to 5 µm, thicknesses ranging from 5 to 15 nm, and widths ranging from 30 to 50 nm. Moreover, cyclohexanone was also used for fiber formation. A network of fibers was obtained by using a dilute solution in cyclohexanone. Overall, this method resulted in device efficiencies of 3.4% for P3HT:PC61BM blends with no further need of annealing.
Figure 1.10 (a) Absorption spectra of a 1 wt% solution of P3HT in p-xylene heated at 80 °C to ensure complete dissolution of the polymeric material and then allowed to evolve after cooling to room temperature for (A) 2, (B) 4, (C) 6, (D) 21, (E) 28, and (F) 48 h. The solutions are cooled at a rate of 20 °C h−1. (b) AFM phase images of a pristine nanostructured P3HT:PC61BM film deposited on glass from a p-xylene solution containing 1 wt% P3HT and 1 wt% PC61BM. (From Ref. [71].)
Due to the toxicity and processing problems of halogenic solvents, replacements with non-halogenic solvents gained interest as a concept to lower safety risks and processing costs while keeping the device performance high. Tetralene (1,2,3,4- tetrahydronaphthalene) was first suggested by Hoth et al. [72]. Tetralene is a high- boiling solvent showing a lower surface tension compared to o-DCB. The tetralene formulation provided reliable inkjet printing, but suffered from poor morphology and significantly rougher surfaces demonstrated in AFM images. This is specific to the inkjet-printed trials since doctor-bladed cells fabricated using tetralene produced cells with PCE of 3.3% for P3HT:PC61BM [73]. Furthermore, toluene was used as a processing solvent. As mentioned previously, device performance is limited due to the lower solubility compared to halogenated solvents resulting in the formation of PC61BM clusters, which restrain the charge separation [62–64].
1.3.2.3 Solvent Blends
Solvent blends can be used for device fabrication since they offer the possibility to adjust the morphology via the different solubility of the solutes in the various systems. For devices containing P3HT and PC61BM, different groups investigated the influence of solvent mixtures. Kawano et al. reported that cells processed with a solvent mixture of o-DCB/CF in 60/40 (v/v) ratio had a better performance than the cells prepared from CB [74]. After annealing at 150 °C for 5 min, the cosolvent system achieved an efficiency of 3.73% compared to 3.34% for CB cells. Short-circuit current density and fill factor increased for the cosolvent system due to larger interfacial area between P3HT and PC61BM. It was found that the cell efficiency improved by adding moderate amount of CF. The highest cell efficiency was obtained, when 40 vol% CF was added into o-DCB. Higher amounts of CF led to a drop in PCE. Furthermore, surface morphology was investigated, which showed that surface roughness was higher for the cosolvent system indicating a higher P3HT chain ordering. Lange et al. investigated the influence of adding TCB to CB as processing solvent [75]. Changes in the absorption spectra compared to the pure solvents where the P3HT absorption maximum occurred between the maxima of the two pure solvents. Therefore, it was concluded that adding TCB with a higher boiling point provides P3HT chains more time to form higher crystalline parts. Chen et al. mixed o-DCB with 1-chloronaphthalene, also providing a higher boiling point compared to o-DCB [76]. Absorption spectra showed again a redshift upon addition of the high boiling point solvent 1-chloronaphthalene indicating higher order due to longer time for self-organization. The cell efficiency peaked for 5 vol% 1-chloronaphthalene at 4.3%. o-DCB and mesitylene formulations (ratio 68 : 32) were used by Hoth et al. for inkjet-printed solar cells achieving PCE of 3% [72]. The solvent blend ratio was chosen to optimize droplet formation properties according to drop volume, velocity, and angularity of the inkjet print head. The combination of o-DCB and mesitylene served two purposes: o-DCB with the higher boiling point of 180 °C was used to prevent nozzle clogging and provided a reliable jetting of the print head, and mesitylene had a lower surface tension and was used to achieve optimum wetting and spreading of the solution on the substrate. Furthermore, it offered a higher vapor pressure and a lower boiling point compared to o-DCB and increased the drying rate of the solvent mixture, which is a critical parameter for phase separation. High-boiling solvents such as o-DCB or tetralene are an essential concept to develop inks for inkjet printing.
Influence of solvent blends was also investigated for blends of a polyfluorene copolymer poly(2,7-(9,9-dioctylfluorene)-alt-5,5-(4′,7′-di-2-thienyl-2′,1′,3′-benzothiadiazole)) (APFO-3) with PC61BM for CF, as well as solvent mixtures containing 1.2% CB, xylene, and toluene offering lower vapor pressure as compared to CF [77]. An increase in photocurrent for CF/CB blends was correlated with a finer phase separation, and a decrease in photocurrent for CF/toluene and CF/xylene was attributed to rougher surface morphologies. Furthermore, time-resolved spectroscopy supported morphological results. Wang et al. used blends of o-DCB and toluene for poly[2,3-bis-(3-octyloxyphenyl)quinoxaline-5,8-diyl-alt-thiophene-2,5-diyl] (TQ1) mixed with PC71BM resulting in PCEs of 4.5% [78]. Results suggest that only 5–20 vol.% o-DCB in the solvent blend system significantly increased JSC, FF, and PCE. The differences to pure o-DCB were quite small. AFM topography images of the spin-coated films showed big grains in the range of 1 µm for films from pure toluene. Grain size decreased to 100 nm by adding 5 vol.% o-DCB, which corresponded well to the improved device performance. Smallest grain size resulted from pure o-DCB with the best efficiencies. The effect of mixed solvent was also studied for PCDTBT and PC71BM by Alem et al. [79]. CF and o-DCB were used as good solvents for these two materials. CF-processed films exhibited larger domains, showing increasing size with higher PC71BM content. The 1 : 1 mixing of CF and o-DCB was used for realizing optimum domain size resulting in power conversion efficiencies of up to 6.1%. Solvent blends containing CB and o-DCB were used for devices made from PCDTBT:PC71BM [69]. Increasing the amount of o-DCB in the CB/o-DCB mixture increased the contribution from PC71BM to the IPCE, showing pronounced peaks around 400 and 450 nm. o-DCB films showed significantly smaller phase separation. Overall, the increased IPCE could be correlated with the nanoscale phase separation.
1.3.2.4 Addition of Poor Solvents
Addition of nonsolvents to solvents can result in aggregate formation, which enhances the field-effect mobility of conjugated semiconductors. Park et al. added acetonitrile to chloroform and changed the P3HT organization from random coil conformation to an ordered aggregate structure [80]. Besides acetonitrile, different solvents such as hexane, acetone, ethanol, and dimethylformamide were added to chloroform-based P3HT inks as conformation modifiers. P3HT aggregation occurred at a certain solvent ratio and an additional redshifted absorption band appeared. Pristine P3HT–chloroform solution contained one peak at 455 nm, which was associated with intrachain π–π∗ transition. For good solvents such as chloroform, P3HT chains were well dissolved, so no sign of molecular ordering occurred. Redshift of the absorption maximum and additional absorption bands was usually associated with ordered aggregates and interchain π–π stacking of P3HT. Both were related to an increased effective conjugation length of the chain segments in the P3HT solution, thereby decreasing energies. Moulé and Meerholz used nitrobenzene (NtB) as nonsolvent for P3HT:PC61BM in CB-based inks [81]. The volume fraction of P3HT aggregates in a P3HT:PC61BM solution could be increased from 60 up to 100% with increasing NtB content (Figure 1.11). Photovoltaic devices from P3HT:PC61BM mixtures with NtB addition resulted in device efficiencies of 4% without further thermal annealing. These experiments proved that a good part of the thin-film morphology can already be introduced on the solution level.
Figure 1.11 (a) UV–Vis spectra of 3: 2 P3HT:PC61BM as-cast PV devices with 0% (solid line), 0.33% (dashed line), 0.67% (dotted line), 1.6% (dashed–dotted line), 3.2% (short dashed line), and 6.3% (solid line) nitrobenzene added into the CB solvent. Offset from the other spectra is the as-cast PV device from the o-xylene dispersion (triangles). (b) j–V curves of as-cast (upper) and heat-treated (lower) 3: 2 P3HT:PC61BM devices. The devices were cast from CB-amorph (triangles), o-xylene-amorph (circles), o-xylene-np (squares), and CB/NtB (stars). (From Ref. [81].)
A further example was presented by Park et al. using blends of acetophenone and mesitylene [53]. The boiling point difference of MS (165 °C) and AP (202 °C) resulted in an increase of concentration of AP during solvent evaporation. The external quantum efficiency nearly doubled from a maximum of 35% at 500 nm for pure MS to 69% for the solvent blend as can be seen in Figure 1.12. One of the difficulties was obtaining the same drying conditions as o-DCB, which limited the ability to fully match the film thickness for different solvent blends. The better device performance of the solvent blended systems was assumed to result from lower series resistance and a superior morphology improving phase separation of P3HT and PC61BM that was analyzed by AFM measurements. The higher boiling point and lower evaporation rate of AP could facilitate reorganization or increase crystallinity.
Figure 1.12 EQE measurement data for devices fabricated from o-DCB, MS, and 80 vol.% MS–20 vol.% AP mixture. (From Ref. [53].)
Oleic acid (OA) was also reported to improve microstructure and device performance of P3HT:PC61BM devices [82]. After thermal annealing, the P3HT:PC61BM blend film with OA showed bigger domain sizes and roughness compared to films without OA. This is a result of enlarged P3HT domains with higher crystallinity analyzed by AFM and XRD measurements. The addition of OA improves the heteromolecular mixture in the solution and induces molecular local ordering in the resulting film. This allowed the formation of well-organized films with high mobility, resulting in high device performance up to 4.3%.
1.3.2.5 Processing Additives
One approach to control the morphology is the addition of small amounts of a high boiling point solvent with selected solubility into a host solvent. The advantages for this type of processing additives are the easy application to polymers with high and low solubility and the fact that no additional processing step is necessary [83]. For example, additives such as alkylthiols or diiodoalkanes are known to selectively help fullerene aggregation due to a better fullerene solubility compared to polymers [17, 84].
Peet et al. analyzed the influence of chain length of different alkane dithiols on the efficiency of P3HT:PC61BM and PCPDTBT:PC71BM solar cells [17]. Small concentrations of alkanethiols formed P3HT aggregations and modified the P3HT:PC61BM phase separation [85]. Since for PCPDTBT thermal or solvent annealing was not successful, additives were used. Addition of 1,8-octanedithiol into CB led to a redshift film absorption peak around 800 nm. This shift to lower energies was associated with enhanced π–π∗ stacking and indicated a PCPDTBT phase with more strongly and improved local structural order as compared to films processed from pure CB. Different chain lengths of alkane dithiols were analyzed. Best cell performance was achievable for the longest alkyl chain, 1,8-octanedithiol, resulting in cell efficiencies of up to 5.5%. AFM pictures showed that a specific chain length is necessary for morphological differences. While for butanedithiol no changes were recognizable compared to no additive processing, hexanedithiol addition showed larger domains.
Lee et al. investigated the use of processing additives on PCPDTBT:PC71BM organic solar cells [84]. Morphological control could be achieved with the criteria of a selective, differential solubility of the fullerene component and a higher boiling point compared to the host solvent. For the additive, different functional end groups of a 1,8-di(R)octane were used, achieving best results of 5.12 and 4.66% for R = I or Br, respectively. Figures 1.13 and 1.14 show the j–V curves of the devices with different additives and the schematic depiction of the role of additives, respectively.
Figure 1.13j–V characteristics of PCPDTBT/PC71BM composite films with various additives: none (black), 1,8-octanedithiol (red), 1,8-dicholorooctane (green), 1,8-dibromooctane (blue), 1,8-diiodooctane (cyan), 1,8-dicyanooctane (magenta), and 1,8-octanediacetate (yellow). (Adapted from Ref. [84].)
Figure 1.14 Schematic depiction of the role of the processing additive in the self-assembly of bulk heterojunction blend materials (a) and structures of PCPDTBT, PC71BM, and additives (b). (Adapted from Ref. [84].)
Moet et al. found by modeling the photocurrent that the use of 1,8-octanedithiol can prevent recombination-limited photocurrent in PCPDTBT:PC61BM solar cells [86]. Modeling showed that the decay rate of bound electron–hole pairs is reduced by additive addition resulting in dissociation probability of 70% at short-circuit current.
The use of processing additives was further investigated by Su et al. for the polymer poly-{bi(dodecyl)thiophene-thieno[3,4-c]pyrrole-4,6-dione} (PBTTPD) in the system PBTTPD:PC71BM [87]. Diiodoalkanes with different chain lengths were added to chloroform solutions and analyzed with GISAXS and GIWAXS measurements. It was concluded that addition of the diiodoalkanes led to an improved dispersion of the PC71BM domains and, therefore, a better network morphology by reducing the grain boundaries of the PC71BM-rich phases. Diiodohexane (DIH) provided the finest dispersion of PC71BM, due to a balance of solubility for PC71BM and the interactions between additive and the polymer molecules. By using DIH, the polymer crystallinity could be increased and the device performance was improved from 5 to 7.3%.
Further GIWAXS measurements including the use of additives were investigated by Rogers et al. who used PCPDTBT in combination with PC71BM and diiodooctane or octanedithiol [83]. By using additives, the device performance could be increased from 3.2 to 5.5%. Both additives have a higher boiling point compared to the host solvent CB and the ability to solvate PC71BM. Absorption measurements suggested increased chain aggregation and improved electrical properties were suggested from mobility and photoresponsivity measurements.
The role of additives in polymer crystallinity was further investigated by Agostinelli et al. using octanedithiol (ODT) for PCPDTBT:PC71BM films [88]. By using GIXRD, absorption spectroscopy, variable angle spectroscopic ellipsometry (VASE), and time-of-flight (TOF) hole mobility measurements, the degree of order was analyzed and accompanied by transient photovoltage (TPV) measurements changes in device performance were monitored. Upon addition of ODT, the polymer crystallinity was increased, resulting in higher charge pair generation efficiency. A series of polymers with alternating thieno[3,4-b]thiophene and benzodithiophene units was investigated by Liang et al. [89]. By using o-DCB/1,8-diiodooctane (97/3, v/v) as solvent, a more finely distributed polymer/fullerene interpenetrating network was obtained and a significantly enhanced solar cell conversion efficiency of up over 6% was achieved.
Chu et al. used a low-bandgap alternating copolymer of 4,4-bis(2-ethylhexyl)dithieno[3,2-b:2′,3′-d]silole and N-octylthieno[3,4-c]pyrrole-4,6-dione (PDTSTPD) and PC71BM as active layer with and without addition of 3% 1,8-diiodooctane (DIO) [90]. Without additive, the device performance dropped significantly below 1.0%. AFM studies of the film morphology showed that PC71BM formed too large isolated domains in the blend film prepared without using DIO. As a result, the JSC dropped from 12.2 to 2.6 mA cm−2 and the VOC and FF also decreased significantly. Addition of DIO to the solution resulted in much more uniform and finer domain structure, ideal for an effective polymer:PC71BM interpenetrating network. As a result, the device performance was greatly improved up to 6.7%. This finding highlights the importance of morphology control for high-performance solar cells.
Morana et al. investigated the effect of ODT on the formation of the charge transfer complex (CTC) for C-PCPDTBT and Si-PCPDTBT [91]. Despite the pristine C-PCPDTBT, no changes were observed in the absorption spectrum of the Si-PCPDTBT films prepared with ODT. Enhanced phase segregation in the C-PCPDTBT films upon addition of ODT caused increase in the molecular luminescence to CT luminescence ratio. This is due to the reduced concentration of CT complexes by a decrease in the contact area between the polymer and the fullerene because of phase separation.
1.3.2.6 Solution Concentration
The influence of solution concentration was investigated by Hoppe and Sariciftci with constant mixing ratio of MDMO-PPV and PC61BM [64]. Besides an increase in layer thickness with increasing concentration, also the fullerene cluster size detected by AFM analysis was increased. Further investigations have been performed by Baek et al. varying the solution concentration from 1 to 3 wt%. All solid film properties such as the crystalline structure formation, the interchain interaction, and the morphology were influenced [92]. P3HT:PC61BM absorption spectra for as-cast and annealed (150 °C for 10 min) films showed decreasing absorption with increasing concentration. Slower evaporation of the solvent at lower concentration of P3HT:PC61BM leads to better crystallization, stronger interchain interaction, and more ordered phase separation of P3HT. This holds for as-cast as well as for thermally annealed films.
1.3.3 Conclusive Outlook
Several approaches have been discussed how the solid-state microstructure of bulk heterojunction composites can be controlled by the design of intelligent solvent systems. Besides the choice of the right solvent, (i) addition of additional good solvents with differing drying properties has been demonstrated to control the domain size of either component, (ii) addition of nonsolvents was shown to trigger the nucleation and subsequent aggregation of individual components, and (iii) addition of processing additives was used to cause a coarsening of the microstructure.
The general ink design for organic semiconductor multicomponent composites is based on a few rules. Generally, the processing solvent has to supply a sufficient solubility, which is typically guaranteed by using halogenated aromatic solvent systems. The processing solvent mainly influences the active layer microstructure. Different PC61BM crystal structures were obtained by using CB, o-DCB, or xylene. Low-solubility solvents, in combination with a gradual variation of the surface energy, allow to control a gradient in the vertical phase separation of the two components. The kinetics of drying does impact the size of the aggregates. Slow drying (i.e., high boiling point solvents such as o-DCB) creates microstructures with an increased crystallinity as compared to lower boiling point solvents due to enhanced reorganization. Multicomponent solvent systems offer significantly more freedom: solvent blends can be used to mimic solubility parameters of a good solvent by using nonhazardous solvents. Furthermore, solvent blends containing high and low vapor pressure solvents allow additional control over the degree of phase separation and interfacial area. Finally, high boiling point additives with selected solubility for one component over the other can trigger more finely distributed microstructures, preventing the aggregation of fullerene clusters. Table 1.1 summarizes the essential parameters for the most frequently used single solvents that are used for processing of organic electronic systems.
Table 1.1 Solvent parameters of different key solvents for OPV.
