Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Design and fabrication of polythiophene-based organic photovoltaic devices on glass and plastic substrate Ebadian, Soheil 2009

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
24-ubc_2009_fall_ebadian_soheil.pdf [ 6.13MB ]
Metadata
JSON: 24-1.0067364.json
JSON-LD: 24-1.0067364-ld.json
RDF/XML (Pretty): 24-1.0067364-rdf.xml
RDF/JSON: 24-1.0067364-rdf.json
Turtle: 24-1.0067364-turtle.txt
N-Triples: 24-1.0067364-rdf-ntriples.txt
Original Record: 24-1.0067364-source.json
Full Text
24-1.0067364-fulltext.txt
Citation
24-1.0067364.ris

Full Text

Design and Fabrication of Polythiophene-Based Organic Photovoltaic Devices on Glass and Plastic Substrate by Soheil Ebadian B.A.Sc. , The University of Tehran, 2000 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in The Faculty of Graduate Studies (Electrical and Computer Engineering) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July, 2009 c Soheil Ebadian 2009  Abstract This thesis reports our various experimental and theoretical work on design and fabrication of organic bulk heterojunction photovoltaic devices based on P3HT:PCBM blends, fabricated on glass and plastic. This work compares several measured characteristics of these devices so as to bring insight on the effect of annealing and regioregularity of the polymer on the photovoltaic characteristic of the devices. More specifically, we present the fabrication method of P3HT:PCBM organic photovoltaic devices on glass and plastic. By comparing the low and high regioregular (RR) P3HT samples used for fabrication of solar cells, we conclude that devices based on higher RR P3HT are more prone to degradation. The low RR device, on the other hand, demonstrates an increase in the hole mobility in post-annealed conditions as thermal annealing causes a more effective π - π stacking. This leads to unchanged or even higher power conversion efficiency (PCE) of the devices with lower regioregularity after thermal annealing. For the same reason, the low RR devices are found to be more stable during a four month degradation analysis while the efficiency of high RR device decreases drastically in the same period of time. We also present a method for reducing the roughness of polyethylene terephthalate (PET) substrates for fabrication of organic photovoltaics. Using PEDOT:PSS ii  Abstract we have been able to reduce the roughness drastically and have increased the yield by decreasing the possibility of top-bottom electrode short circuit. In this method the PEDOT:PSS is deposited on the substrate in a 2-step method that makes the substrate smoother and more uniform before P3HT:PCBM deposition.  iii  Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ii  Contents  iv  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii  Dedication  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  xii  Statement of Co-Authorship . . . . . . . . . . . . . . . . . . . . . . . . . xiii  1 Introduction  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  1  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  1  1.1  Motivation  1.2  Thesis Overview  1.3  Solar Cells 1.3.1  . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  3  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  4  Basic Concepts . . . . . . . . . . . . . . . . . . . . . . . . . .  6 iv  Contents The p-n Junction . . . . . . . . . . . . . . . . . . . . . . . . .  6  Photogeneration of Charge Carriers  . . . . . . . . . . . . . .  7  Charge Carrier Separation . . . . . . . . . . . . . . . . . . . .  8  1.3.2  Macroscopic Equivalents and Efficiency Parameters . . . . . .  9  1.3.3  Working Principle of the Organic Solar Cells  . . . . . . . . .  10  . . . . . . . . . . . . . . . . . . . . . .  10  Exciton Generation and Separation . . . . . . . . . . . . . . .  11  Device Structures  12  Conjugated Polymers  1.3.4  . . . . . . . . . . . . . . . . . . . . . . . .  Efficiency Characteristics  . . . . . . . . . . . . . . . . . . . .  15  J-V Characteristic . . . . . . . . . . . . . . . . . . . . . . . .  15  Maximum - Power Point . . . . . . . . . . . . . . . . . . . . .  16  Fill Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . .  16  Quantum Efficiency Bibliography  . . . . . . . . . . . . . . . . . . . . . . .  17  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  18  v  Contents 2 The Effect of Annealing on High and Low Regiregular PolythiopheneBased Bulk Heterojunction Organic Photovoltaics  . . . . . . . . .  20  2.1  Brief . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  20  2.2  Introduction  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  21  2.2.1  OPV Devises . . . . . . . . . . . . . . . . . . . . . . . . . . .  21  2.2.2  P3HT:PCBM Bulk Heterojunction Device . . . . . . . . . . .  24  Experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  26  2.3.1  Photovoltaic Device Fabrication  . . . . . . . . . . . . . . . .  26  2.3.2  Measurements  . . . . . . . . . . . . . . . . . . . . . . . . . .  28  2.4  Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . .  28  2.5  Conclusion  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  36  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  38  2.3  Bibliography  3 Reducing the Roughness of Transparent Electrodes in Organic Photovoltaic Devices on Plastic Substrate by PEDOT:PSS Treatment 42 3.1  Brief . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  42  3.2  Introduction  43  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  vi  Contents 3.3  Experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  45  3.4  Measurements  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  47  3.5  Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . .  47  3.6  Conclusion  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  52  Bibliography  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  53  4 Conclusion  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  55  4.1  Experiment on Glass  4.2  Experiment on Plastic  4.3  . . . . . . . . . . . . . . . . . . . . . . . . . .  55  . . . . . . . . . . . . . . . . . . . . . . . . .  56  Future Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  57  4.3.1  Nano-Particle Mixture of Active Layer . . . . . . . . . . . . .  57  4.3.2  Replacing ITO and Inverted Structure . . . . . . . . . . . . .  58  4.3.3  Experiments with Modified Conjugated Polymers . . . . . . .  59  Bibliography  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  60  vii  List of Figures 1.1  The cross section and energy diagram of a silicon based solar cell. Each interface causes an electrical field in the device which enhances charge separation and reduces the chance of charge recombination. . .  1.2  5  The common donor and acceptors used in fabrication of organic solar cells.  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  7  1.3  The electron hole pair generation in semiconductors.  . . . . . . . . .  8  1.4  The equivalent circuit of the solar cell. . . . . . . . . . . . . . . . . .  9  1.5  Schottky organic solar cell based on one layer of polymer semiconductor sandwiched between two metal layers. . . . . . . . . . . . . . . . .  1.6  Bilayer organic solar cell based on a layer of donor polymer and a layer of acceptor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  1.7  13  14  Bulk heterojunction device maximized the interface between the donor and acceptor and increases the efficiency of the device. . . . . . . . .  15  viii  List of Figures 1.8  J-V curve of the solar cells. The efficiency of the PV devices is the power conversion at maximum current and voltage point as shown. .  2.1  16  The energy diagram of the donor/acceptor charge separation resulting from the built-in charge at heterojunction interface. The electron and hole separation is further enhanced by PEDOT:PSS as it increases the Anodes work function resulting in a net increase in the anode/cathode work function difference. . . . . . . . . . . . . . . . . . . . . . . . . .  2.2  23  The cross section of the bulk heterojunction OPV device on ITO coated glass. The PEDOT:PSS increases the work function of the anode, leading to a better charge separation. . . . . . . . . . . . . . .  2.3  26  The J-V curves for high and low RR devices before and after annealing. As the graph shows, the annealing has a minor effect on the Isc and Voc of the low RR device while affecting those of high RR device drastically. 29  2.4  Absorption spectrum for P3HT:PCBM film after annealing for high and low RR P3HT. The absorption for 98% RR device is red-shifted in comparison to the absorption profile of cells fabricated based on 94% RR P3HT . This graph shows a higher optical density for high RR P3HT, which is in contradiction with EQE analysis as a result of higher phase-segregation in high RR device. . . . . . . . . . . . . . .  30  ix  List of Figures 2.5  The EQE curves based on different wavelengths for high and low RR devices before and after annealing. EQE drops at high RR P3HT while is almost similar for low RR cells. . . . . . . . . . . . . . . . . .  2.6  32  Log V - Log J diagram for hole mobility calculation in high and low RR devices before and after annealing. The intercept with J axis at the region with slope equal to 2 results the mobility in accordance with constant part of logarithm of Eq. 1. . . . . . . . . . . . . . . . .  2.7  33  PCE degradation atlow and high RR device in a 4-month period. Low RR device is more stable and degrades with a lower rate in comparison to the high RR device. . . . . . . . . . . . . . . . . . . . . . . . . . .  3.1  OPV device on PET substrate. P3HT:PCBM is used as the semiconductor layer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  3.2  35  44  The microscope image of the film a) after one layer deposition and before rinse b) after PEDOT:PSS rinse with DI water. Conductive polymer remains in some spots as depicted in (b) by dark color. (a) shows the 1-layer deposition of PEDOT:PSS. . . . . . . . . . . . . . .  3.3  46  AFM analysis of PEDOT:PSS Films deposited on PET substrates: (a) Film 1, before deposition and after sonication in iso-propyl alcohol, (b) Film 2, deposition of the first PEDOT:PSS, (c) Film 3, rinsing Film 2 with water for several minutes, (d) Film 4, after deposition of second PEDOT:PSS and annealing. . . . . . . . . . . . . . . . . . . . . . . .  48 x  List of Figures 3.4  Frequency transfer results for the measured AFM roughness for a) Film 1, b) Film 2 and c) Film 4. . . . . . . . . . . . . . . . . . . . . .  3.5  50  The J-V curves of the devices fabricated using 1:1 ratio of P3HT to PCBM measured at 100 mW/cm2 in the light and dark conditions. It also shows the light J-V curve of the device deposited by single layer deposition of PEDOT:PSS. . . . . . . . . . . . . . . . . . . . . . . . .  51  xi  Dedication I want to dedicate this work to my parents and to Negah, my wife.  xii  Statement of Co-Authorship Chapters 1 and 4 were written by the author. Chapter 2 is our work as a submitted paper co-authored with my supervisor, Dr. Peyman Servati and Bobak Gholamkhas from the chemistry department in Simon Fraser University. Chapter 3 is a conference paper published in IEEE Nanotechnology Materials and Devices Conference, which is co-authored by Dr. Peyman Servati and me. On both works I have performed the experiments, designed the test setups, generated and prepared the figures, written the first draft of the paper, and participated in the collaborative effort that led to the final version of the paper. For chapter 2, the idea of the experiment and the analysis of the work is partly by me and partly by Bobak Gholamkhash. On chapter 3, the idea of the experiment is by me and the analysis is done by Dr. Servati and me.  xiii  Chapter 1 Introduction 1.1  Motivation  Organic Photovoltaic (OPV) Devices have attracted enormous attention in the past few years. The huge demand for sustainable and renewable energy resources requires methods and materials that are cheaper than the conventional silicon fabrication methods. The organic semiconductors, are light weight and can be deposited on different substrates by spin coating and dip casting as well as printing and roll-to-roll processes, thus making the product more economical for production. Moreover, organic devices can be fabricated on plastic and other flexible substrates that could be employed by variety of industries and applications such as military and automobile industries. Demand for renewable energy resources has increased drastically during the past decade. The soaring demand for energy in developing countries has increased the fossil fuel prices. The growing concerns for global warming have increased the need for sustainable sources of energy. Photovoltaic (PV) industry is one of the most promising renewable energy sources. Since the year 2000, the PV power generation has almost multiplied 10 times worldwide and has increased for more than 50% in the year 2007 to 4.5 GW [1].  1  1.1. Motivation The high growth rate of thin film production and the increase of the total market share indicate that the thin film technology is gaining more and more acceptance. A thin film market share of 25 to 30% in 2010 seems realistic. In PV industry, the Inorganic materials such as CdTe, CuInSe (CIS) and amorphous silicon have dragged more attention among other thin film technologies. These are considered as the second generation of solar cells, following the crystalline silicon technology. Organic PVs are the third generation of solar cells, comprising small molecules and conjugated polymers. This generation of solar cells benefit from various advantages over the inorganic counterparts including light weight and ease of processing. The organic semiconductors can be dissolved in solvents, therefore they can be deposited on substrates by printing and other common solvent-based techniques such as spin-coating. Moreover, these organic semiconductors are deposited on curved and flexible substrates which result in several new applications. Organic materials, also, are semitransparent and can be made in variety of colors for different applications such as deposition on glass and other transparent substrates. Despite all the advantages, the organic semiconductors are less efficient than the inorganic thin film counterparts and are prone to degradation when exposed to air. The highest reported efficiency of these devices is 5% [2]. It is estimated that in 2015 the efficiency of these devices will reach 12%, which is comparable to other thin film technologies (15% for a-Si, 13% for CdTe and 10-15% for CIGS) [1]. It has been estimated that cell degradation will decrease efficiency from 5% per 1000 hours at research scale to 2% per 1000 hours in commercial modules [1]. The hybrid use of nano-particle/organic material, also, seems to be promising [3] as the nano-material based devices are going through the proof-of-concept phase. We have chosen poly(3-hexylthiophene) (P3HT) and 6,6-phenyl C61 -butyric acid methyl ester (PCBM) based solar cells as it demonstrates highest efficiency among other materials used for organic solar cells. This thesis addresses two main issues in development  2  1.2. Thesis Overview of organic solar cells which are degradation and fabrication on flexible substrates. Chapter 2 presents our work and fabrication methods for development of organic solar cells on glass. This chapter focuses on the effect of thermal annealing and regioregularity of P3HT polymer in the performance and degradation of these devices. By comparing several characteristics of low and high regioregular P3HT based solar cells, we discuss the advantages and disadvantages of each material. Also we analyze the effect of thermal annealing on both low and high regioregular P3HT. Moreover we have analyzed the degradation in both devices during a 4 month period. Chapter 3 presents our work for design and fabrication of solar cells on flexible substrates. We address the issues regarding device deposition on polyethylene terephthalate (PET) substrates. Plastic substrates coated with indium oxide are an order of magnitude rougher than ITO coated glass and common methods of solar cell deposition result in lower fabrication yield due to bottom and top electrode short circuit. We developed a method to decrease the roughness by a 2-layer deposition of PEDOT:PSS which leads to a more uniform PEDOT:PSS layer as well as a less rough surface for deposition of organic active layer.  1.2  Thesis Overview  This thesis comprises one journal paper and one conference paper. Chapter 2 is a manuscript submitted paper to Solar Energy Materials and Solar Cells Journal. It provides a review on the methods of solar cell deposition and an introduction to the effect of regioregularity and thermal annealing on the performance and efficiency of P3HT:PCBM based solar cells.  3  1.3. Solar Cells This paper also has an emphasis on the hole mobility and external quantum efficiency comparison of the two instances. To calculate the hole mobility we use space charge limited current (SCLC) theorem. We also compare the degradation in the P3HT devices and discuss the reasons behind degradation and compare the degradation profile for devices made by high and low regioregular P3HT polymers. The third chapter, which is a paper presented in IEEE Nanotechnology Materials and Devices Conference (NMDC 2009), is a study of surface roughness in indium oxide coated substrate with focus on fabrication of P3HT:PCBM organic solar cells. In this chapter we have presented a 2-step method for deposition of PEDOT:PSS conductive polymer which leads to a smooth and uniform layer proper for deposition of organic solar cells. We argue that using this method reduces the short circuit probability in the device and increases the yield by decreasing the roughness. We report the effect of this method by comparing the J-V characteristic of devices made by this method and by ordinary 1-layer spin-coating. A spacial frequency analysis also helps to demonstrate the effect of two layer deposition.  1.3  Solar Cells  In the simplest explanation, photovoltaic (PV) cell is a device that absorbs the incident light energy and generates electricity. The introduction of semiconductor as the materials with an energy gap in the band diagram brought up the hope to generate electron hole pairs (EHP) and separate them by creation of a built-in electrical field before their recombination. Using the internal intrinsic electric field in the device, the generated EHPs are extracted and transferred to the electrodes. Electrons are knocked loose from their atoms, allowing them to flow through the material to produce electricity. The generated positive charge goes the  4  1.3. Solar Cells opposite direction to the anode. In general two steps are involved in generation of electricity in the devices: electron hole pair charge generation and charge transfer to the electrodes. These two steps are the challenges for generation of electricity and each photovoltaic device needs an efficient method for performing theses two steps. It also should be mentioned that photovoltaic is a general name for the devices generating electrical energy from photons ranging from photosensors to solar cells. In this thesis we are focusing on solar cells based on organic conjugated polymers.  Figure 1.1: The cross section and energy diagram of a silicon based solar cell. Each interface causes an electrical field in the device which enhances charge separation and reduces the chance of charge recombination. In solar cells, the semiconductor structure forms a panel that converts solar energy into electricity. Silicon based cells still are the most common devices deployed by industry. The production costs for crystalline silicon cells are high as they use the conventional device  5  1.3. Solar Cells fabrication methods. The second generation of solar cells consists of polycrystalline and amorphous silicon based solar cells. Because of lower production costs, this generation of solar cells has raised hopes for fabrication of low cost and efficient solar cells. The increase in the efficiency of these devices in the recent years, also, has added to their importance in the industry. Moreover, these devices do not require high fabrication temperature during their production, which leads to the deposition of these materials on glass or even plastic substrates. This slight change has the potential to make these devices highly economical. This require a conductive also transparent electrode, which is typically indium tin oxide (ITO). Other new technologies for fabrication of PV devices are organic and polymeric PVs. The polymer solar cells are mostly based on the conjugated polymers with semiconductor characteristics [4] and are the subject of intense research because of the low cost in production and manufacturing of these devices. The OPVs are even cheaper than amorphous silicon deposition and can be spin coated or printed which makes them even more feasible.  1.3.1  Basic Concepts  The p-n Junction In order to effectively separate and transfer charges from solar cells, the presence of a p-n junction is essential. The built-in field at the p-n junction region, enhances charge separation and leads the separated charges to the corresponding electrodes. The p-n junction in the silicon and poly silicone based devices is generated by diffusion of n or p type materials in the structure. As Fig. 1.1 shows, there are several junction interfaces made in the silicon based devices in order to enhance charge transfer to the electrodes. The p+  6  Embedded Nanowires and Na  S Electrical and Computer Eng  1.3. Solar Cells  and n+ regions , the p-n interface and the use of Al and Ag as the electrodes all enhance the charge transfer across the device. The p region in the silicon device is designed to be  Organic Photovoltaic (OPV)Devices thicker as the hole mobility in silicon is lower than electron mobility and a thicker p-type region helps charge separation and limits the charge recombination across the device. solar devices (mainly based on silicon and polysilicon) areThe • Conventional  expensive are not (<and 15% efficiency). p and and n materials are efficient called donor acceptor in the organic solar cells. The electron and polymeric PVs have lower manufacturing cost and can be • Organic and hole pairs generated in organic devices are separated by the donor/acceptor interface deposited on unconventional substrates such as plastic, fabric and paper. at these Thehave effective formation of donor acceptor interface solardue cells to is of PV devices. Devices significantly lowerand efficiency (< 2%in[1]), the • Polymer low mobility of electrons andefficiency holes device in the[5]. polymer matrix. great importance for a high donor and acceptor conjugated polymers, distributed junctions are • By using Fig. 1.2 can showsextract commongenerated donors and acceptors in organic photovoltaic devices. formed that electron hole pairs.  a) P3HT  c) C60  b) POPT  d) 5,6-PCBM  Figure 1.2: The common donor and acceptors used in fabrication of organic solar cells. Example of donor (a and b) and acceptor (c and d) type conjugated polymers  Photogeneration of Charge Carriers  Na  •A  a a •E n a m •D N ch •K m co P •A S co •M in  Load  Mo Depending on the wavelength of incident light, the photons that hit the semiconductor layer are absorbed, reflected or transmitted through the cell. The absorption density is one of the Al  major parameters that affects the performance of solar cells. The photons that contribute CONJUGATED POLYMER  in generation of solar energy are only those with energies higher than the band gap of the PEDOT:PSS  silicon. These photons excite the electrons ITO from the valance band to the conduction band  •U  re •D an •H by  Glass  7 Sun Light A general cross section schematic for OPV devices  • Organic semiconductors are intrinsically p-type. By adding acceptor elements,  νi  U  •M  l •T  1.3. Solar Cells  Figure 1.3: The electron hole pair generation in semiconductors. and generate an electron hole pair (EHP), as figure 3 shows. The electrons that are excited to energies higher than the band gap of the material would create phonons which leads to heat generation [5]. A phonon is a quantized mode of vibration occurring in a rigid crystal lattice, such as the atomic lattice of a solid.  Charge Carrier Separation There are two main modes for charge carrier separation in a solar cell: • Drift of carriers, is the EHP separation which is caused by the electric field generated by the p-n junction and the depletion region in the device. Through this field, the holes move in the direction of electrical field and the electrons in the reverse direction. • Diffusion of carriers, from zones of high carrier concentration to zones of low carrier concentration. This is the movement of particles from high concentration regions to  8  1.3. Solar Cells lower concentration regions. The latter, as will be discussed, is the dominant factor in the polymer based devices, as the movement of electrons is mostly random and based on brownian random walk in polymer structures [6].  1.3.2  Macroscopic Equivalents and Efficiency Parameters  Fig. 1.4 shows the macroscopic equivalent circuit of a solar cell. The elements of the circuit are as follows:  Figure 1.4: The equivalent circuit of the solar cell.  • A current source that represents the photocurrent generated within the cell. This current flows in the opposite direction compared to the forward equivalent diode and depends on the voltage across the device. • The dark current diode which is in the inverse direction with respect to the current source.  9  1.3. Solar Cells • A series resistance Rs that gathers the ohmic contributions of the electrode and the contact between the organic semiconductor and the metal. Rs presents the capability of the organic bulk to transport charge carriers. Rs has to be lowered to ensure a maximum efficiency. • A shunt resistance Rsh that describes the potential leakage current through the device. Unlike Rs , Rsh has to be maximized to reach high efficiency cells. This resistance is typically due to manufacturing defects, rather than poor solar cell design. Low shunt resistance causes power losses in solar cells by providing an alternate current path for the light generated current. Such a diversion reduces the amount of current flowing through the solar cell junction and reduces the voltage from the solar cell.  1.3.3  Working Principle of the Organic Solar Cells  Conjugated Polymers The organic devices are based on conjugated polymers. Conjugated polymers are formed due to the presence of conjugated double bonds along the carbon backbone of the polymer. These bonds are alternately single and double. Each of these bonds contain a localized sigma σ bond which is strong. The polymer also contains a double bond which is less strong and is known as π bond. This single bond-double bond structure based on sp2 -hybridized carbon atoms, is the major reason for conductivity of these materials. Due to the resonance effects these π electrons are delocalized, resulting in high electronic polarizability. The Peierls instability splits the originally half-filled Pz band into two, the π and π ∗ bands.  10  1.3. Solar Cells Upon light absorption electrons may be excited from the bonding π into the anti-bonding π ∗ band. This absorption corresponds to the first optical excitation from the highest occupied molecular orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO). The optical band gaps of most conjugated polymers are around 2 eV. Since the introduction of conjugated polymers in 1977 [7] by H. Shirakawa et al., there has been an intensive effort for development of organic electronic devices. Organic light emitting diodes (OLED) and organic transistors have been a major subject for research. Organic photovoltaic (OPV) devices are another application of conjugated polymers that have attracted attention during the past few years. The efficiency of these devices, as low as 0.1% in 1999 [8, 9], has been increased to 4-5% [4, 10] recently. Fig. 1.2 shows the common kinds of conjugated polymers used in solar cells. When dissolved in organic solvents and sometimes even water, these polymers form a solution which can be deposited on the substrates as ”ink”. This results in much cheaper fabrication processes than the conventional semiconductor production process. The ink-like product can be spin coated, printed or even deposited in a roll-to-roll process which leads to lowcost devices. This is of excessive importance in solar cell production which requires large area devices and needs to be cheap with respect to the larger size of the device.  Exciton Generation and Separation As Fig. 1.5 shows, the incident light, if is absorbed by the organic layer, may generate an exciton, or electron-hole-pair (EHP), in the structure. This step is different with the EHP generation in inorganic devices. The greater de-localization of the carrier wave functions (wider bandwidth, lower-effective mass), and the narrower Coulomb potential well in the  11  1.3. Solar Cells inorganic photovoltaics, commonly result in the photoproduction of free electrons and holes at room temperature [11]. In OPVs, however, localization of the carrier wave function and the larger Coulomb potential well lead to photoproduction of bound electron-hole pairs, or excitons. These excitons are capable of moving the polymer structure by 5-15 nm, before recombination [12]. The internal field effect caused by the electrode is not sufficient to dissociate these charges effectively as charges are placed close to each other (∼ 1 nm) and are bound with a 0.25 eV Columbic energy [11]. To separate these charges a localized field effect is essential which is produced by introduction of donors and acceptors to the structure. In this work, P3HT and PCBM are used as the donor and acceptor respectively. P3HT based OPV devices are the most promising and efficient devices among polymerbased solar cells. This is despite higher band gap of P3HT polymer (1.9 eV) in comparison with the solar spectrum peak at 1.8 eV. One reason for outstanding performance of P3HT based devices is the efficient formation of the P3HT and PCBM mixture and the relatively uniform dispersion of PCBM in the polymer structure [13]. The highest PCE among all OPV devices is reported for P3HT:PCBM BHJ devices [4].  Device Structures Since the introduction of the organic photovoltaics several structures and material have been used for design and fabrication of an efficient and stable device. The first generation of these devices is shown in Fig. 1.5 . As shown, these device are comprised of a polymer layer sandwiched between an indium tin oxide conductive transparent electrode and Al. The incident light at the organic semiconductor layer, transfers the electrons from HOMO to LUMO. As Fig. 1.5 shows, the band bending at the Al causes a built-in field in the Al and polymer interface which enhances charge separation at the interface. The low electron  12  1.3. Solar Cells  Figure 1.5: Schottky organic solar cell based on one layer of polymer semiconductor sandwiched between two metal layers. mobility in these devices, and the short width of active layer results in a poor charge transfer and low efficiency of devices made by this method. Because of the reasons mentioned above, the bilayer structure of a donor and an acceptor heterojunction is used to fabricate organic solar cells. The bilayer device provides a heterojunction structure that enhances charge separation by the built-in field generated at donor/acceptor interface. Fig. 1.6 shows the device structure for bilayer device. It also shows the energy diagram at the interface of donor and acceptor. As Fig. 1.6 shows, at the heterojunction interface charge separation happens more effectively due to the electric field at the interface. Separated charges then hop through the donor and acceptor and contribute to the current and voltage in the device. This device has a higher efficiency in comparison to the single polymer structure as the electric field in the heterojunction interface accelerates charge separation drastically. Despite the charge separation at the  13  1.3. Solar Cells  Figure 1.6: Bilayer organic solar cell based on a layer of donor polymer and a layer of acceptor. interface, the efficiency is still low because of the relatively small length of the interface at the heterojunction which is limited to the areas close to the heterojunction interface. To overcome this issue, the donors and acceptors are blended to maximize the interface between them. The devices that are made by mixing the donor and acceptor are called bulk heterojunction (BHJ) devices. In order to fabricate an organic BHJ device, the donor and acceptor are mixed in the solvent and usually stirred on a hot plate for 12-36 hours. As shown in Fig. 1.7, the active layer of BHJ devices is spread across the whole length of these devices. As a result of the high donor and acceptor interface, there is a high chance of charge dissociation for excitons generated in the BHJ cells. Using this structure,  14  1.3. Solar Cells  Figure 1.7: Bulk heterojunction device maximized the interface between the donor and acceptor and increases the efficiency of the device. charge separation is not limited merely to the single interface of donor and acceptor and the chance of charge separation increases drastically. The highest performance of organic solar cells is reported by the use of this structure.  1.3.4  Efficiency Characteristics  J-V Characteristic Fig. 1.8 shows the generic J-V characteristic of a solar cell. The VOC and ISC are the short circuit current and open circuit current of the device. Based on this graph the following characteristics can be identified for the solar cells. Solar cells are operating between open-  15  1.3. Solar Cells Voc  J (mA/ cm2 )  Vm  Jm  Jsc Voltage (V)  Figure 1.8: J-V curve of the solar cells. The efficiency of the PV devices is the power conversion at maximum current and voltage point as shown. circuit and short-circuit condition (fourth quadrant in the currentvoltage characteristics), as shown in Fig. 1.8.  Maximum - Power Point The maximum-power point, is the point that maximizes J × V product, hence maximizes the power. That is, the load for which the cell can deliver maximum electrical power at that level of irradiation. In Fig. 1.8, Vm and Im shows the maximum power point. The power is zero at VOC and ISC . The proper load should be placed in order to achieve highest performance.  Fill Factor Fill Factor of the device is the ratio of the maximum power generated in the device and VOC × ISC . In another word, the ratio between the darkly shaded and brightly shaded  16  1.3. Solar Cells areas in Fig. 1.8.  FF =  Voc ×Isc VM ×IM  The overall efficieny of a solar cell can be expressed as follows:  PCE =  Voc ×Isc Pin  × FF  where Pin is the incident light power and FF is the fill factor of the devices.  Quantum Efficiency As described above, when a photon is absorbed by a solar cell it is converted to an electronhole pair. This electron-hole pair may then travel to the surface of the solar cell and contribute to the current produced by the cell; such a carrier is said to be collected. Alternatively, the carrier may give up its energy and once again become bound to an atom within the solar cell without reaching the surface; this is called recombination, and carriers that recombine do not contribute to the production of electrical current. Quantum efficiency refers to the percentage of photons that are converted to electric current (i.e., collected carriers). External quantum efficiency is the fraction of incident photons that are converted to electrical current, while internal quantum efficiency is the fraction of absorbed photons that are converted to electrical current. Mathematically, internal quantum efficiency is related to external quantum efficiency by the reflectance of the solar cell; given a perfect anti reflection coating, they are the same.  17  1.3. Bibliography  Bibliography [1] ArnulfJger-Waldau, PV Status Report, Joint Research Center, Renewable Energy Unit, 21, 2008. [2] W. Ma, C. Yang, X. Gong, K. Lee, A. J. Heeger, Thermally Stable, Efficient Polymer Solar Cells with Nanoscale Control of the Interpenetrating Network Morphology, Adv. Funct. Mater., 15:1617–1622, 2005. [3] S. Berson, R. Bettignies, S. Bailly, S. Guillerez, B. Jousselme, Elaboration of P3HT/CNT/PCBM composites for organic photovoltaic cells ,Adv. Func. Mat., 17, 3363, 2007. [4] A.J. Heeger et al., Nobel Prize in Chemistry, 2000. [5] D. Pulfrey, PV Photvoltaic Power Generation, Van Nostrand Reinhold, ISBN 0-44226640-5 , 1978. [6] BA. Gregg, Excitonic solar cells , J. Phys. Chem. B, 107:4688, 2003. [7] H. Shirakawa, E.J. Louis, A.G. MacDiarmid, C.K. Chiang, A.J. Heeger, Synthesis of Electrically Conducting Organic Polymers - Halogen Derivatives of Polyacetylene (CH)X, J Chem Soc Chem Comm () 579. [8] S. Karg, W. Riess, V. Dyakonov, M. Schwoerer, Elecrical and Optical Characterization of Poly(Phenylene-Vinylene) Light-Emitting-Diodes 54 (1993) 427. [9] RN. Marks, JJM. Halls , DDC. Bradley , RH. Friend , AB. Holmes, The Photovoltaic Response in Poly(Phenylene-Vinylene) Thin-Film Devices 6 (1994) 1379.  18  1.3. Bibliography [10] J. Peet, J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses, A. J. Heeger, G. C. Bazan, Efficiency Enhancement in Low-Bandgap Polymer Solar Cells by Processing with Alkane Dithiols, Nat. Mater. 6 (2007) 497. [11] BA. Gregg, MC. Hanna, Comparing Organic to Inorganic Photovoltaic Cells: Theory, Experiment and Simulation, J. Appl. Phys. , 93 (2003) 3605. [12] A. Haugeneder, M. Neges, C. Kallinger, W. Spirkl, U. Lemmer, J. Feldmann,U. Scherf, E. Harth, A. Gugel, K. Mullen, Exciton Diffusion and Dissociation in Conjugated Polymer/Fullerene Blends and Heterostructures, Phys. Rev. B, 59 (1999) 15346. [13] C. Woo, B. Thompson, B. Kim, M. Toney, J. Frechet, The Influence of Poly(3hexylthiophene) Regioregularity on Fullerene-Composite Solar Cell Performance, J. Am. Chem. Soc , 130 (48), (2008), 16324.  19  Chapter 2 The Effect of Annealing on High and Low Regiregular Polythiophene-Based Bulk Heterojunction Organic Photovoltaics 1  2.1  Brief  Organic photovoltaic (OPV) devices function based on photogeneration of excitons followed by separation of hole and electron at donor/acceptor heterojunction interfaces. The interA version of this chapter will be submitted for publication. Ebadian, S., Gholamkhas, B. and Servati, P. The Effect of Annealing on High and Low Regiregular Polythiophene-Based Bulk Heterojunction Organic Photovoltaics. 1  20  2.2. Introduction face between poly(3-hexylthiophene) (P3HT) and 6,6-phenyl C61 -butyric acid methyl ester (PCBM) is affected drastically by the fabrication method and especially the P3HT:PCBM drying time after active layer spin coating. The post-fabrication annealing, also, plays a major role in the performance of P3HT:PCBM based OPV devices. This work compares the effect of annealing on the performance of devices made by lower regioregular (RR) P3HT (94%) and higher RR P3HT (98%). We compare the efficiency, hole mobility, external quantum efficiency and light absorption as well as J-V curves in the two devices, before and after annealing. In general, devices based on higher RR P3HT are more prone to degradation as the nano-phase segregation between the donor and acceptor molecules is stronger due to its higher tendency for crystallization. The low RR device, on the other hand, demonstrates an increase in the hole mobility in post-annealed conditions as thermal annealing causes a more effective π - π stacking with lower rate of nano-phase separation. This leads to unchanged or even higher power conversion efficiency (PCE) of the devices with lower regioregularity after thermal annealing. Due to the same explanation, the low RR devices are found to be more stable during a four month degradation analysis while the efficiency of high RR device decreases drastically in the same period of time.  2.2  Introduction  2.2.1  OPV Devises  Since the introduction of conjugated polymer in 1977 [1], the OPV devices have emerged and improved from > 0.1% efficiency in 1993 [2, 3] to ∼ 4 − 5% of power conversion efficiency in 2005 [4, 5]. This enhancement is partly due to the improvements in de-  21  2.2. Introduction vice structure and partly due to the introduction of more stable conjugated polymers and molecules for development of OPVs. Introduction of a donor-acceptor structure [6] and fabrication of bilayer heterojunction device instead of single semiconductor polymers [2], increased the efficiency of OPV devices drastically. This bilayer structure enables charge dissociation in the polymer chain as a result of built-in field effect at heterojunction interface. Also, mixing donor and acceptor molecules and polymers and fabrication of a blend of bulk composite has increased PCE of the OPV devices. As Fig. 2.1 shows, the electrons in the polymer chain, become excited upon light absorption from the bonding π into π ∗ anti-bonding band. This absorption refers to the optical excitation from highest occupied molecular orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO) of donor. The donor-acceptor structure accelerates effective dissociation of generated excitons at the interface of donors and acceptors by internal field effect which should be > 106 V/cm in order to separate electron holes tightly bound (∼ 1 nm distance) by coulomb energy equal to 0.25 eV [7]. This dissociation is limited to a thin interface between donor and acceptors usually termed as the exciton diffusion length. The lower electron mobility in the conjugated polymer chain, requires a balanced dispersion of donor and acceptor in the PV’s active layer to enhance charge separation in bulk heterojunction (BHJ). This is a prominent factor for separating electrons and holes from excitonic state to free charge carriers. The donor-acceptor blend helps the electrons, which have lower mobility, to move effectively toward the cathode by hopping in the acceptor sites( ∼ 4 × 10−4 cm2 .V−1 .s−1 for holes in P3HT compared to ∼ 2 × 10−3 cm2 .V−1 .s−1 for PCBM) [8] . This increases the electron injection in the device considering the fact that short circuit current in polymer based solar cells is limited by the electron extraction in the devices [9].  22  2.2. Introduction  Light  LUMO LUMO  Al  Donor  Energy  Acceptor ITO  PEDOT HOMO :PSS HOMO  Heterojunction Surface  Figure 2.1: The energy diagram of the donor/acceptor charge separation resulting from the built-in charge at heterojunction interface. The electron and hole separation is further enhanced by PEDOT:PSS as it increases the Anodes work function resulting in a net increase in the anode/cathode work function difference.  The conjugated polymers also have improved drastically and several different polymers and conjugated molecules have been studied since the introduction of organic solar cells. Various combination of materials have been studied for fabrication of high efficiency and stable bulk heterojunction devices [10, 11]. This includes the polymer-polymer structures such as MEH-PPV [12] as well as fullerene based devices made of C60 derivatives such as PCBM [5].  23  2.2. Introduction  2.2.2  P3HT:PCBM Bulk Heterojunction Device  P3HT based OPV devices are the most promising and efficient device among polymer-based solar cells such as MDMO-PPV and PFB. This is despite the larger band gap of P3HT polymer (1.9 eV) in comparison with the solar spectrum peak at 1.8 eV. One reason for outstanding performance of P3HT based devices is the efficient formation of the P3HT and PCBM mixture and the relatively uniform dispersion of PCBM in the polymer structure [13, 14]. The highest PCE among all OPV devices is reported for P3HT:PCBM BHJ devices [4]. Several different fabrication methods have been studied for P3HT:PCBM based devices. This includes several different solvents and co-solvents ranging from dichlorobenzene(DCB) and chlorobenzene(CB) to chloroform(CHL). Several different annealing methods have been studied to increase polymer chain crystallization on P3HT:PCBM devices [15, 16]. Also several different weight ratios of P3HT and PCBM as well as solution concentrations have been examined to optimize device performance. Considering all these experiments and background works mentioned above, we have used a 1:1 wt % P3HT:PCBM (20 mg of each)dissolved in 1mL of DCB. Although some groups have reported better results with 1:0.7 and 1:0.95 ratio of P3HT:PCBM [21, 22] , we have noticed a better efficiency for 1:1 ratio of polymer and fullerene comparing to the devices made with other wt rations. To dry out the active layer after spin coating, several routines have been employed, ranging from a slow, room temperature processes, to thermal annealing in a vacuum oven. The devices made at room temperature and during a slow annealing process show a higher performance [17]. This is because the slow solvent-annealing leads to a more crystalline P3HT structure compared to fast annealing [17]. Moreover, devices that are annealed during a slow annealing process have a higher optical density which leads to a higher light absorption, hence a higher external quantum efficiency (EQE). High efficiency devices, also  24  2.2. Introduction employ a thin layer of Ca or LiF before evaporation of Al which increases the Cathode’s work function. This leads to an increase in open circuit voltage and fill factor of the device. In this work we did not use Ca or LiF because we focus on comparing the effect of further thermal annealing on devices made by low and high RR P3HT polymer. Additionally, we wanted to eliminate the effect of Ca oxidation in comparison between the degradation of low and high RR devices over a period of 4 months. It should be noted that Al oxidation is much slower than that of Ca. P3HT:PCBM based solar cells exhibit a variety of different responses in the thermal annealing process. The efficiency and short circuit current is reported in many papers to increase as a result of thermal annealing [15, 16]. The solvent and the regioregularity of the P3HT chain used for fabrication of OPV devices, seem to have an eminent role in the performance of devices before and after thermal annealing. In devices fabricated by CB, the efficiency and short circuit current increases as devices are annealed at 140 ◦ C for 1 hr [20]. In this work we have fabricated P3HT:PCBM OPV devices with two instances of P3HT polymer with high (98%) and low (94%) regioregularity dissolved in DCB. The devices are measured for J-V characteristic, external quantum efficiency and optical density and the power conversion efficiency and hole mobility for each device is calculated. These parameters are also calculated for the same devices annealed at high temperature. Devices are also characterized during a 4-month period for efficiency.  Through this work, we have analyzed and compared the characteristics of devices made by high and low RR P3HT before and after thermal annealing and discussed the effect of regioregularity on the pre and post annealed devices. We have also compared the device  25  2.3. Experiment  +V-  Al  Al  P3HT:PCBM Al PEDOT:PSS ITO Glass  Figure 2.2: The cross section of the bulk heterojunction OPV device on ITO coated glass. The PEDOT:PSS increases the work function of the anode, leading to a better charge separation.  degradation and efficiency decline in these devices with respect to the regioregularity of the samples.  2.3 2.3.1  Experiment Photovoltaic Device Fabrication  Indium tin oxide (ITO) (Delta Technologies, 25 Ω/sq) is used as the transparent electrode. The ITO is first masked by Kapton tape (DuPont) and etched from the middle of the substrate by using a solution of 20 wt% HCl and 5 wt% HNO3 and water. It is cleaned subsequently in acetone, iso-propanol and water and wiped out with kimwipe to remove all glass particles remained from cut. The substrate is then sonicated in iso-propanol for 20 minutes at 50 ◦ C followed by acetone sonication at 60 ◦ C and in a solution of ammonia and water. This step decreases surface energy of the substrate and increases its wetting properties. The substrate is then sonicated in water for 20 minutes followed by oxygen  26  2.3. Experiment plasma cleaning for 15 minutes. Plasma cleaning further changes the surface energy by increasing density of oxygen bound onto the surface. PEDOT:PSS (Clevios P VP Al 4083, HC. Stark) filtered through 0.45 µm PP syringe filter is spin coated on the substrate at 4000 rpm resulting in a ∼ 50 nm layer, measured by Fimetrics thickness measurement device. The PEDOT:PSS then annealed in the oven at 140 ◦ C for 10 min. For the semiconductor layer, P3HT purchased from Rieke Metals ( > 98% for high RR with Mw = 64, 000 g/mol and ∼ 94% for low RR with Mw = 50, 000 g/mol) and PCBM (purchased from Aldrich) were used as supplied. Solution was prepared in the golvebox by mixing a 1:1 wt ratio of P3HT and PCBM in DCB (Aldrich Anhydrous 99%). About 20 mg of P3HT and PCBM were separately dissolved in 500 µL of solvent and were stirred on the hot plate at 50 ◦ C for 20 minutes.The two solutions are then mixed and the total P3HT:PCBM is further stirred and heated over night at 40 ◦ C. The solution was then filtered through 0.45 µm PP Syringed filter. On average a 120 nm of P3HT:PCBM is spin coated at 200 rpm for 5 seconds and at 1000 rpm for 15 seconds for a uniform deposition of active layer. The layer is thicker for high RR P3HT samples (130 nm vs. 115 nm for low RR). The layer is then dried slowly over 45 minutes under a petri dish. Slow drying of the semiconductor polymer layer allows the P3HT polymer to crystallize and stack more effectively for a better electronic conduction [10]. This results in an OPV device with darker color, in comparison with the devices made by fast drying and thermal annealing. The darker color of the semiconductor layer is a key factor in achieving a better efficiency for the OPV devices [15, 20]. Device structures were completed by evaporation of an 80 nm thick Al as cathode layer under a vacuum of 2×10−6 Torr [21, 22]. After characterizing the devices to measure pre-annealing conditions, all cells are thermally annealed for 30 minutes at 140 ◦ C and are characterized for post-annealing conditions. Devices are all kept in the glovebox and are periodically tested for overtime degradation characteristics.  27  2.4. Results and Discussion  2.3.2  Measurements  The PCE was calculated from the current density−voltage (J−V) characteristics under air mass 1.5 solar simulated light irradiation of 100mW/cm2 . JV characteristics were measured by the semiconductor characterization system (KEITHLEY 4200, Keithley Co. Ltd.). Devices are directly transfered from N2 filled glovebox to a home made chamber with vacuum outlet. The chamber’s light absorption is calibrated and essential corrections are made for accordance to AM1.5 solar spectrum and power. For EQE measurements, the cells were illuminated with monochromatic light of 0.5 − 1mW/cm2 . The wavelength of the incident light was changed by computer-controlled monochromator. Film thicknesses were measured by Filmetrics F20-UVX and a Alpha Step 200 profilometer.  2.4  Results and Discussion  Three samples of devices with low and high RR P3HT were fabricated in the glove box. The active area of each device is 0.104 cm2 . Fig. 2.3 shows the J-V characteristic of the fabricated devices, before annealing. The fugure also contains post-annealing results after thermal treatment at 140 ◦ C for 30 min. As the Fig. 2.3 depicts, the high RR device has the highest short circuit current among all samples. Higher Isc is due to higher molecular weight of high RR P3HT sample [19]. As shown, the Isc and Voc for the high RR device reduces steeply after annealing. As a result of this current and voltage drop, the efficiency of high RR device decreases by 39 % from original 2.5% to 1.52% after thermal annealing. Table.1 shows the Isc and Voc , PCE and fill factor for each device.  28  2.4. Results and Discussion 0.1 0.2 High RR before Annealing High RR after Annealing Low RR before Annealing Low RR after Annealing  0.3  0.4  0.5  0.6  Current Density (mA/ cm2 )  0  -6.75  -9  Voltage (V)  Figure 2.3: The J-V curves for high and low RR devices before and after annealing. As the graph shows, the annealing has a minor effect on the Isc and Voc of the low RR device while affecting those of high RR device drastically.  The low RR device is quite stable before and after annealing. As Fig. 2.3 and Table.1 show, the Voc remains the same for low RR device before and after annealing while there is a slight change in Isc after annealing. In total, the efficiency of devices made with low RR P3HT is almost unchanged or increased in some samples. The Fill factor (FF) is also affected in the annealing process. The FF of the high RR device decreases from 52% to 45% while the FF in the low RR device does not change or even increases slightly ( ∼ ±1 − 2%). Although the low RR device is stable during the annealing process, its PCE is on average 10% lower than the high RR device. Of the three samples made for each category of devices ( 3 for low RR and 3 for high RR P3HT), the efficiency decreased in all high RR devices while that of low RR devices either increases slightly or is unchanged. The average of efficiency for high RR devices is 2.5% before annealing and 1.41% after annealing. Meanwhile, for devices made by a low RR P3HT sample, PCE is 2.27% before and 2.25% after annealing, which demonstrates a higher stability and lower degradation  29  2.4. Results and Discussion Table 2.1: Device Characteristic for Different regioreguarity and Annealing schemes 1) high RR before annealing 2) high RR after annealing 3) Low RR before annealing 4) Low RR after annealing Device Isc (mA) Voc (V) PCE(%) FF(%) 1 9.3 0.54 2.5 52 2 6.75 0.44 1.52 45 3 8.82 0.54 2.29 48 4 8.29 0.55 2.22 50 2  Wavelength/nm  Absorption (A.U.)  1.6 1.2 0.8 0.4 350  Low RR P3HT:PCBM Device High RR P3HT:PCBM Device 440 525  613  700  Figure 2.4: Absorption spectrum for P3HT:PCBM film after annealing for high and low RR P3HT. The absorption for 98% RR device is red-shifted in comparison to the absorption profile of cells fabricated based on 94% RR P3HT . This graph shows a higher optical density for high RR P3HT, which is in contradiction with EQE analysis as a result of higher phase-segregation in high RR device.  in these devices. These findings are in contradiction with previous works that show the increase in performance and characteristics of P3HT:PCBM devices after annealing [20]. This is mostly due to the use of CB as the solvent which has a lower boiling point and causes a faster drying of the active layer. This fast drying leads to less effective crystallization which may be enhanced by further thermal annealing [10].  Fig. 2.4 shows the UV-Vis result of the low and high RR P3HT coated substrates. Al-  30  2.4. Results and Discussion though the post annealing results show a higher efficiency for devices made by low RR P3HT, the UV-Vis result demonstrates higher optical density in 98% RR P3HT sample after annealing. As shown, absorption is higher for most of the visible wavelengths. The absorption profile in the devices fabricated with high RR P3HT is also red-shifted in comparison to the low RR counterparts. Despite higher optical absorption and the tendency toward higher crystallinity in the high RR device, the low RR samples have comparable efficiency and much better stability after annealing and over the period studied. This is because of higher phase-segregation between donor and acceptor molecules in high RR devices. The tendency in higher RR samples for π-π stacking at polymer plains, pushes the PCBM out of the boundary of P3HT sites and decrease the interface between donor acceptor molecules, essential for charge separation in the structure [25].  This finding is also supported by the Monte Carlo simulation results of bulk heterojunction (BHJ) devices [23, 24]. Based on this model, there is an optimum phase separation for BHJ devices. Larger or smaller domains of donor and acceptor decreases the efficiency, the mobility in the device and the effective interface between donor and acceptor, resulting in a weak built-in field, essential for charge separation. This nano-scale phase separation is accelerated by the thermal annealing for high RR device while helps the low RR device to crystallize more efficiently. The EQE results for both cases, before and after annealing is in accordance with these findings shown in Fig 2.4 and 2.5. The high RR device has higher EQE in all wavelengths while EQE decreases for this sample drastically after annealing. EQE curves are mostly similar for low RR device before and after annealing. Additionally comparison between  31  2.4. Results and Discussion 70  Wavelength/nm  52.5 EQE (%)  35  17.5  300  High RR before annealing High RR after annealing Low RR before annealing Low RR after annealing 425 550  675  800  Figure 2.5: The EQE curves based on different wavelengths for high and low RR devices before and after annealing. EQE drops at high RR P3HT while is almost similar for low RR cells.  UV-Vis results of post annealed samples with EQE results of the same samples, shows that optical density of higher crystalline device does not lead to a higher external quantum efficiency which we argue is because of higher phase segregation between donor and acceptor sites [25]. EQE in all four cases drops to ∼ 2% at 675 nm.The high RR P3HT cell, as depicted, has the highest EQE which corroborates the PCE results. The high RR curve also has two shoulders at 360 and 630nm. The annealed cell with high RR follows the same pattern as the non-annealed one with a 10% decrease on average in EQE. On the other hand, the low RR device shows a similar EQE curve in most of the wavelengths with a slight increase in 500-580 nm for the post annealed device. Because of the slow drying phase in the fabrication of both devices and use of DCB as the solvent, we expect to have effective crystal formation of P3HT polymer with an efficient π-π stacking between polymer plains in the structure. This leads to a semi-uniform dispersion of PCBM sites in the structure. As the built-in field in the device is limited to the difference  32  2.4. Results and Discussion 4  LOG (V) High RR before annealing High RR after annealing Low RR before annealing Low RR after annealing  LOG (J)  ×10−4 Cm2 /V.s 3.93  0  3.19  High RR Low RR  2.35  Pre Annealing  -3 -1  -0.65  -0.31  0.03  Post Annealing  0.375  Figure 2.6: Log V - Log J diagram for hole mobility calculation in high and low RR devices before and after annealing. The intercept with J axis at the region with slope equal to 2 results the mobility in accordance with constant part of logarithm of Eq. 1.  between HOMO and LUMO of the donor acceptor, the geometry of device and placement of P3HT chain and PCBM molecules becomes eminent. In the high RR device the tendency for a more crystalline structure, forces the PCBM molecules to phase segregate. On the other hand, lower RR device, which has less effective π-π stacking, is likely forming a more crystalline structure. The non-ideal crystallization in low RR device, allows PCBM molecules to embed in the polymer structure without a major large-scale phase separation in this condition. As the crystallization continues in this case, the hole mobility increases which leads to slight increase in the PCE of the device. To justify these assumptions, we have also studied the hole mobility in both cases before and after annealing, based on the space-charge limiting current theorem (SCLC) [9, 26]. Based on SCLC, the unbalanced hole and electron mobility causes an excessive amount of holes in the device to be piled at the anode, causing an increase in the anode-cathod potential. The SCLC is calculated from equation 1.  33  2.4. Results and Discussion  JSCL =  9 V2 µ 0 r h 8 L3  Eq.(1)  Fig. 2.6 shows the Logarithmic plot of current density - potential in each of the four devices. Based on the above, the hole mobility for each of the four devices is calculated. As the inset of Fig. 2.6 shows, hole mobility decreases for the high RR P3HT device from 3.93 ×10−4 cm2 .V−1 .s−1 for pre annealed sample to 2.35 ×10−4 cm2 .V−1 .s−1 in post annealed device. Meanwhile, the hole mobility increase for the low RR device from 3.19 ×10−4 cm2 .V−1 .s−1 in pre annealed sample to 3.74 ×10−4 cm2 .V−1 .s−1 in post annealed device. These results confirm the efficiency analysis above. The hole mobility in high RR device decrease because of the defective effect of PCBM larger crystals in the structure and decreasing the alignment between separate P3HT sites in the device. This results in lower PCE in the device. On the contrary, the formation of a more crystalline polymer in low RR device along with effective nano scale interface between P3HT and PCBM in this device, leads to a higher hole mobility in low RR device. As a result of higher hole mobility, the PCE of the device increases. This is despite lower absorption in low RR device as shown in Fig. 2.4.  The results of this work shows a discrepancy with the efficiency of devices fabricated by using CB as solvent. W. Ma et al. report a thermally stable P3HT:PCBM device that demonstrates highest efficiency at 150 ◦ C [4]. This is due to the difference in the formation of polymer crystal structure. The use of CB as the solvent (boiling point = 130 ◦ C) causes the active layer to effectively dry out after spin-coating and solvent evaporation happens much faster. Therefore, the uniform and complete π-π stacking of the polythiophene back-  34  2.4. Results and Discussion Days Low RR P3HT:PCBM Device High RR P3HT:PCBM Device  2.7 2.05  PCE (%)  1.4 0.75 0.1 0  40  80  120  Figure 2.7: PCE degradation atlow and high RR device in a 4-month period. Low RR device is more stable and degrades with a lower rate in comparison to the high RR device.  bone does not take place efficiently. In their work, the crystallization of P3HT polymer continues by further annealing which leads to a higher efficiency devices. On the contrary, after spin-coating of the DCB (boiling point = 190 ◦ C) the solvent slowly evaporates and gradually forms the P3HT polymer chain. This causes the efficiency to either remain stable (in low RR devices) or decrease drastically ( in high RR devices) as phase segregation becomes dominant factor.  The phase separation continues after the fabrication. This phenomenon, causes further degradation in both devices. Although the devices are kept in the glove box under N2 and tested periodically, they both demonstrate a drop in PCE. It should be noted that low RR device show a more stable PCE over time in comparison with high RR device. The efficiency of high RR device drops from 2.7% to 0.2% while the low RR device becomes stable after an early decline in the first 20 days. The efficiency of low RR device decrease to 1.9% from the 2.4% after 120 days. Fig. 2.7 shows the PCE changes in a 4 month period  35  2.5. Conclusion as devices degrade.  2.5  Conclusion  The effect of annealing on the performance of device is highly related to the fabrication method, solvent used for mixing P3HT:PCBM and more importantly, regioregularity of P3HT used. Using DCB as the solvent and slow drying of the substrate causes a rather efficient crystallization of the polymer structure. This results in higher PCE devices in comparison with devices that dry fast. Devices made by this method demonstrate different aging characteristics depending on the regioregularity of P3HT used. If the P3HT used in fabrication of the device, is of higher RR ( 98% in this experiment) the short circuit current and open circuit voltage drop drastically, as the crystallinity of the P3HT leads to separation of PCBM molecules from the the nano scale interface in the bulk heterojunction. This decreases the built-in field caused by the energy difference between the HOMO and LUMO of donor and acceptor. This built-in field is the major reason for electron-hole separation of the excitons. On the other hand, the low RR device shows a stable or even increased efficiency as the crystallization continues in this structure but does not force PCBM molecules to diffuse and make larger crystals. The same phenomenon takes place as the device ages and accelerates the aging of high RR device while the low RR device remains more stable in comparison with the high RR device. The efficiency of low RR device drops in the first 20 days and remains almost stable for the first 120 days while the high RR device demonstrates a drastic drop in the efficiency from 2.7% to 0.2% over a 120 days period. It is a delicate matter to select proper regioregularity for the OPVs as the high RR P3HT shows a better PCE in the beginning as a matter of higher optical density  36  2.5. Conclusion and hole mobility while devices made by lower RR P3HT, remains more stable over longer period of time. The trade-off between crystallinity and phase segregation leads to a device with optimum levels of efficiency as well as stability.  37  2.5. Bibliography  Bibliography [1] H. Shirakawa, E.J. Louis, A.G. MacDiarmid, C.K. Chiang, A.J. Heeger, Synthesis of Electrically Conducting Organic Polymers - Halogen Derivatives of Polyacetylene (CH)X, J Chem Soc Chem Comm (1977) 579. [2] S. Karg, W. Riess, V. Dyakonov, M. Schwoerer, Electrical and Optical Characterization of Poly(Phenylene-Vinylene) Light-Emitting-Diodes ,54 (1993) 427. [3] RN. Marks, JJM. Halls , DDC. Bradley , RH. Friend , AB. Holmes, The Photovoltaic Response in Poly(P-Phenylene Vinylene) Thin-Film Devices, J. Phys-Con Matter. , 6 (1994) 1379. [4] W. Ma, C. Yang, X. Gong, K. Lee, A. J. Heeger, Thermally stable, efficient polymer solar cells with nanoscale control of the interpenetrating network morphology ,Adv. Funct. Mater. 15 (2005), 1617. [5] J. Peet, J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses, A. J. Heeger, G. C. Bazan, Efficiency enhancement in low-bandgap polymer solar cells by processing with alkane dithiols ,Nat. Mater. 6 (2007) 497. [6] N. S. Sariciftci, D. Braun, C. Zhang, V. I. Sardanov, A. J. Heeger, F. Wudl, Semiconducting Polymer-Buckminsterfullerene Heterojunctions - Diodes, Photodiodes, and Photovoltaic Cells,Appl. Phys. Lett. 62 (1993) 585. [7] B. A. Gregg, M. C. Hanna, Comparing organic to inorganic photovoltaic cells: Theory, experiment, and simulation, J. Appl. Phys. 93 (2003) 3605.  38  2.5. Bibliography [8] V. D. Mihailetchi, J. K. J. van Duren, P. W. M. Blom, J. C. Hummelen, R. A. J. Janssen, Electron transport in a methanofullerene, Adv. Funct. Mater. 3 (2003) 43. [9] V. D. Mihailetchi,, J. Wildeman, P. W. M. Blom, Space-charge limited photocurrent,Phys. Rev. Lett. 94 (2005) 126602. [10] NS. Sariciftci, L. Smilowitz, AJ. Heeger, F. Wudl, Photoinduced Electron-Transfer from a Conducting Polymer to Buckminsterfullerene,Science 258 (1992) 1474. [11] CR. McNeill , A. Abrusci, J. Zaumseil, R. Wilson, MJ. McKiernan, Dual electron donor/electron acceptor character of a conjugated polymer in efficient photovoltaic diodes ,Appl. Phys. Lett. 8 (2006) 3557. [12] G. Yu, J. Gao, JC. Hummelen, F. Wudl, AJ. Heeger, Polymer Photovoltaic Cells - Enhanced Efficiencies via a Network of Internal Donor-Acceptor Heterojunctions ,Science 270 (1995) 1789. [13] B. Thompson, Y. Kim, T. McCarley, J. Reynolds, Soluble narrow band gap and blue propylenedioxythiophene-cyanovinylene polymers as multifunctional materials for photovoltaic and electrochromic applications ,J. Am. Chem. Soc. 128 (2006) 12714. [14] M. Scharber, D. Muhlbacher, M. Koppe, P. Denk, C. Waldauf, A. Heeger, C. Brabec, Design rules for donors in bulk-heterojunction solar cells - Towards 10 % energyconversion efficiency ,J. Adv. Mater 18, (2006) 789. [15] G. Li, V. Shrotriya, Y. Yao, Y. Yang, Investigation of annealing effects and film thickness dependence of polymer solar cells based on poly(3-hexylthiophene), J. of Appl. Phys. 98 (2005) 043704.  39  2.5. Bibliography [16] T. Yamanari, T. Taima, K. Hara, K. Saito, Investigation of optimum conditions for high-efficiency organic thin-film solar cells based on polymer blends ,J. of Photochemistry and Photobiology A: Chemistry 182 (2006) 269. [17] S.-H. Jin, B. Vijaya, K. Naidu, H. S Jeon, Optimization of process parameters for high-efciency polymer photovoltaic devices based on P3HT:PCBM system,Solar Eng. & Mat. & Sol. Cel. 91 (2007) 1187. [10] G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery, Y. Yang, Investigation of annealing effects and lm thickness dependence of polymer solar cells based on poly3hexylthiophene,Nat. Mater. 4, (2005), 864. [19] P. Schilinsky, U. Asawapirom, U. Scherf, M. Biele, C. J. Brabec, Influence of the molecular weight of poly(3-hexylthiophene) on the performance of bulk heterojunction solar cells ,Chem. Mater. 17, (2005), 2175. [20] T. Clarke, A. Ballantyne, J. Nelson, D. Bradley,J. Durrant, Free Energy Control of Charge Photogeneration in Polythiophene/Fullerene Solar Cells: The Influence of Thermal Annealing on P3HT/PCBM Blends ,Adv. Funct. Mater. , 18, (2008), 4029. [21] K. Sylvester-Hvid, T. Ziegler, M. Riede, N. Keegan, M. Niggemann, A. Gombert, Analyzing poly(3-hexyl-thiophene): 1-(3-methoxy-carbonyl)propyl-1-phenyl-(6,6)C61 bulk-heterojunction solar cells by UV-visible spectroscopy and optical simulations ,J. App. Phys. 102, (2007), 054502. [22] M. Reesea, Anthony, J. Morfaa, M. Whitea, N. Kopidakisa, S. Shaheenb, G. Rumblesa, D. Ginleya, Solar Eng. & Mat. & Sol. Cel. 92 (2008) 746.  40  2.5. Bibliography [23] R.A. Marsh, C. Groves, N. C. Greenham, A microscopic model for the behavior of nanostructured organic photovoltaic devices ,J. Appl. Phys. 101 (2007) 083509. [24] C. Groves, R.A. Marsh, N. C. Greenham, Monte Carlo modeling of geminate recombination in polymer-polymer photovoltaic devices ,J. Chem. Phys. 129 (2008) 114903. [25] C. Woo, B. Thompson, B. Kim, M. Toney, J. Frechet, The Influence of Poly(3hexylthiophene) Regioreigularity on Fullerene-Composite Solar Cell Performance ,J. Am. Chem. Soc., 130 (48), (2008), 16324. [26] A. M. Goodman and A. Rose, Double Extraction of Uniformity Generated ElectronHole Pairs from Insulators with Non-injecting Contacts ,J. Appl. Phys. 42 (1971) 2823.  41  Chapter 3 Reducing the Roughness of Transparent Electrodes in Organic Photovoltaic Devices on Plastic Substrate by PEDOT:PSS Treatment 2  3.1  Brief  This work presents our results on surface treatment of flexible polyethelene terephthalate (PET) substrates for fabrication of organic photovoltaic devices (OPVs). The PET A version of this chapter has been published. Ebadian, S. and Servati, P. Reducing the Roughness of Transparent Electrodes in Organic Photovoltaic Devices on Plastic Substrate by PEDOT:PSS Treatment. Nano Material and Devices Conference Proceeding:221-224. 2  42  3.2. Introduction substrates coated with transparent electrodes (such as indium oxide, Au and Ag) have a surface roughness with a high amplitude and spatial frequency, resulting in low device yield due to the high chance of short circuit between the top and bottom electrodes. Deposition of a single layer of poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) does not suppress the roughness amplitude and the microscopic images show a non-uniform and defective PEDOT:PSS layer. This work presents a double layer method that reduces the surface roughness drastically, thus improving the yield and uniformity of poly 3-hexylthiophene (P3HT) and [6,6]-phenyl-C61 -butyric acid methyl ester (PCBM) based photovoltaic devices made on PET substrates.  3.2  Introduction  Solar energy is abundant and clean. However, today’s photovoltaic devices (mainly based on silicon and polysilicon) used for conversion of solar energy to electricity are expensive and inefficient (less than 9% conversion efficiency) [1]. Organic and polymeric photovoltaic devices are attractive due to their lower manufacturing cost and the possibility of room temperature processes (e.g. spin casting, spraying and printing) that allows use of cheap unconventional substrates such as plastic [2]. Conjugated polymers, also, have lower weight and are soluble in a wide range of organic solvents. Therefore, they can be printed, spin coated or even sprayed on flexible substrates such as PET. The fabrication methods on flexible substrates are more complicated as the Indium Tin Oxide (ITO) coated plastic films are an order of magnitude rougher than ITO coated glass [3]. Especially the high spatial frequency of surface roughness makes fabrication of flexible devices more prone to defects and pin holes, leading to a higher probability of short circuit between the top  43  3.2. Introduction  Figure 3.1: OPV device on PET substrate. P3HT:PCBM is used as the semiconductor layer. and bottom electrodes. To increase the hole injection in organic solar cells, a layer of PEDOT:PSS is often deposited to increase the Anode’s work function, as shown in Fig. 3.1. This layer, also, makes the ITO coated substrate smoother and more suitable for organic semiconductor deposition. A blend of P3HT:PCBM is then spin coated on the PEDOT:PSS. The P3HT and PCBM, which are called donor and acceptor, respectively, form nano scale heterojunctions that enable effective dissociation of light charged excitons [4]. The PCBM molecules attract the electron in the charge separation process, preventing the electron hole pairs from recombination [5]. The separated electrons and holes then move between the PCBM molecules and in the P3HT structure respectively. Although subsequent coating of the transparent electrode with PEDOT:PSS, reduces the spatial frequency of the surface roughness, the spin coating of this conductive polymer does not drastically reduce the amplitude of roughness and microscopic image shows a  44  3.3. Experiment non-uniform and uneven surface after deposition of a single layer of PEDOT:PSS. In this work, we present a method that overcomes this issue and results in a uniform and smooth layer of PEDOT:PSS that prevents defects in OPV devices and increases the device yield.  3.3  Experiment  The PET coated with indium oxide (In2 O3 )/Au/Ag (delta technologies, 10 Ω/square) is used as the transparent electrode. It is cleaned subsequently in iso-propanol and water. The substrate is then sonicated in iso-propanol for 20 minutes at 40 ◦ C. Sonication in iso-propanol decreases surface energy of the substrate and increases its wetting properties. The substrate is then sonicated in water for 20 minutes followed by oxygen plasma cleaning for 15 minutes. Plasma cleaning further changes the surface energy by increasing density of oxygen bonds on the surface. This layer is then modified by PEDOT:PSS layer (Clevios P, HC. Stark) diluted with 4 wt% iso-propanol and filtered through 0.45 µm PP syringe filter. PEDOT:PSS is spin coated on the substrate at 4000 rpm resulting in a 170 nm layer, measured by Fimetrics thickness measurement device. Before letting the PEDOT:PSS dry, the substrate is rinsed in DI water for several minutes until the layer is wiped out completely. The substrate is then dried using a Nitrogen gun. As the microscope image in Fig. 3.2 shows, the substrate after 1-layer deposition is uniformly covered with PEDOT:PSS (Fig. 3.2a). Fig. 3.2b, on the other hand shows that the PEDOT:PSS is not wiped out uniformly after rinsing with DI water. The darker parts depict the residues of the polymer that result in a smoother surface for another  45  3.3. Experiment 3nm  −6nm  1µm  a  b Fig. 2. The microscope image of the film a after PEDOT:PSS rinse with DI water. Conductive polymer remains in some spots as depicted in the figure by dark color.  0  +6nm  Figure 3.2: The microscope image of the film a) after one layer deposition and before rinse b) after PEDOT:PSS rinse with DI water. Conductive polymer remains in some spots as depicted in (b) by color. (a) shows the 1-layersurface deposition ◦ at 40dark C. Sonication in iso-propanol decreases energy of PEDOT:PSS. of the substrate and increases its wetting properties. The  a  1  −6nm  1µm  substrate is then sonicated in water for 20 minutes followed by oxygen plasma cleaning for 15 minutes. Plasma cleaning PEDOT:PSS layer deposition. further changes the surface energy by increasing density of oxygen bonds on the surface. This layer is then modified by PEDOT:PSS layer (Clevios P, HC. Stark) diluted with 4 1 0 c At this point another layer of PEDOT:PSS is spin coated on the surface 3500 wt% iso-propanol and filtered through 0.45atµm PP rpm. syringeThis filter. PEDOT:PSS is spin coated on the substrate at 4000 rpm ◦ C. layer is dried on a covered hot plateresulting for an hour at nm 85 layer, in a 170 measured by Fimetrics thickness Fig. 3. AFM analysis of PEDO (a) Film 1, before deposition and measurement device.  Film 2, deposition of the first PE  letting the PEDOT:PSS dry, the substrate is rinsed water for several minutes, (d) Film For the semiconductor layer, P3HT Before and PCBM purchased from Aldrich were used as and annealing. in DI water for several minutes until the layer is wiped out completely. The substrate is then dried by Nitrogen gun. As supplied. Cosolvents were prepared in the golvebox by mixing the dichlorobenzene (Aldrich the microscope image in Fig. 2 shows, the PEDOT:PSS is III. M not wiped out uniformly. The darker parts depict the residues Anhydrous 99%) in 1:1 wt P3HT:PCBM ratio. The solution was stirred and heated for 48Photovoltaic performanc of the polymer that result in a smoother surface for another a calibrated AM1.5 solar PEDOT:PSS layer deposition. ◦ hours at 40 C and was filtered through 0.45 µm PP Syringed filter. A 120 nm of active mW/cm2 . Current-Voltage At this point another layer of PEDOT:PSS is spin coated using a Keithley 4200-SCS the surface at 3500 rpm. Thisfor layer dried on aThe covered layer is spin coating at 200 rpm for 5onseconds and at 800 rpm 15 isseconds. layer by is Filmetrics F20-UVX an hot plate for an hour at 85 ◦ C. sured in a vacuum probe −6 then dried slowly for 45 minutes under dish. Slow of PCBM the semiconductor Forthe the petri semiconductor layer, drying P3HT and purchased and at 3 × 10 Torr. from Aldrich were used as supplied. Cosolvents were prepared polymer layer allows the P3HT polymer crystallize and stack more effectively(Aldrich for better IV. R ESULT in thetogolvebox by mixing the Dicholorobenzene Anhydrus 99%) in 1:1 wt P3HT:PCBM ratio. The solution Figure 3 shows the atom conduction [10]. This results in an OPV device color,atin with the ◦ was stirred andwith heateddarker for 48 hours 40comparison C and was filtered for four different films. F through 0.45 µm PP Syringed filter. A 120 nm of active layer PET, after sonication in devices made by fast drying. The darker keyfor factor in is spincolor coatingofatthe 200semiconductor rpm for 5 seconds layer and at is 800a rpm 15 DOT:PSS deposition. Film seconds. The layer is then dried slowly for 45 minutes under PEDOT:PSS film. The laye achieving a better efficiency for the the OPV devices [6, 7]. Device structures werepolymer completed petri dish. Slow drying of the semiconductor 3 is the result of rinsing fi layer allows the P3HT polymer to crystalize and stack more and Film 4 is after spin co effectively for better conduction. This results in an OPV device PEDOT:PSS deposition. 46As seen for Film 1, the P with darker color, in comparison with the devices made by fast drying. The darker color of the semiconductor layer is a key high frequency and ampli factor in achieving a better efficiency for the OPV devices of roughness is between [7], [6]. Device structures were completed by evaporation of of 8 × 107 m−1 . This ro an 80 nm thick Al as cathode layer under a vacuum of 2×10−6 uneven and non-uniform d Torr. Devices are then annealed for 1 hour at 45 ◦ C inside the 2. Although the roughness golvebox [8], [9]. PET substrate, the amplitud  3.4. Measurements by evaporation of an 80 nm thick Al as cathode layer under a vacuum of 2 × 10−6 Torr. Devices are then annealed for 1 hour at 45 ◦ C inside af golvebox [8, 9].  3.4  Measurements  Photovoltaic performance of the devices were measured by a calibrated AM1.5 solar simulator (150W Newport) at 100 mW/cm2 . Current-Voltage (J-V) characteristics were recorded using a Keithley 4200-SCS. Film thicknesses were measured by Filmetrics F20-UVX and a profilometer. Devices are measured in a vacuum probe station at room temperature at 25 ◦ C and at 3 × 10−6 Torr.  3.5  Results and Discussions  Fig. 3.3 shows the atomic force microscopy (AFM) image for four different films. Film 1 is the In2 O3 /Ag/Au coated PET, after sonication in Propyl Alcohol and prior to PEDOT:PSS deposition. Film 2 is the result of deposition of first PEDOT:PSS film. The layer is spin coated at 4000 rpm. Film 3 is the result of rinsing film 2 in water for several minutes and Film 4 is after spin coating and annealing of the second PEDOT:PSS deposition. As seen for Film 1, the PET and transparent electrode have high frequency and amplitude of roughness. The amplitude of roughness is between ±3 nm with a spatial frequency of 8 × 107 m−1 . This roughness is the major reason for uneven and non-uniform deposition of PEDOT:PSS in Film 2. Although the roughness in Film 2 is much lower than the PET substrate, the amplitude of film roughness is still high for deposition of the P3HT:PCBM  47  3.5. Results and Discussions  3nm  +6nm  −6nm  −6nm  1µm  0  +6nm  1µm  a  1µm  +6nm  −6nm  b  1µm  −6nm  1µm  0  0  1µm  c  1µm  0  d  1µm  Figure 3.3: AFM analysis of PEDOT:PSS Films deposited on PET substrates: (a) Film 1, before deposition and after sonication in iso-propyl alcohol, (b) Film 2, deposition of the first PEDOT:PSS, (c) Film 3, rinsing Film 2 with water for several minutes, (d) Film 4, after deposition of second PEDOT:PSS and annealing. 48  3.5. Results and Discussions layer. Devices deposited on this layer are less likely to work and have a higher probability of short circuit between the top and bottom electrodes. The lower yield of devices made by this method and the linear characteristic of bad devices proves this fact. In addition, the P3HT:PCBM layer does not become thin enough due to the roughness of the substrate, affecting the device performance. The PEDOT:PSS layer here is 170 nm thick and this would not significantly change even at higher spin coating speeds. Film 3 results after rinsing Film 2 in water for several minutes. As the AFM image shows, this film is less rough due to the residual PEDOT:PSS on the surface. The residual PEDOT:PSS smoothens the substrate and makes the subsequent deposition of a conductive electrode more uniform. As the figure shows, Film 4 is less rough in terms of both amplitude and spatial frequency than Film 2. The amplitude of roughness is ±2 nm comparing to ±3 nm in Film 2. In general, it also has less fluctuations than all other samples in the experiment. The thickness of the layer, also, can be managed to be thiner than Film 2 as the substrate has become much smoother. The PEDOT:PSS layer is 70 nm at 3500 rpm spin coating. This uniformity results in a higher yield in device fabrication and higher device efficiency. Figure 3.4 shows the frequency response resulted from Fourier transform of the surface roughness signal in Films 1, 2 and 4. As the figure shows, the PET substrate contains a frequency peak at 60 µm−1 . After deposition of the first PEDOT:PSS layer, as shown in Fig. 3.4b, the number of peaks increases, showing the increase in the roughness texture. In this substrate, the previous peak and some new ones appear at 50 and 100 µm−1 which is due to the original roughness of the substrate and the fact that PEDOT:PSS does not dry uniformly during the spin coating process. As depicted, after deposition and annealing of the second PEDOT:PSS layer, major frequency peaks disappear, indicating a smoother finished surface. This corroborates the spatial domain AFM images that show a smooth  49  3.5. Results and Discussions ×10−7 Frequency Amplitude (A.U)  1  Film 4  0 3 2 1  Film 2  0 3 2 1  Film 1  0 50  100 Spatial Frequency (µm−1 )  150  Figure 3.4: Frequency transfer results for the measured AFM roughness for a) Film 1, b) Film 2 and c) Film 4. surface for the final PEDOT:PSS layer. The residual PEDOT:PSS, shown in Fig. 3.2, partially fills the cavities and smoothens the surface as a pre-coating for the deposition of the second PEDOT:PSS layer. Fig. 3.5 shows the J-V characteristics of P3HT:PCBM devices fabricated with 1:1 ratio of P3HT and PCBM on PET substrate under AM1.5 100 mW/cm2 light source. The device fabricated by this method has Isc = 1.3 mA and Voc = 0.42 V. As the figure also shows, The Isc and Voc are not drastically different for devices made by 1-layer and 2-layer deposition methods ( Isc = 1.1 mA and Voc = 0.39 V for 1-layer PEDOT:PSS deposition) while the efficiency of the devices made using 2-layer deposition is an order of magnitude higher (PCE = 0.25-0.3% for 2-layer deposition comparing to 0.02-0.05% for 1-layer PEDOT:PSS deposition). This is due to the difference in fill factor (FF = 0.47% for suggested method comparing to 21% for 1-layer deposition). On the other hand, devices made by this method  50  Current Density (mA/ cm2 )  3.5. Results and Discussions  0  -0.5  -1 AM1.5 100 A/ cm , 2-Layer Dark J-V 2 AM1.5 100 A/ cm , 1-Layer 2  -1.5 0  0.15 Voltage (V)  0.3  0.45  Figure 3.5: The J-V curves of the devices fabricated using 1:1 ratio of P3HT to PCBM measured at 100 mW/cm2 in the light and dark conditions. It also shows the light J-V curve of the device deposited by single layer deposition of PEDOT:PSS. have a higher yield (∼80% for 2-layer deposition comparing to ∼10% for 1-layer method). Devices made on PET substrate also are around an order of magnitude lower in efficiency in comparison with the identical devices made on glass [11]. This is due the higher Isc , Voc and FF in devices deposited on glass. The reason for this discrepancy is the lower reflection of the incident light on glass substrate. Also, smoother surface of ITO coated glass leads to more optimized electrode design leading to effective built in electrical field in devices made on glass[4]. We have also repeated the aforementioned method for more than 2 layers. In this effort the spin coated PEDOT:PSS layer was rinsed after the second deposition and another layer of PEDOT:PSS was deposited. We noticed that depositing more layers of PEDOT:PSS increases the thickness of the layer to more than 80 nm and does not help to smoothen the surface any further. The thicker layer of PEDOT:PSS causes more absorption of light and decreases the efficiency of the device.  51  3.6. Conclusion  3.6  Conclusion  The surface roughness is one of the major issues for device deposition on flexible substrates. A 2-step method for deposition of PEDOT:PSS layer suggested here, drastically reduces the amplitude of the roughness. As the microscope and AFM results show, after deposition of a single layer of PEDOT:PSS and wiping the same layer using DI water, there is residual PEDOT:PSS on the surface of polymer that reduces the roughness of the substrate. The deposition of the second PEDOT:PSS layer, shows even a lower roughness which makes it suitable for deposition of organic semiconductor for organic photovoltaic devices. This method results in lower defects in the substrate and lower possibility of any shorts between top and bottom electrodes, hence a higher yield in device fabrication. Repeating the process for more than 2 layers does not change the surface characteristics much and leads to less control over the thickness of the PEDOT:PSS layer as the polymer chains may stack, especially on parts of the substrate with higher spatial roughness.  52  3.6. Bibliography  Bibliography [1] C. R. McNeill et al. Influence of nanoscale phase separation on the charge generation dynamics and photovoltaic performance of conjugated polymer blends - Balancing charge generation and separation. J. Phys. Chem.:111, 19153, 2007. [2] G. Dennler et al. Flexibe Conjugated Polymer-Based Plastic Solar Cells: From Basics to Applications Proc. of IEEE : 93, 8, 2005. [3] Jae-Hyeong Lee Structural and optical properties of CdS thin lms on organic substrates for exible solar cell applications J Electroceram.: 1108, 2006. [4] N.S. Saricifitci et al. Semiconducting polymers (as donors) and bulkminsterfullerene (as acceptor): Photoinduced electron transfer and heterojunction devices Synth. Met.: 59, 333, 1993. [5] B.A. Gregg et al. Comparing organic to inorganic photovoltaic cells: theory, experiment and simulation J. Appl. Phys.:93, 3605, 2003. [6] T.M. Clarke et al. Free Energy Control of Charge Photogeneration in Polythiophene/ Fullerene Solar Cells: The Inuence of Thermal Annealing on P3HT/PCBM Blends. Adv. Func. Mater. :18, 4029, 2008. [7] G. Li Investigation of annealing effects and lm thickness dependence of polymer solar cells based on poly3-hexylthiophene. J. App. Phys.:98, 043704, 2005. [8] Sung-Ho Jin et al. Optimization of process parameters for high-efciency polymer photovoltaic devices based on P3HT:PCBM system. Solar Energy Materials and Solar Cells:91, 1187, 2007.  53  3.6. Bibliography [9] K.O. Sylvester-Hvid Analyzing poly3-hexyl-thiophene:1-3-methoxy-carbonylpropyl-1phenyl- 6 , 6C61 bulk-heterojunction solar cells by UV-visible spectroscopy and optical simulations. J. App. Phys.:102, 054502, 2007. [10] G. Li et al. High-efficiency solution processable polymer photovoltaic cells by selforganization of polymer blends. Nature Mater:4, 864, 2005. [11] C.H.Woo  The Influence of Poly(3-hexylthiophene) Regioregularity on Fullerene-  Composite Solar Cell Performance. J. App. Chem. Soc: 130, 16324, 2009.  54  Chapter 4 Conclusion This chapter mainly focuses on the conclusions derived from the experiments performed on glass and plastic substrates in the previous chapters. Also, we present a number of future works that could be done to continue the research on P3HT:PCBM based solar cells.  4.1  Experiment on Glass  We have fabricated P3HT:PCBM devices on ITO coated glass with different regioregularity of P3HT. The spin-coated solutions of P3HT:PCBM are allowed to dry slowly in the room temperature. We have observed the efficiency and hole mobility in both low and high regioregular P3HT. After fabrication, the devices made by high RR P3HT, show a higher efficiency and hole mobility in comparison to the low regioregular counterparts. This is due to higher crystallization of high RR device because of more efficient π-π stacking in the polymer chain. We have also observed the efficiency of high and low RR devices after a thermal annealing. As shown in post annealing results, the efficiency of low RR device increases or remains steady while that of high RR device decreases drastically. This is due to the nano scale phase segregation which takes place more rapidly in high RR device as it has a higher tendency for crystallization and pushes the PCBM molecules out of the P3HT  55  4.2.  Experiment on Plastic  structure. This segregation decreases the effective interface area between the donor and acceptor molecules which leads to lower charge separation in these devices. We have also analyzed the degradation of both devices in a period of 4 months. Although devices are kept in the glove box, the efficiency of both devices decrease as a matter of phase separation. It is shown that the efficiency drop is more drastic for high RR device while the efficiency of low RR devices remains steady during the 4-month period after an early efficiency drop in the first 20 days. The hole mobility analysis in each case is performed by space charge limited current method. The results show that the hole mobility for low RR device drops more drastically in comparison to the high RR device. This is also due to the faster phase segregation in high RR device.  4.2  Experiment on Plastic  We have fabricated solar cells on flexible transparent electrodes. One of the major obstacles in flexible electronic is the roughness of plastic substrates such as PET. The Plastic substrates are an order of magnitude rougher than glass and especial treatments are necessary for high yield device fabrication. In this work, we present a two-step method which reduces the roughness drastically. As shown, we have spin-coated a layer of PEDOT:PSS and rinsed it with DI water after deposition and before it dries out. PEDOT:PSS particles remain on the substrate and make the substrate less rough as this step is done. The second PEDOT:PSS layer then deposited on the substrate. This layer is smoother in comparison to the one-step method and shows better contact between the active layer and electrode  56  4.3. Future Works as well as higher yield as a result of lower probability of top and bottom electrode contact.  4.3  Future Works  The research presented in this thesis covers fabrication of P3HT:PCBM solar cells. We have mostly focused on the degradation profile of the OPV’s based on regioregularity and thermal annealing. The following projects are some of the ideas for continuing the research for more efficient and low cost solar cells. Also, as a challenge in solar cell industry, one of the issues that should be addressed is devices that can be deposited on substrates that are not transparent, hence deploying inverted structure devices.  4.3.1  Nano-Particle Mixture of Active Layer  One of the main issues in organic solar cells is the low efficiency of these devices. One reason for low efficiency is the fact that mobility in general and particularly electron mobility is low. Low mobility of charges, accelerates the recombination. Development of nano-particle structure that increases the whole mobility in the structure is one of the challenging and popular field of research. One common method that addresses this issue is mixing the polymer blend with nano particles. It is shown that adding certain amounts of CNT or other nano particles to the structure increases the efficiency of device. Different weight ratio and CNT types drastically change the performance of the device and characterization of devices based on P3HT and PCBM. [1]  57  4.3. Future Works Other nano particles such as TiO2 and silver nano particles also have attracted attention. TiO2 and silver have show a great light absorption which could lead to higher efficiency. [2]  4.3.2  Replacing ITO and Inverted Structure  One major issue in organic and thin film solar cells is the use of ITO as the conductive transparent electrode. ITO is expensive and fragile. It also limits the substrates that could be used to plastic and glass. The fragility of the ITO substrate also limits the possibility of fabrication of flexible and bendable device. To address all the above issues, there is trend toward replacing the ITO with a , preferably, organic transparent electrode that could be spin coated on glass and plastic. PEDOT:PSS is the first candidate for this purpose although the resistivity of PEDOT:PSS may not be lowered as it could be for ITO. Mixing PEDOT:PSS with nano particles such as CNT is one way to make the electrodes more conductive. Despite the preliminary results [3], these electrodes need to be modified for higher conductivity and device efficiency. One other benefit of development of conductive organic electrode is the possibility of inverted structure of device. The organic devices are all fabricated on glass or plastic transparent substrates. Although this is a viable solution for modules of solar cells, there is a huge demand for devices that can be deposited on non conventional and not necessarily transparent substrates. Development of such devices, paves the way for depositing the ink-like organic solutions on the body of airplanes, automobiles even on the facade of the buildings. To make it possible, the Al electrode could be deposited on the surface and after deposition of active polymer layer, the top transparent electrode is deposited. [3]  58  4.3. Future Works  4.3.3  Experiments with Modified Conjugated Polymers  The nano-phase separation that is addressed in chapter 2, is a major reason behind the degradation of organic solar cells. This phase segregation, can be eliminated by binding the donor and acceptor molecules by grafting the acceptor molecule on donor during the polymer synthesis. This binding eliminates phase separation as nano phase separation happens when P3HT molecules crystallization continues and enhance phase separation. Characterization of graphed molecules and optimizing the solar cells made by these molecules is a promising method for decreasing device degradation.  59  4.3. Bibliography  Bibliography [1] S. Berson, R. Bettignies, S. Bailly, S. Guillerez and B. Jousselme Elaboration of P3HT/CNT/PCBM Composites for Organic Photovoltaic Cells. Adva. Func. Mat., 17, 33633370, 2007. [2] S. Kim, W. Kim, A. Cartwright, P. Prasad Self-Passivating hybrid (organic/inorganic) tandem solar cell. Solar Energy Materials & Solar Cells, 93, 657661, 2009. [3] E. Kymakis, G. Klapsis, E. Koudoumas, E. Stratakis, N. Kornilios, N. Vidakis, Y. Franghiadakis Carbon nanotube/PEDOT:PSS electrodes for organic photovoltaics. Eur. Phys. J. Appl. Phys., 36, 257259 2007.  60  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0067364/manifest

Comment

Related Items