Large-scale full-solution, vacuum-free gravure printed ITO-free flexible organic solar cells

Contents

1 Introduction
1.1 The need for solar energy
1.2 Generations of solar cell technology
1.3 Research fields of organic solar cells

2 State of art and motivation

3 Fundamentals
3.1 Working principle of OPV device
3.2 Device geometries of OPV device
3.3 ITO-free OPV device
3.4 Materials in OPV device
3.5 Characterization of OPV device
3.5.1 J-V curve
3.5.2 LBIC
3.6 Large-scale manufacturing methods for OPV
3.6.1 Coating technologies
3.6.1.1 Slot-die coating
3.6.1.2 Blade coating
3.6.1.3 Spray coating
3.6.2 Printing technologies
3.6.2.1 Screen printing
3.6.2.2 Gravure printing
3.6.2.3 Flexographic printing
3.6.2.4 Inkjet printing
3.6.3 Summary of coating and printing techniques
3.6.4 R2R concept and manufacturing strategies

4 Experimental
4.1 Gravure printing of silver back cathode
4.2 Gravure printing of ZnO
4.3 Gravure printing of active layer
4.4 Gravure printing of PEDOT:PSS
4.5 Gravure printing of silver front anode
4.6 Deposition of remaining layers
4.7 Characterization

5 Results and discussion
5.1 Gravure printed silver back cathode
5.2 Gravure printed ZnO
5.3 Gravure printed active layer
5.4 Gravure printed PEDOT:PSS
5.5 Gravure printed silver front anode
5.6 Summary of results

6 Challenges

7 Conclusion and outlook

Bibliography

List of Abbreviations

illustration not visible in this excerpt

List of Figures

1.1 Global energy consumption

1.2 Chart of power conversion efficiencies

1.3 Organic solar cell research fields

3.1 Band diagram and BHJ structure of an OPV device

3.2 Principal device geometries of BHJ organic solar cells

3.3 Modified inverted ITO-free layer stack (IOne)

3.4 Chemical structure of common materials used in OPV devices

3.5 Characteristic J-V curve of a solar cell

3.6 Schematic illustration of LBIC experimental setup

3.7 Schematic illustration of slot-die coating

3.8 Schematic illustration of blade coating

3.9 Schematic illustration of spray coating

3.10 Schematic illustration of flatbed screen printing

3.11 Schematic illustration of rotary screen printing

3.12 Schematic illustration of gravure printing

3.13 Schematic illustration of flexographic printing

3.14 Schematic illustration of DoD piezoelectric inkjet printing

3.15 Process route principles

4.1 Gravure printed ITO-free layer stack

4.2 Laboratory-scale R2R coating/printing system

4.3 Two-roller gravure printing setup

4.4 Mini Roll Coater™

4.5 Slot-die head

4.6 Gravure printing of silver back cathode

4.7 Gravure printing of ZnO

4.8 Gravure printing of active layer

4.9 Gravure printing of PEDOT:PSS

4.10 Grid pattern of second gravure cylinder

4.11 Gravure printing of silver front anode

List of Figures

4.12 Deposition of remaining layers

4.13 Experimental setup for J-V characterization

5.1 Gravure printed silver back cathode

5.2 Silver labels

5.3 J-V curves of cells with gravure printed back silver cathode

5.4 Gravure printed ZnO

5.5 LBIC image of viscous fingering phenomenon

5.6 Solar cell characteristics - gravure printed silver and ZnO

5.7 Gravure printed active layer

5.8 Active layer labels

5.9 Absorbance curves of slot-die coated and gravure printed active layers

5.10 J-V curves of cells with gravure printed silver, ZnO and active layer

5.11 Gravure printed PEDOT:PSS

5.12 Transmittance curves of slot-die coated PEDOT:PSS layers

5.13 Gravure printed silver front anode

5.14 Completed gravure printed solar cell sample

5.15 Gravure printability and two-dimensional patterning

6.1 Modified gravure printing setup

6.2 Doctor blade position

7.1 Completed organic solar cells

List of Tables

2.1 Reports in literature on gravure printing for OPV 10

3.1 Comparison of film-forming techniques 33

At the beginning of this thesis the question why there is need for using renewable energies such as solar energy is answered, in addition the different generations of solar cell technology are specified and eventually the major research fields of organic solar cells are shortly described.

1.1 The need for solar energy

The global energy consumption and therefore the demand is increasing steadily while traditional energy sources are running out sooner or later and are coupled with tremendous hidden costs. Energy efficiency and conservation can temper a portion of the global demand but alone they far fall short of matching the expected need. Today we consume energy of approximately 18 terawat (TW), a number that is projected to rise to over 25 TW by 2035 and widely believed to reach 30 TW by the year 2050. In other words, the projected energy demand in 35 years from now is about twice the amount of energy we consume today.[1] Thus the societal questions where we can obtain this amount of energy and what consequences humanity has to face are posed. Incidentally, TW is an unit normally used for power but here it is adopted as unit for energy use.

In Figure 1.1 the global energy consumption in 2013 breakdown by each energy source is illustrated. Fossil fuels collectively representing over 80 % of the total energy supply dominate today’s energy mix. Renewable energies represent on the contrary only 11 % which is contributed mostly by hydroelectric power. Wind, solar and geothermal energy are currently tiny pieces of the action.[1]

Beside that fossil fuels are a limited resource they are associated with tremendous hidden costs. Using fossil fuels especially coal causes emissions from fossil fuel power plants, particulates and ozone, air, water and thermal pollution. These impacts directly affect the substantial cost in health insurance, food prices, among others which are not considered yet. However, the biggest drawback of consuming fossil fuels is the greenhouse effect and climate change which is generated mainly by carbon dioxide (CO2) contributing to delirious effects such as sea level rise and floods.[1]

illustration not visible in this excerpt

Figure 1.1: Global energy consumption breakdown by energy source in 2013.[1]

Nuclear energy is one of the major secondary energy sources in use today (Figure 1.1). Nevertheless, it has beside some of the mentioned ones above several hidden costs of its own: dual-use capabilities which equates to nuclear weapons, challenges in long-term management of radioactive waste and safety issues disclosed by nuclear accidents like recently happened in Fukushima including its consequential charges.[1] Beyond fossil fuels and nuclear power the only other major energy source are the renewable ones. They imply hugely less amount of drawbacks and hidden costs, though the impacts on local wildlife are important to consider. However, the small role of renewable energy sources in today’s energy mix is still caused by their ap­parently higher financial costs. Therefore they are regularly subsidized by local governments to support an energy transition. Many renewable energy sources are relatively simple to implement in the developing world due to the distribution in nature without a need of an established centralized grid infrastructure.[1]

The renewable hydroelelectric power, wind, geothermal energy and others such as biomass or ocean energy are only technically feasible to contiguously produce electricity to a limit probably in the range of 8 TW which is far below the global demand of 30 TW projected for 2050.[1]

Solar electricity is another renewable energy source that contributes to the world’s demand but still represents a small part of the energy production. There are myriad ways in which energy from the suns can be utilized but the focus of the thesis is on photovoltaic (PV). The huge benefit of solar PV energy is that the sheer unlimited amount of accessible solar power received fairly consistently across the earth’s surface from the sun is sufficient to fully substitute the global primary energy demand.[2]

Over 65 TW of energy are achievable by only solar PV supply when assuming a usage of just 2 % land area and a power conversion efficiency (PCE) of 12 %.[1] Although solar energy alone can theoretically provide the whole global energy demand, a mix including other (renewable) energy sources is a practical and favorable necessity for energy stability due to the intermittent availability and the ongoing lack of adequate energy storage.

PV will have to serve as the backbone of solar energy contribution in the next few decades. With the knowledge of nanotechnology the energy challenge of the future can be faced and the way for safer sources of clean and sustainable energy will be smoothed.

1.2 Generations of solar cell technology

Solar PV cells are devices that convert sunlight directly into electricity at the nanoscale without emitting CO2. PV devices have been fabricated from a wide variety of materials with differing power conversion efficiencies which is shown in Figure 1.2. However, todays PV technology can be categorized into three genera­tions of solar cells:

1st generation: The first generation cells are based on thick mono- or polycrystalline silicon wafer (100 μ^,). The efficiencies are breaking the 25 % value for lab-scale devices.[3] Commercial roof top modules have an efficiency of ;20 %. Manufacturing the modules requires energy intensive processes with high temperatures and cutting wafers from ingots results in large material loss. Furthermore, high initial costs of modules are essential.[4,5]

2nd generation: The second generation of solar cell technologies addresses the ma­terial consumption during fabrication and focuses on thin-film deposition tech­nology such as physical vapour deposition (PVD) and chemical vapour deposi­tion (CVD) amongst others. A variety of materials such as amorphous silicon, copper indium gallium selenide (CuInxGai_xSe2 or CIGS) and cadmium tel­luride/cadmium sulfide (CdTe/CdS) are utilized.[6] Lab-scale efficiencies of around 20 % are achievable while module-based are in the range of 16 %.[3,7] Nevertheless, the bottleneck of these types is the high cost for PV installa- tion.[5]

3rd generation: The category of cheaper emerging photovoltaic technologies, the third generation, is mainly based on organic materials which is a broad concept. In detail, the genre includes polymer solar cells[8-10], dye-sensitized solar cells

illustration not visible in this excerpt

Figure 1.2: Chart of certified power conversion efficiencies of best research solar cells from 1976 through 2012 for various photovoltaic technologies.[11]

(DSSC) and small-molecule solar cells.[5] Record efficiencies for lab-scale sized organic photovoltaic (OPV) devices beyond 10.5 % were achieved.[3].

The current research all around the world is still focused on lab-scale sized devices ^1 cm2, ITO-glass substrates, and spin coating as the main fabrica­tion method. These OPV devices are far from any practical application until considering the large-scale production compatibility.[4,5]

The third generation solar cell technology summarizes in addition further con­cepts such as perovskite and quantum dot cells or drafts to increase the effi­ciency beyond 30 % as displayed in Figure 1.2 due to absorption of larger parts of the solar spectrum by stacking multiple devices and absorption layers.[12,13]

Polymer solar cells are of particular interest because they can be produced very fast from solution by coating or printing processes at temperatures less than 140 °C and much lower energy pay back time (EPBT), which is the time required to produce the energy invested during its device life time, than conventional PV cells of the first or second generation of solar cell technology.[14-16] Moreover, polymer solar cells own outstanding potential to reduce the overall cost of energy production due to minimization of manufacturing and materials expenses by low energy con­sumption during production and high abundance of raw materials, high degree of automation and implementation of large-scale production by high-speed and high throughput roll-to-roll (R2R) processing. Apart from that there are also environ­mental advantages such as no use of toxic substances or by-products, no rare-earth elements, precious and heavy metals, toxic substances, and fully recyclability with low carbon footprint during the module production process.

The history of developing organic and polymeric photovoltaic goes back to 1986 and would be beyond the scope of this thesis. It can be found elsewhere.[17] Since more than 25 years of research an tremendous amount of developments of device architectures and materials are made which are covered in numerous reviews and books.[4,9,10,18-21]

The focus of the thesis is on polymer solar cells which are part of the organic solar cells group.

1.3 Research fields of organic solar cells

Research in the field of organic solar cells is still in an exciting phase with a rapid progression of improvements. The main direction in terms of research trends has been towards the improvement of power conversion efficiency under simulated sun­light. Of course, the PCE of organic solar cells is highly important in order to compete with the mature silicon technology of first generation solar cells. Further­more, consistently new records of power conversion efficiencies justify the research. However, beside the PCE there are at least two other crucial factors, stability and process, that are becoming more considerable for the success of organic solar cells. The individual areas of stability and process have been given relatively little at­tention compared to the efficiency. Another parameter is the cost of material and processing.[22,23]

illustration not visible in this excerpt

Figure 1.3: The challenge of unifying the three major organic solar cell research fields for one material: stability, process and efficiency.[22]

The three circles representing diversified research fields overlapping each other summarized in Figure 1.3, which illustrates the unification challenge, highlights that research in only one of the three areas will not necessarily lead to a useful organic photovoltaic technology due to the need of some extent of all three factors. Each feature has been demonstrated individually but their combination for one material is still a huge ongoing challenge. As long as only one of these areas is addressed in research a breakthrough towards application of the technology is expected to be slow.[4,22,23]

Nevertheless, the focus of the studies of this bachelor thesis is only on the research field of processing on the basis of performing a feasibility study including evaluating the possibility to fabricate large-scale organic solar cells by gravure printing.

2 State of art and motivation

In this chapter the state of art of two-dimensional gravure printed organic photo­voltaic (OPV) which is the main focus of this bachelor thesis and its motivation is shortly summarized. The use of gravure printing for the preparation of organic solar cells has only been referred in very few cases of literature to this day as listed in Table 2.1. Scientific publications of other polymer-based devices like organic light- emitting diode (OLED) that employ a similar layer stack to OPV including polymer light-emitting diode (PLED), organic thin film transistor (OTFT), organic photo­diode (OPD) and printed electronics are excluded from the table and can be found elsewhere[24-42]. The state of art in slot-die coating regarding OPV is beyond the scope of the topic of this thesis and is also not implied although it was regularly used in the studies as tool for fast testing under real production conditions, verifying and reproducibility purposes.

In OPV, gravure printing has been recently used for manufacturing the active layer (AL), electrodes or the intermediate layers as homogenous, thin films (Table 2.1). Most of the listed reports describe the printing process of the poly(3,4-ethylenedioxy- thiophene):poly(styrenesulfonate) (PEDOT:PSS) and the poly(3-hexylthiophene): [6,6]-phenyl-C6i-butyric acid methyl ester (P3HT:PCBM) layer discussing wetting behavior, solvent compositions, printability and morphology.

Beside printing the hole transport layer (PEDOT:PSS) as well as the AL as P3HT:PCBM of single cells (19 mm2)[43] resulting in a PCE of 2.8 %, small mod­ules with a size of 9.6 cm2 and a PCE of 1.9 % have been successfully printed in normal device architecture on a tabletop gravure proofer by Kopola et al.[44] This was accomplished amongst others by optimization of ink and processing conditions for the gravure processed P3HT:PCBM layer. On a larger module with eight inter­connected cells (active area: 15.45 cm2) a PCE of 1.58 % was scored. Unfortunately the electrodes were evaporated and the process was not roll-to-roll (R2R). Never­theless the PCE number of 2.8 % is by now the highest reported efficiency for partly gravure printed organic solar cells with P3HT:PCBM and PEDOT:PSS.

Studies in improving the morphology of the AL by adjusting the print process parameters like ink concentration, speed and drying time were pursued by Koidis et

al.[45,46] A PCE value of 1.0 % was reached. However, only the PEDOT:PSS and the P3HT:PCBM layer were printed by gravure. Eventually Koidis et al. carried out surface treatment and drying studies on PEDOT:PSS using R2R gravure printing on a small laboratory R2R printer.[47]

Furthermore, Voigt et al. reported payoff in producing sheet-to-sheet gravure printed organic solar cells on polyethylene terephthalate (PET) in an inverted struc­ture. Three subsequent layers (Titanium oxide (TiOx), P3HT:PCBM and PE- DOT:PSS) were fabricated by gravure but the cells were rounded off by evaporating the back electrode from gold in vacuum. However, a PCE of 0.6 % of a 4.5 mm2 device has been achieved.[48]. Another paper of Voigt et al. includes an ink and wetting study where the necessity of optimizing the ink and processing parameters to obtain a smooth and homogenous film is shown. The maximum PCE of 1.2 % was feasible.[49]

Beside the previous gravure papers mentioned above another scientific report pre­sented by Hübler et al. comprises the production of fully R2R-processed solar cells on paper substrate using not only gravure but also flexographic printing. Small devices with gravure printed AL and a size of 0.09 cm2 showed a maximum PCE of 1.31 %. The architecture used is ITO-free on opaque Zn/ZnO coated paper.[50]

The effects of different solvents on the fabrication of gravure printed organic solar cells were reported by Lee et al. A polymer bulk hetero junction (BHJ) comprising In­dium tin oxide (ITO)/PEDOT:PSS/P3HT:PCBM/Aluminium (Al) was produced. The AL was printed by gravure in a nitrogen filled glove box. The Al electrode was evaporated. Chloroform as solvent of the active layer resulted in the best PCE of 2.21 %.[51]

In addition Cho et al. investigated the mechanical flexibility of transparent PE- DOT:PSS films, gravure printed onto a flexible PET substrate. Organic solar cells fabricated on the PEDOT:PSS electrode showed a power conversion efficiency of 2 %.[52]

The paper of Logothetidis et al. focuses on the investigation of the optical prop­erties of R2R gravure printed nanolayers for OPV by R2R inline spectroscopic ellip- sometry in the ultraviolett region. The reviewed PEDOT:PSS transparent electrodes and the P3HT:PCBM blends were gravure printed onto flexible PET substrates. In conclusion the effects of the processing parameters on the optical properties and the quality of the printed nanolayers were discussed.[53]

Furthermore, Yang et al. successfully fabricated flexible OPV modules by gravure printing in ambient air using a drum-based industrial-scale printing proofer. Both the hole trasport layer and the AL were printed on ITO coated PET. The Al top electrode was evaporated. In some devices zinc oxide (ZnO) was applied on top of the P3HT:PCBM active layer by gravure printing. A PCE of 1.02 % of 45 cm2 sized modules composed of five cells connected in series was presented.[54]

Tuomikoski et al. investigated parameters affecting the printability of the active P3HT:PCBM and PEDOT:PSS layer. The films were processed by gravure printing in a R2R pilot line on ITO. A maximum PCE of 2.4 % for a single cell was achieved. Also an 1.6 % PCE value of an interconnection of three single solar cell with an active area of 6 cm2 was shown.[55]

Moreover, Ding et al. studied the fabrication of P3HT:PCBM based polymer bulk heterojunction photovoltaic cells on plastic substrates plus transparent conductive oxide (TCO) using gravure printing. By verifying the dependencies of device perfor­mance on material and process parameters and performing a wetting study implying contact angles, ink concentrations and viscosities, solvent characteristics and print­ing parameters optimized hole transport and active layers were printed. A maximum PCE of 1.68 % was reached.[56]

R2R reverse gravure printing technique was used to produce thin homogenous films for organic bulk heterojuction solar cells by Tobjörk et al. The PEDOT:PSS and P3HT:PCBM layer were successfully reverse gravure printed on an ITO covered plastic substrate in ambient air. The top electrode was thermal evaporated. The PCE of the organic solar cells was determined to 0.74 %.[57]

Finally, Apilo et al. analyzed the printability, thin-film formation and its influence on the OPV module performance as gravure printing is transferred from laboratory scale to R2R pilot production. Here, OPV modules were produced on top of ITO- coated PET. The hole transport layer of of PEDOT:PSS was R2R gravure printed on top of an ITO surface in continuous stripes. The photoactive layer of P3HT:PCBM was deposited likewise. The electron transport layer and electrode were evaporated in a glove box. A maximum PCE of 1.7 % was reached.[58]

Depending on the ink materials and surface properties there are different tech­niques that are more optimal for the deposition of one layer in the OPV stack than another. Thus a combination of varied methods is the better choice. Nevertheless, there is effort and even success in finding an all-solution processed functional solar cell layer architecture, on or without ITO and with R2R concept or not, fully pro­duced by only one practice such as slot-die coating[59,60], spray coating[61,62] and inkjet printing[63, 64]. For the case of gravure this goal is not yet achieved or at least was not reported in scientific literature. Most of studies concerning gravure printing for OPV were performed on desktop gravure tester and were not fully R2R. In addition the majority of experiments printing the AL by gravure were carried

illustration not visible in this excerpt

Table 2.1: Overview of important studies that employ gravure printing as a man­ufacturing method for the AL, electrode or the intermediate layer for OPV. Adapted from [5].

out with a high solid concentration. In all of the above-listed references excluding [50] the sample cells required a vacuum process for evaporating the back electrode. Several times plasma cleaning, ITO as substrate was utilized and gravure print­ing was often used for printing full layers or continuous stripes where cells were built up on later. Full gravure printing for OPV may be an attractive alternative to flexographic printing and a suitable technique for fast commercial processing of large-area, flexible organic solar cells with high throughput allowing feasible manu­facturing of arbitrary-shaped modules. In the research field of OPV one stands on the door sill between laboratory and large-area R2R production scale. Large-scale full-solution, non-vacuum, without glove box, processed, ITO-free, fully R2R (with­out benchtop equipment like tabletop tester) and completely gravure printed (all layers, not only the hole transport and AL) flexible organic solar cells were never fabricated so far. For the last-mentioned reasons this was chosen as goal of the stud­ies and topic of this bachelor thesis: ’’Large-scale full-solution, vacuum-free gravure printed ITO-free flexible organic solar cells”. Two-dimensional patterning for cell structuring by means of a gravure cylinder which was shown only by a minority of the reports[43,44] was favorable evaluated by printing text labels beside the regular printing of stripes.

In this chapter the fundamentals of an OPV device are presented: The section in­cludes the working principle of converting solar energy (photons) into electric power (charge flow), device geometries, followed by the concept of ITO-free organic solar cells and the materials used in an OPV device. Afterwards large-scale manufactur­ing methods for organic solar cells and finally characterization procedures for OPV are reviewed. Advanced chemistry and physics of OPV devices are not described in detail since the focus of the thesis is on large-scale fabrication technologies.

3.1 Working principle of OPV device

The field of solution-processed OPV covers various types of semiconducting polymer donor:acceptor material systems in which the acceptor part can be fullerenes, poly­mers, semiconductors nanoparticles, -crystals, respectively, or metal oxides. The latter are also referred to as hybrid solar cells. The scope in this chapter is on poly- mer:fullerene cells working by the bulk heterojunction (BHJ) principle. Clear and brief that means that two different semiconducting materials, one as electron donor (D) and the other as electron acceptor (A), is mixed in an organic solvent such as chlorobenzene and deposited on a conductive substrate. Both components can ab­sorb light. Brought together and after evaporation of the solvent and post-treatment steps microphase separation is taking place. The AL of the solar cell comprising an interpenetrated network of D and A is formed. The interface is randomly and unisotropic dispersed throughout the volume of the AL. The interconnected D and A domains of an OPV cell are continuously linked to the top and bottom electrodes (Figure 3.1b) allowing efficient charge transport to anode and cathode.

The function of an organic solar cell is described in abbreviated version below. For further details references [10,65-68] are recommended by the author. The working principle of a BHJ OPV device includes four fundamental steps whereby the band diagram and BHJ structure is illustrated in Figure 3.1:

1. Exciton generation:Upon illumination an incident photon with an energy that corresponds at least to the band gap energy of the active material is mainly

illustration not visible in this excerpt

Figure 3.1: (a) Simplified band diagram of photocurrent generation due to photon absorption, exciton generation, exciton diffusion, exciton dissociaton and charge carrier transport to the electrodes in a BHJ solar cell.

(b) Corresponding intermixed BHJ structure with charge carrier trans­portation paths.[5]

absorbed in the D material. It excites an electron from the highest occu­pied molecular orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO) and at the same time a positive charge carrier, so-called hole, re­mains in the HOMO level. In conjugated polymers the promoted electron has a reduced mobility due to the fact that both charge carriers are attracted to each other and bound by Coulomb forces (binding energy circa 0.4 eV[65]). Hence an electron-hole pair which is regarded as a quasi-particle, the exciton, is formed.

2. Exciton diffusion:The generated excitons diffuse inside the D phase to the

interface of the D and A material. Exciton decay or charge carrier self­recombination after a certain time (around 1 ns [10]) may appear if they are generated too far from the interface leading to emission of photoluminescence or/and thermal with corresponding reduction in PCE. The maximum distance an exciton can move within its lifetime is called exciton diffusion length. The BHJ concept of two intermixed materials decreases the diffusion length com­pared to a stacked bilayer structure. Therefore, the decay rate of excitons is being reduced. The dimensions of the two phases of blended materials should be in the range or smaller than the diffusion length showing a large variation of 4-20 nm for various (conjugated) organic materials.[69-73]

3. Exciton dissociation:In case an exciton reaches the D-A interface, the electron

is attracted by the energetic lower LUMO level of the A wherefor the LUMO level of the A needs to be on the order of the exciton binding energy. Thereby,

the exciton can dissociate resulting in a free electron and a free hole at the interface of D and A material: holes remain in the HOMO of the donor. Charge separation is happening. This overall process costs energy on behalf of the PCE. Thus the LUMO levels of the D and the A should match in such way that the LUMO level of the A is at least 0.3-0.4 eV lower than the LUMO level of the D to allow efficient electron transfer. Moreover, the A HOMO level must lie under the D HOMO level that the hole is not attracted by the electron acceptor material and not lead to recombination with electrons.[4] Note that the large interface and complexity in nanoscale morphology of D and A in the range of the exciton diffusion length is very crucial and largely improves the charge separation and transport, decreases recombination losses of charge carriers and effects a satisfying exciton-to-charge conversion efficiency. If the D-A phase are not continuous, the charges may be trapped in the AL by dead ends.

4. Charge carrier transport The free charge carriers transport is driven by an in­ternal electric field due to the different work functions of the front and back electrode. The charges are transported through the D (holes) and A material (electrons) by hopping between localized states whereat the charge carriers are collected at the corresponding electrodes: electrons at the cathode and holes at the anode. The photocurrent is generated by short circuiting or applying a load to an external circuit.[4,5,74]

Each step is crucial for enabling an efficient power generation and high PCE values. To attain a large PCE the key parameters such as the optical band gap or HOMO/LUMO energy levels of the materials and blend microstructure have to be fulfilled to maximize light absorption, charge separation and charge carrier transport. Organic materials with large absorption range (low band gap) can be synthesized and directly influence the first step, the exciton generation. Thus pho­tons at longer wavelength can be harvested and efficiency due to higher currents is improved. Tuning of HOMO and LUMO energy levels of donor and acceptor affects higher voltage values which rises the efficiency [75]. The microphase separation can be largely affected by processing parameters.[4,5,74]

3.2 Device geometries of OPV device

A standard geometry of an OPV device incorporates two electrodes, the front elec­trode at illumination side and the back electrode, with a photoactive layer sand-

illustration not visible in this excerpt

(a) Normal geometry

wiched in between. It is necessary that one of the two electrodes offers transparency that can be accomplished by using indium tin oxide (ITO) as conductive electrode. Usually ITO is sputtered or evaporated on a transparent substrate such as glass or PET. ITO is commercially available. Today, the BHJ concept with intermixed lay­ers of D and A in the AL is the most commonly used whereby other structures such as bilayer and highly ordered interdigitated heterojunctions are possible as well.[4] Intermediate or buffer layers, the electron and the hole transport layer (ETL, HTL), between the active layer and the electrodes support the charge selective transport of either electrons or holes but blocks in the converse case.[4,5,74]

Two device geometries named normal and inverted with light illumination through the transparent conductive electrode are actually used to manufacture OPV devices as illustrated in Figure 3.2.

The normal structure PET/ITO/PEDOT:PSS/BHJ/LiF/Al (Figure 3.2a) is most often used for processing lab-scale sized devices for testing new materials. One drawback is the utilization of a vacuum processing step while evaporating the low work function Al electrode on top of the active layer. The normal device architecture as the traditional geometry is known for its satisfying efficiency and relatively simple production.[4,5,74]

By flipping the layers stack named inverted architecture (Figure 3.2b) comprising for example PET/ITO/ZnO/BHJ/PEDOT:PSS/Ag, full-solution processing of the anode is made possible. The inverted stack offers the benefit of avoiding vacuum steps during deposition of all layers except the ITO and is as has been proved the better choice for large-scale production.[16,76]

After processing the complete OPV layer stack the device fabrication is finished by encapsulation procedures to protect the solar cells from environmental impacts by water and oxygen, such as photo-oxidation and photo-chemical processes, and to offer long functional device stability. Rigid glass would be the ideal barrier due to its increase of the gas diffusion path length but is not handled in a R2R process. Glass is

so far used for encapsulation of small lab-scale devices.[5] Stability and degradation issues are beyond the scope of this thesis as mentioned before and details can be found elsewhere.[22,23] Reference [77] where different large-scale R2R encapsulation procedures for polymer solar cells, namely Ultraviolet(UV)-curable adhesive, hot- melt and pressure sensitive adhesives, are evaluated is recommended.

3.3 ITO-free OPV device

Until now ITO was the material of choice for an electrode in current state-of-art OPV devices because of its high transparency combined with a low sheet resistance. ITO is a semiconductor with a wide band gap that absorbs only highly energetic photons and therefore has a high light transmittance in the visible range. But the fabrication of ITO electrodes involves vacuum steps and subtractive patterning processes. These highly energy intensive procedures rise the embodied energy higher than without ITO and consequently it is not feasible for low cost production of organic solar cells. Moreover, indium resources are scarce and high demand from display industry result in cost fluctuations. Furthermore, the transparent electrode needs high preparation temperatures.[4,5] These issues including energy pay back time (EPBT) and cost­driving factors are very crucial but go beyond the scope of this thesis and has been shown in a few life cycle assessment (LCA) studies.[15,78-80]

Anode(Ag) HTL (PEDOTPSS) BHJ active layer ETL (ZnO) FlextrodeH PEDOTPSS Ag grid Substrate

illustration not visible in this excerpt

Figure 3.3: Modified inverted layer stack (lOne) that replaces ITO and PEDOT:PSS.[5]

The research of OPV concentrates amongst others on finding a reasonable re­placement of ITO to fabricate a full-solution based solar device on large-area.[60,81] The IOne stack PET/Ag-grid/PEDOT:PSS/ZnO/BHJ/PEDOT:PSS/Ag with light entering from the bottom as illustrated in Figure 3.3 shows one successful ap- proach[82-84] under several research fields such as polymer, metal, combination of polymer and metal and carbon nanotubes and graphene.[85] In this architecture, ITO is substituted by silver (Ag) grids, the highly conductive polymer PEDOT:PSS on top and a full layer of ZnO as ETL. This transparent conductive electrode, named

Flextrode, is fully printable using low-temperature processes. Even special highly conductive PEDOT:PSS solution alone can act as replacement for ITO.[5]

The again modified ITO-free inverted layer stack of which the design and fabrica­tion takes center stage in all studies for this bachelor thesis is detailed further down in Chapter 4. To gain a deeper insight into the topic of ITO replacement in organic solar cells references [59,83,85-88] are advisable.

3.4 Materials in OPV device

OPV devices are named organic but only the HTL poly(3,4-ethylenedioxy-thio- phene):poly(styrenesulfonate) (PEDOT:PSS) layer and active photoconversion layer are in principle fully organic. In all the other layers except the substrate metals and metals oxides beside conductive and semi-conductive conjugated polymers and molecules are utilized.

PEDOT:PSS as structural formula illustrated in Figure 3.4a is mostly in use as HTL in normal geometry devices and is dissolved in aqueous solution. The layer only absorbs light in the deep UV region. The function of powerful charge extraction of the ETL to the Al cathode is fulfilled by calcium or lithium fluoride (LiF) as demonstrated in Figure 3.2a.[4,5]

In the inverted architecture (Figure 3.2b), the ETL coated or printed, respectively, either from solution-based precursors or nanoparticles are typically metal oxides like TiOx or ZnO. Beside PEDOT:PSS, one can use metal oxides from molybdenum or vanadium as HTL in inverted OPV devices.[5] In the latter Ag acts as back electrode or as current collection grid in front electrodes of the ITO-free Flextrode in the IOne layer stack (Figure 3.3).[4,5]

The basic part of an organic solar cell is the light absorbing AL. The most likely best known and researched D-A material combination is poly(3-hexylthiophene):[6,6]- phenyl-C6i-butyric acid methyl ester (P3HT:PCBM) whereby the chemical struc­tures of both components are shown in Figure 3.4b and 3.4c. They are easy to produce commercially which is important for R2R large-scale manufacturing ap­proaches. Note that they are dispersible in various solvents such as chlorobenzene, dichlorobenzene or chloroform.[4, 5] By means of this active material combination PCEs up to circa 5 % were reached.[89,90] The spectrum of P3HT is not matching the solar spectrum and absorbs only wavelengths below 650 nm. Assuming that in a P3HT:PCBM blend the polymer of the composite defines the optical band gap which is circa 1.9 eV[75] one can calculate the absorbed photon density and the power density. A P3HT:PCBM layer can absorb a maximum of 27 % of the avail-

illustration not visible in this excerpt

(a) PEDOT:PSS (b) P3HT (c) PCBM

able photons and 44.3 % of the available power. The PCE maximum is predicted as a value of 34.6 %. Deeper information about the calculation are specified in reference [91]. There is a huge research field in OPV to find new polymers with improved absorption characteristics. Further details about for instance low band gap polymers which offer besides higher efficiencies these properties are not topic of this thesis and are reported elsewhere in literature.[92]

3.5 Characterization of OPV device

Two characterization methods for OPV devices comprising J-V curve and light beam induced current (LBIC) for photocurrent measurements are explained. Further procedures to evaluate the performance of an organic solar cell like incident photon- to-electron conversion efficiency (IPCE) are excluded due to the fact that there were not used in the studies.

3.5.1 J-V curve

In the dark, the solar cell acts like a simple diode but under light illumination the typical current density-voltage characteristics (J-V curve) for an OPV device are shown in Figure 3.5 where the current density J is plotted versus voltage V. A voltage is applied and the current is measured. Furthermore, the key parameters open circuit voltage Voc (in V ), short-circuit current Isc (in A), short-circuit current density Jsc (in mA cm-2), fill factor FF (in %) and current density Jmpp and voltage VMPP for the maximum power point MPP are illustrated. The mentioned parameters used to characterize the performance of an OPV device are described as follows.

VOC is the cell voltage under open circuit conditions, the cell voltage in sunlight

when no current is flowing and represents the maximum output (photo)voltage that the measured solar cell can provide under ideal circumstances. In heterojunction devices VOC depends on the frontier orbital energy levels of the organic materials, hence on the LUMO level of the A and the HOMO level of the D as well as the interface contact of the active material and electrode. Moreover, other factors such as charge recombination can influence the VOC.

Note that the ISC is the current flowing through an illuminated cell when there is no external resistance.

Another parameter is JSC which describes the photocurrent density extracted from the device under illumination and zero bias. It represents the maximum (photo)current that can be obtained in a solar cell. JSC is influenced by the ab­sorption coverage of the organic molecules within the solar spectrum. It is linear or sublinear with intensity of the illuminated sunlight. Charge transport properties of the used organic semiconductors play a significant role as well to reach a high JSC value.[4,66]

The most important performance parameter for device evaluation, the power con­version efficiency (PCE, η), often mentioned before is defined by the ratio between the maximum electrical power output Pout (in W m-2) produced by a cell under illumination and the power or intensity of the incident light Pin (in W m2) on a given active area A:where the fill factor (FF) is defined as the ratio of the actual largest output power or

illustration not visible in this excerpt

where the fill factor (FF) is defined as the ratio of the actual largest output power or practical produced power (JMPP · VMPP), respectively, to the product of JSC and VOC (theoretically possible power output if both current (ISC) and voltage VOC are at their maximum), i.e:

illustration not visible in this excerpt

The FF value demonstrates the quality of the solar cell and is determined by the photogenerated charge carriers and the fraction thereof that reach the electrodes. It depends on the competition between charge carrier recombination and transport. It is shown in Figure 3.5 as maximum area within the J-V curve and where voltage VMPP and current IMPP are at the MPP. An ideal solar cell would have a rectangular shaped J-V curve and therefor the fill factor should be approaching the unity or as high as possible. It ranges for state of the art OPV devices from 60 to 70 % and is dependent on parasitic resistances such as series resistance Rseries including all

Figure 3.5 Characteristic J-V curve of a solar cell

illustration not visible in this excerpt

resistances at the interfaces in the layers, the conductivity of the semiconductors and the electrodes. Furthermore, in contrast to Rseries the shunt resistance Rsh needs to be high and comprises all current leakages through shunts due to defects in the layers. The fact that shunting leads to an inefficient cell is obvious.[4,5,66]

Besides experimentally characterizing the performance of an organic solar cell the VOC can be estimated from electrochemical measurements. It is given by the differ­ence between the D HOMO level EHOMO,D (in eV) and the A LUMO level ELUMO,A (in eV) together with an additional loss factor. This is a result of the discrepancy between the HOMO-LUMO energy levels and the electrochemical potential at which the charges are extracted from the solar cell[4,5,93,94]:

illustration not visible in this excerpt

where q is the elementary charge.

The typical spectrum of the light impinging on the surface of the earth on a cloudless day is given by the ASTM Standard G173-03 and named Air Mass 1.5 (AM1.5G). The so-called AM1.5G that is the overall reference for solar cell charac­terization during illumination cumulates an integrated power density of 1000 W m-2 (100 mW cm-2) and a photon flux distributed over a large range of wavelengths (280-4000 nm).[91]

3.5.2 LBIC

Light beam induced current (LBIC) is a well-established high-resolution non-de­structive optical characterization technique for mapping solar cells. A focused beam of light, usually a low power laser, illuminates localized areas of a solar cell area while