Chapter 1 Introduction
1.4 Objectives and Arrangements of the Study
Based on the aforementioned points, we chose atomic layer deposited Al2O3 over-layers as the shell layers, on the nanoporous TiO2 electrodes of dye-sensitized solar cells (DSSCs).
Previous explanations for the improvement of DSSCs with the Al2O3 over-layers were based solely on the assumptions made on the energy levels at core/shell interface and quality of the coating on the TiO2 electrodes which, consequently, were rarely accurate, because the interfacial energy levels may change significantly due to chemical reaction and the coverage of shell layers may be decreased by the island growth. The future improvement of DSSCs will be determined, in part, by the extent of the understanding developed on the Al2O3/ TiO2
interface through the analysis of real energy levels and nano-structures.
To achieve a comprehensive study of the physical, chemical, and electrical properties of the Al2O3 over-layers, we studied three aspects of the ALD Al2O3/TiO2 electrodes: interfacial energy levels, surface coverage, and growth mode. We examine a low-temperature ALD process for forming the Al2O3 barriers to achieve compatibility with low-temperature DSSCs fabrication processes. The relation of interfacial energy level, surface coverage, and growth mode to power conversion efficiency (PCE) was investigated by using TEM, ultraviolet photoelectron spectroscopy (UPS), x-ray photoelectron spectroscopy (XPS), and reflective electron energy loss spectroscopy (REELS). The nano-structure and influence of ALD Al2O3
layers on nanoporous TiO2 electrodes in DSSCs were also studied using these characterization techniques.
Chapter 1 introduces the motivation and objective of this study.
Chapter 2 reviews the literature concerning basic concept of solar energy, recent progress of DSSCs, characterization and previous studies of core/shell electrodes.
Chapter 3 describes the experimental method and analytic techniques.
Chapter 4 discusses the interfacial energy levels of Al2O3 films on TiO2 electrodes of DSSCs. The related properties are also discussed.
Chapter 5 shows a calculated model that measures the coverage of the core/shell electrode of DSSCs from XPS signals. The model calculated the coverage from isolated spheres to muti-layers core/shell materials, which can evaluate the ideal PCE of DSSCs with the core/shell electrodes.
Chapter 6 discusses the growth modes of Al2O3deposited on TiO2electrodes of DSSCs.
Obtaining a comprehensive understandings of the influencing factors of the ALD Al2O3
deposited on TiO2electrodes of DSSCs was investigated.
Chapter 7 concludes the experimental results and suggests future work worthy for pursuing.
Chapter 2
Literature Review
This chapter reviews the properties of solar energy and power, development of dye-sensitized solar cells, and the characterization of core/ shell electrodes.
2.1 Solar Energy and Power Conversion Efficiency
The solar irradiance, also named as solar constant, is defined as the power density of sunlight outside the atmosphere of earth and is regarded as a constant, 1367 W/m2, by the World Radiometric Center (WRC) [38]. However, when sunlight reaches the earth’s surface, the solar radiation changed due to atmosphere effects (i.e., absorption and scattering), weather, latitude, time etc. The measurement of power conversion efficiency of solar cells is affected by many factors, hence, it is necessary to introduce the standard conditions to evaluate the power conversion efficiency of solar cells.
2.1.1 Air Mass
The maximum radiation strikes the earth’s surface when the sun is directly overhead, having the shortest path length through the atmosphere at clear skies. The path length is called the air mass (AM) and can be calculated by the equation AM = 1/cos, where is the zenith angle as shown in Fig. 2.1. The standard solar spectrum outside the earth’s atmosphere is called AM 0. The path length of standard solar spectrum used for efficiency measurements of solar cells is AM 1.5 G (global), given that = 48.19° [39]. Fig 2.2 presents the photon
The spectrum is normalized such that the integrated irradiance (i.e., the amount of radiant energy received from the sun per unit area) is 1000 W/m2. In the diagram, the maximum photocurrent is the current at short-circuit conditions for a solar cell device converting all incident photons below the absorption onset wavelength into electric current. For example, the maximum short-circuit current for a solar cell with an absorption onset of 700 nm is 20 mA/cm2.
Figure 2.1 Illustration of the path length of solar radiation and the zenith angle .
Figure 2.2 Photon flux of the AM 1.5 G spectrum at 1000 W/m2, and calculated accumulated photocurrent. (ASTM G173-03)
2.1.2 Power Conversion Efficiency
The power conversion efficiency (PCE, η) of the dye-sensitized cell is determined by the following equation [40]:
η= Jsc x Voc x FF / Is (2-1)
where the Jsc is the photocurrent density measured at short circuit, the Voc is open-circuit photo-voltage, FF is the fill factor of the cell and equal to JMaxx VMax / Jscx Voc, and Is the intensity of the incident light, as shown in Fig. 2.3. The maximum power (PMax) is equal to JMax VMax where the power output of the cell is maximal. Then the fill factor of the cell can be calculated as follows:
FF = PMax / Jsc x Voc (2-2)
Figure 2.3 Typical shape of the current-voltage curve of a solar cell.
2.1.3 Standard Measurement of PCE
Accurate efficiency measurement of a solar cell depends on the international standard reporting conditions (SRC) such as 100 mW/cm2 total irradiance, AM 1.5G reference spectrum, and 25 ℃ cell temperature [41]. The current AM 1.5G reference spectrum used by the international terrestrial photovoltaics community is from International Electrotechnical Commission (IEC) Standard 60904-3, as shown in Fig. 2.4, and American Society for Testing and Materials (ASTM) Standard G173-03. Typically, the irradiance incident on the solar cell is measured with a reference cell. The spectral error in the measured short-circuit current (JSC) of the solar cell is induced by the mismatch between the spectral irradiance of the light source and the reference spectrum. In addition, the difference between the spectral responses of the reference detector and test cell needs to be corrected. Therefore, a correction for these errors should be applied. The errors can be expressed as a spectral
where ERef() is the reference spectral irradiance, ES() is the source spectral irradiance, SR() is the spectral responsivity of the reference cell, and ST() is the spectral responsivity of the test cell, all of which are a function of wavelength (). The range of 1 and 2 in the above equation should cover the full spectra and spectral responses to avoid error.
Figure 2.4 Spectral irradiance for AM 1.5 G reference spectrum (IEC 60904-3) and the typical source irradiance of an Oriel 150 W solar simulator. Intensities of the spectra have been normalized to 100 mW/cm2 [41].
2.2 Developments in Dye-Sensitized Solar Cells 2.2.1 Evolution of Dye-Sensitized Solar Cells
In 1873, H. Vogel discovered the “dye sensitization” by adding some organic dyes into silver halide to enhance the green and red light absorption [42]. The silver halides used in photography are insensitive to visible light and have band gaps in the range of 2.7-3.2 eV, just as the TiO2 used in DSSCs. The dye-sensitized phenomenon was later explained as the electron transfer from the organic dyes to the silver halide. What followed was the first sensitization of photo-electrode using a similar chemistry was performed in 1887 [43].
Research in the sensitization in solar cells then commenced [44-47] and showed cell conversion efficiencies below 1% in the following century.
The milestone of high-efficiency DSSCs was established by B. O’Regan and M. Grätzel in 1991 [10]. The cell construction with 7% solar power conversion efficiency composed of nano-crystalline TiO2 electrode and ruthenium complex dye is still used now in modern DSSCs. The surface-area of the TiO2 film introduced was believed to induce the enormous power conversion efficiency by loading larger amounts of dye and increasing light absorption.
The great improvement achieved in their research was not only the introduction of the nano-crystalline TiO2 electrode, to achieve the enormous power conversion efficiency, but also determining the main structure for the best-performing modern DSSCs. In recognition of this contribution, these types of DSSCs are generally referred as “Grätzel’s Cell”. In the decade after the original publication, the record efficiency for the Grätzel’s cell increased from 7.1% to 10%, however, in the subsequent decades the efficiency has only increased to 11.5% [48].
2.2.2 Operational Principles of Dye-Sensitized Solar Cells
The main operating principle of the DSSCs is presented in Fig. 2.5 [49]. From left to right, the cell consists of a transparent conductive oxide (TCO) substrate, normally doped SnO2 with fluorine on glass. A porous transparent semiconductor layer is coated onto the substrate, normally 10-15 μm of ~20 nm TiO2 particles, which gives an internal surface area of ~1000 cm2 per cm2 substrate. Dye molecules are absorbed onto the internal surface of this semiconductor to form approximately one monolayer. The typical dye is N719. The substrate is joined to a counter electrode, usually a few nanometers of platinum on another
TCO glass. A seal is formed around the cell, followed by injection of electrolyte through a hole which is subsequently sealed. So far all DSSCs have been based on an iodine/iodide electrolyte in organic solvent. A typical electrolyte consists of methoxyproprionitrile (MPN) with 0.6 M propylmethylimidazolium iodide, 0.1 M LiI, 0.1 M tert-butylpyridine, and 0.1 M iodine.
Figure 2.5 Energy band diagram of a typical DSSC employing an iodide/triiodide-based redox eletrolyte and N719 as a sensitizer dye [49].
As shown in Fig. 2.5, the sensitizer attached to the surface of a porous film absorbs the energy from sunlight (hv). The injection of electrons into the conduction band of the oxide is induced by the process of photo-excitation of the dye. Subsequently, the electron from electrolyte regenerates to the dye on the surface of porous films. The electrolyte contains the negative ion of iodide/triiodide (I–/I3–) couple as a redox couple. Reduction of positive ion of dye (S+) by iodide (I–) regenerates the original form of the dye (So) while producing triiodide
ions (I3–). This prevents any significant increase of S+, which could recapture the conduction band electron at the surface. The iodide is regenerated by the reduction of the triiodide ions at the counter-electrode, where the electrons are acquired from the external circuit through the Pt. The whole reaction can be represented by following processes.
Anode: So + hv → S* _Absorption S* → S+ + e-(TiO2) _Electron Injection 2S+ + 3I– → 2So + I3– _Regeneration Cathode: I3– + 2e-(Pt) → 3I–
Cell: e-(Pt) + hv → e-(TiO2)
Where the So represents the dye, the S* indicates the negative ion of dye. Thus, the device is generating electricity from light without any permanent chemical transformations.
Recently, a second type of operating principle of the DSSCs was proposed [50-52]. As shown in Fig. 2.6, the type II DSSCs are based on direct electron injection from the ground state (HOMO) of the sensitizer (dye) into the conduction band of TiO2. Unlike the conventional DSSCs, photoexcitation of the type II dye results in the direct electron injection from the ground state of the sensitizer into the conduction band of the TiO2, omitting the charge transfer that electron jumping from the ground state to the excited state (LUMO) of dye and then injecting into the conduction band (ECB) of the TiO2. Noted that no photoexcited dye states are involved in the type II DSSCs, in contrast to conventional DSSCs.
Typical type II sensitizers are organic molecules composed of endiol ligands that form a chelating bond with an under coordinated tetrahedral Ti(IV) surface state [52].
Figure 2.6 Illustrations comparing the operational principles of a (a) conventional DSSC and (b) Type II DSSC [52].
2.2.3 Surface Area of the Nano-Crystalline Electrodes
The specific surface area of the nano-crystalline TiO2 structure is about 2000 times larger than that of the bulk TiO2 structure [10] and directly increases both the light harvesting area and dye-electrode interface enormously. Figure 2.7 compares the incident photon to charge carrier efficiency (ICPE) of dye-sensitized solar cell with flat electrode and nano-crystalline electrode, the improvement is about 600 times increase for the nano-crystalline electrode [5]. The incident-photon-to-current conversion efficiency is plotted as a function of wavelength with both flat electrode and nano-crystalline electrode.
The IPCE value obtained with the single-crystal electrode is only 0.13% at 530 nm, where the value with the nanocrystalline electrode reaches 88% — more than 600 times greater.
Figure 2.7 Comparison of the incident wavelength to charge carrier efficiency (ICPE) of dye-sensitized solar cell with (a) flat electrode and (b) nano-crystalline electrode [5].
As mentioned above, there are many ways to improve the surface of porous electrodes to optimize the efficiency. One of the most successful methods is to coat the nano-crystalline electrode with a thin shell layer has been research over last decade. However, although many variants of the fabrication, including surface layers on nano-crystalline particles, have been studied, the lack of characterization limited the development in DSSCs with a core/shell electrode. For example, although a 0.2 nm thick shell layer can be deposited on a nano-crystalline or core/shell electrode, it is difficult to measure the surface coverage of this shell layer [53]. The following sections will discuss the problems in detail.
2.3 Core/Shell Electrodes in DSSCs
As shown in Fig 2.8, two core/shell structures have been developed to create the electrodes of DSSCs [54]. One includes the fabrication of nanoparticles and then forming a shell layer on the surface of nanoparticles. This leads to the formation of core/shell structured
nanoparticles to form the photoelectrode film (Fig 2.8a). However, such a structure creates an energy barrier not only at the nanoparticle/electrolyte interface but also between the individual core nanoparticles. In another structure, the photoelectrode film is composed of nanoparticles prepared prior to the deposition of shell layer, as show in Fig. 2.8(b). The latter structure is obviously beneficial to electron transport, it is a challenge to fabricate a shell layer in term of ideal uniformity of the shell material.
Figure 2.8 Core/shell structures used in DSSCs. (a) The shell layer is formed prior to the film deposition, (b) the shell layer is coated after the film deposition [54].
In this study, we defined the core/shell electrode like previous study [55] as the nanoporous inorganic semiconductor electrode that is covered with a shell of other metal oxide and coated after the film deposition. These electrodes can slow the recombination processes by the formation of an energy barrier at the TiO2 surface. The conduction band of the shell should be higher than that of the core semiconductor to generate an energy barrier for the reaction of the electrons present in the core with the oxidized dye or the redox mediator in solution. Nevertheless, there exist the varying conclusions on the path of electron transfer through the shell layer.
2.3.1 Operational Principles of Core/Shell Electrodes
The first core/shell electrode was developed by A. Zaban et al. in 2000 [11]. They reported the fabrication of a TiO2/Nb2O5 nanoporous electrode, which improved the performance of dye sensitized solar cells by more than 35%. In the mechanism described as the illumination of a DSSC, an electron is injected from the dye into the TiO2 film followed by a hole transfer to the electrolyte. The injected electrons must cross the TiO2 film and reach the conducting substrate, while the oxidized ions diffuse towards the back electrode, where they are re-reduced. The porous geometry that permits the presence of the electrolyte through the entire electrode provides a high surface area for recombination between the injected electrons and the holes in solution. In the absence of an energy barrier at the electrode–electrolyte interface, the rate of this recombination process may be very high depending on the properties of the hole carrier. As illustrated in Fig. 2.9, the energy level differences form an energy barrier at the electrode–electrolyte interface, which can reduce the rate of recombination processes of the photoinjected electrons. A comparison of two similar DSSCs that differ only in their nanoporous electrode, shows that the core/shell electrode is superior to the standard one with respect to all cell parameters. This demonstrated superiority, measured from many cells results in a 35% increase of the overall conversion efficiency.
Figure 2.9 A schematic view of the new nanoporous electrode which consists of a nanoporous TiO2 matrix covered with a thin layer of Nb2O5 [11].
The second mechanism suggested by E. Palomares [14,15], where the conformal growth of an overlayer of Al2O3 on a nanocrystalline TiO2 film, is shown to result in a 4-fold retardation of interfacial charge recombination, and a 30% improvement in photovoltaic device efficiency. The main charge-transfer events that take place at the TiO2/dye/electrolyte interface are depicted in Fig. 2.10. Visible light is absorbed (1) by the sensitizer dye. Electron injection (2) from the excited state of the dye into the conduction band of the TiO2 by tunneling effect is followed by the subsequent regeneration of the dye by I–/I3– red/ox couple (4). Efficient operation of the DSSC device relies upon the reduction of the possible recombination pathways occurring at the TiO2/dye/electrolyte interface, allowing efficient charge transport through the TiO2 film and electrolyte and subsequent charge collection at the device contacts. There are two possible recombination losses to consider. The injected electrons may recombine either with oxidized dye molecules (3) or with the oxidized redox couple (5).
Figure 2.10 Illustration of the interfacial charge-transfer processes occurring at the nanostructure TiO2/dye/electrolyte interface of DSSC [14,15].
The difference between these two mechanisms can be seen at the path (2) of electron transfer. The first mechanism claimed electron injection (2) from the dye into TiO2 without any energy barriers, but the second mechanism assumed electron injection (2) from dye into TiO2 by tunneling effect. This understanding is based on the assumption of that ultra-thin over-layers maintain the same energy levels after coating on the TiO2 electrodes as in the bulk structure; consequently, the energy diagrams of DSSCs with Al2O3/ TiO2 interface structures have been determined mostly by inserting the energy data from bulk material and assuming the vacuum levels of all of the layers are equal.
In 2002, Diamand et al. used SrTiO3as a shell layer to coat on the TiO2 electrode [55].
They found that SrTiO3layer shifted the conduction band of the TiO2(ECB0) in the negative direction, rather than formed an energy barrier at the TiO2 surface, as shown in Fig. 2.11.
This type of photoanode modification cannot suppress the back recombination rate as a function of the electron density, but can suppress the back recombination as a function of the applied potential. As a result, the forming “surface dipole” reduced Jsc slightly and significantly increased the Voc and the devices conversion efficiency was improved by 15.3%.
As shown in Fig. 2.11(b), the formation of a potential step subjected to the surface dipole shifted the conduction band of TiO2 in the negative direction.
Figure 2.11 Energy diagram representing the movement of (a) the TiO2conduction band by the SrTiO3 coating and (b) the effect of the band shift at applied potential indicated the coated system is more resistive than the bare electrode [55].
The core/shell electrode has also been used in the type II DSSC [52]. Figure 2.12 shows the operation of the SrTiO3 barrier layer as a barrier of recombination in the type II DSSC using catehol as a dye; therefore, the cell is named a catechol-sensitized type II DSSCs (CSSCs). The SrTiO3’s conduction band is higher than the conduction band of TiO2 by 0.2 eV, forming an energy barrier between the TiO2 and the catechol. Due to excitation, an electron is injected from the HOMO level of the catechol molecule into the conduction band of the SrTiO3. The energy difference between the SrTiO3 and TiO2 conduction band edges creates a tendency for the electron transfers from the SrTiO3layer to the TiO2 core, resulting in the reduction of the back electron transfers to the oxidized catechol compared to the uncoated TiO2 particles.
Figure 2.12 Schematic drawing of a CSSC with a SrTiO3 barrier layer (left). The band diagram shows the different electron-transfer processes (right) [52].
Another method to improve electron injection and suppress electron recombination in core/shell electrodes is by reducing the surface state sites of electrode via passivation [56,57].
Another method to improve electron injection and suppress electron recombination in core/shell electrodes is by reducing the surface state sites of electrode via passivation [56,57].