Literature Thesis Perovskite Solar Cells

1 downloads 0 Views 11MB Size Report
hexylthiophene) (P3HT) and poly(triarylamine) (PTAA). They used CH3NH3PbI 3-xClx as perovskite layer and heated the perovskite films coated with these ...
UNIVERSITY  OF  AMSTERDAM    

MSc  Chemistry   Master  Track   MDSC       Literature  Thesis    

Perovskite  Solar  Cells    

Stability,  design  architecture,  photophysical  properties,  and   morphology  of  the  film  in  organometal  halide  Perovskite-­‐ based  Photovoltaics  

    By     Aram  Farawar   Student  ID:  10480765     Nov.  2015   12  EC’s  

Research  period:  Sep.-­Nov.  2015           Edited  version  by  R.M.  Williams,  7-­‐1-­‐2016   Supervised  by:  Dr.  René  M.  Williams   Molecular  Photonics  Group   Van  ‘t  Hoff  institute  for  Molecular  Sciences   Universiteit  van  Amsterdam  

Summary Perovskite based photovoltaics (PV) have had a rapid and an unprecedented evolution in the past few years. These light-harvesting materials are of huge interest to the academic community in order to make more efficient solar cells which are expected to reach more than 20% PCE. They have both much lower processing costs and high power conversion efficiency (PCE) in comparison to other third generation thin film technologies. Since the operational methods of perovskite PV’s are still relatively new, there is great opportunity for further research in basic physics and chemistry around them. An important objective for the further improvement of the performance of perovskite-based PV’s is to extend their optical-absorption onset further into the red to enhance solar-light harvesting. In the past few years, scientists of this field have developed many methods and techniques to reach this important target. To explain how the production of improved solar cells can be achieved, in this thesis the most important emissive and photophysical properties of the known perovskites of methylammonium lead triiodide (CH3NH3PbI3), CH3NH3PbI2Br, formamidinium lead triiodide (NH2CHNH2PbI3) and related materials like the Cl, Br as well as the Sn variations will be discussed. The rise of perovskites in photovoltaics, some design architectures and preparation techniques of perovskite solar cells will also be discussed in this thesis.

 

2  

Content   Summary……………………………………………………………………………2 Content……………………………………………………………………………...3 1. Introduction………………………………………………………………....4 1-1: Structure name perovskite……………………………………………..4 1-2: Solar cells SQ-limit………………………………………………….5/6 2. The rise of perovskites in photovoltaics………………………………….7-9 3. Most prominent perovskites………………………………………………..10 3-1 Methylammonium lead triiodide (MAPbI3)…………………………...10 3-2 Formamidinium lead triiodide (FAPbI3)………………………………10 3-3 Synthesis of CH3NH3PbI3…………………………………………….10 3-3-1 Synthesis of CH3NH3I………………………………………10 3-4 Properties of MAPbI3 and FAPbI3……………………………………11 3-5 Chemical stability of perovskites ……………………………………12-16 3-6 Photochemical stability of perovskites………………………………17-22 4. Perovskite structure ABX3, and the role of each component……………..23 4-1 Effect of ‘A’ type cation on performance of Perovskite PV…………23-25 4-1-1 Effect of mixed-organic-cations…………………………...26-33 4-2 Effect of ‘B’ type cation on performance of Perovskite PV………….34-37 4-3 Effect of ‘X’ type anion on performance of Perovskite PV………….38-39 5. Perovskite devices………………………………………………………….40 5-1 Design architecture of perovskite devices……………………………..40 5-1-1 Functions of components of perovskite solar cell……….......41

6. 7. 8. 9.

5-2 Hole conductor-free perovskite devices……………………………..42/44 5-3 Crystallinity……………………………………………………………45 Discussions and Conclusions…………………………………………….46/48 Acknowledgement…………………………………………………………48 The list of definitions, symbols and chemical structures…………………..49 The structure of some important molecules………………………………..50

References…………………………………………………………………….....51/53

 

 

3  

1- Introduction 1-1 Structures of perovskites: The name perovskites refers to a group of materials that share the same crystal structure as calcium titanium oxide (CaTiO3). Perovskite is the name of a naturallyoccurring mineral first discovered by Gustav Rose in 19th century in Ural Mountains of Russia. This mineral is named Perovskite after Russian mineralogist L. A. Perovski (1792-1856) (ref. 1). All photovoltaic perovskite materials have the general chemical formula of ABX3. In this formula ‘A’ and ‘B’ are two cations of different sizes (A cations are larger than the ‘B’ cations), and X is an anion (usually oxygen) that bonds to ‘A’ and ‘B’ cations. Perovskites have a cubic structure and in an idealized cubic structure, ‘A’ cations are positioned at the cube corner, ‘B’ cations at the body center and oxygen atoms at face center of the perovskite cube structure. Cation ‘A’ is an alkaline earth or rare earth element as Calcium, Cesium and Natrium and cation ‘B’ could be a 3d, 4d or 5d transition metal as Ti and Fe. In this cubic structure cations ‘A’ are surrounded by 12-anions in the cubo-octahedral coordination and cations ‘B’ are surrounded by 6-anions in the octahedral coordination (figure 1-1a). The perovskite lattice arrangement is also shown in figure 1-1b, but in crystallography as many other structures, it can be represented in multiple ways. The simplest way to think about a perovskite lattice structure is as a large atomic or molecular cation (positively charged) of type ‘A’ in the centre of a cube. The corners of the cube are then occupied by atoms ‘B’ (also positively charged) and the faces of the cube are occupied by a smaller atom X (not shown in figure 1-1b) with negative charge (anion). The relative atom (ion) sizes of A and B are very important for the stability of the ideal cubic-symmetry structure. This means that small strain and distortion can produce several lower-symmetry distorted versions, in which the coordination of A, B or both is reduced. A large number of metallic elements could be stable in the perovskite structure if the tolerance factor is in the range of 0.75-1.0 (ref. 5). This is generally discussed with the Goldschimidt tolerance factor since 1926 (ref. 2). For tolerance factor see the list of definitions A: cation (e.g. Ca, Ce, Na) B: cation (e.g. Ti, Fe) X: anion (usually O or Cl, Br, I)

Figure 1-1: (a) perovskite cubic structure, (b) perovskite lattice structure (ref. 4)

Crystals of perovskites could be black, brown, gray and orange to yellow color. These crystals appear as cubes, but they are actually pseudocubic.

 

4  

In recent years scientists have discovered, by choosing suitable materials for the components A, B and X (O in figure 1a), that it is possible to create a stable material which has the ability of converting the light into electricity. The choice of material combinations will also be crucial for determining the optical and electronic properties of the perovskite structure (e.g. band gap and absorption spectra, mobility, diffusion length, etc.). All of these factors will be discussed in the next sections.

1-2 Shockley-Queisser-limit Not all the incoming sunlight can be converted into the electricity. The average amount of the sun’s radiation that penetrates the atmosphere and reaches the earth is 51% of the total incoming energy (30% is reflected back into space and 19% is absorbed by the atmosphere and clouds (ref. 73)). From this 51% of the radiation that reaches the earth, theoretically calculated by Shockley and Queisser (SQ), for a silicon solar cell only 33% can be converted to the electricity (SQ limit). For SQ limit see definitions list

The ratio of the light energy, which is able to induce electric current in the cell, to the total light energy that reaches the cell, is referred to as the efficiency of a cell. The process, in which the energy of a photon from the sun is transformed into other forms of energy as electricity or heat, is called “Absorption of electromagnetic radiation”. This is what happens in photovoltaic devices. Photovoltaic devices basically work by absorption of light by a semiconducting or light absorbing materials used in them. Not all light that reaches a solar cell can be used for electricity generation. The ability to induce electric current in a solar cell depends on the wavelength of the sunlight and the band-gap of the semiconductor material. For a semiconductor electron to move into a current circuit, its energy level must be increased from its valence level (tightly bound to atom) to its higher conduction level (free to move around). The amount of energy needed to send a valence electron to a higher level is called the “band-gap” energy. Only photons with at least the band-gap energy will be able to create a current. Sunlight photons with lower energy level than the band-gap energy will pass through the solar cell and photons with excess band-gap energy can generate a current (free electrons and holes), but their extra energy gets dissipated as heat in the solar cell. Maximum theoretical solar cell efficiencies vary with the difference in the bandgap of the semiconductor material used in the solar cells. Figure 2 shows the solar radiation spectrum. In this spectrum only the mustard colored photons can create electricity in a crystalline silicon cell (this area is a picture of the SQ Limit applied to silicon as Shockley and Queisser calculated in 1961). The yellow colored wavelengths have too much energy and the red color ones do not have enough energy to create electricity. Perovskite solar cells exhibits a relatively wide absorption from 250 nm to 800 nm. A lot of effort of scientists in the field of perovskite solar cells is to push the value of the optical absorption of perovskite materials to the infrared part of the solar spectrum.

 

5  

Figure 1-2-1: Solar radiation spectrum (picture taken from ref. 73)

 

6  

2- The rise of perovskites in photovoltaics Perovskite Solar Cells have shown a rapid efficiency improvement in PCE’s from 2.2% in 2006 to 20.1% in 2015 (ref. 6). This very high PCE is related to the excellent optoelectronic properties of this material such as: direct band gap, broad absorption spectra, and high mobility of charge carriers (ref. 38). Photovoltaic technologies are generally divided into two main categories: wafer-based PV’s or first generation PV’s, and thin-film solar cells based on thin-film technologies (figure 2-1). Both single and multi-crystalline silicon solar cells and Galium Arsenide (GaAs) single junction solar cells belong to the first generation photovoltaics (ref. 52). The latter seem to have the highest efficiency among all single junction waverbased PV’s. First generation PV (c-Si) includes cells consisting Silicon or Germanium. These materials are usually doped with phosphorus or Boron. Silicon cells have a relatively high efficiency and dominate the commercial market but they are expensive due to their energy-required fabrication process. Crystalline silicon PV modules have fallen in price from $76.67/W in 1977 to $0.4-0.5/W in early 2015 (ref. 52). Thin-film PV’s includes second-generation and third generation PV’s. Their production is about 10-100 times more efficient than silicon solar cells, due to the use of films of only a few microns thick, and they are made out of organic and inorganic materials. Hydrogenated amorphous silicon, cadmium telluride (CdTe), copper indium gallium (di) selenide (CIGS) cells and copper zinc tin sulphide (CZTS) all belong to the second-generation thin-film PV’s. Cadmium telluride (CdTe) technology has been successfully commercialized, with more than 20% cell efficiency and 17.5% module efficiency record.  

The third generation PVs include dye-sensitized solar cell (DSSC), organic photovoltaic (OPV), quantum dot (QD) PV and perovskite PV. Perovskite Solar Cells include a perovskite material as their active layer. The most common structure of this active layer is an organic-inorganic lead or tin halide based material.

Figure 2-1: Classification of photovoltaics (taken from ref. 52)

 

7  

Since the discovery of high-temperature superconductivity in layered copper oxide perovskites by Bednorz et al. in 1994 (ref. 10), there was a considerable technological interest in this class of materials. At that time only a few other examples of conducting layered perovskites were known such as (La1-x Srx)n+1MnnO3n+1 and Lan+1NinO3n+1 by Mohan Ram et al. in 1986 and Ban+1PbnO3n+1 by Cava et al. in 1992 (ref. 11). With increasing the number of perovskite layers (n), all of these materials exhibit a semiconducting to metallic behavior. Methylammonium lead triiodide (CH3NH3PbI3) and formamidinium lead triiodide (CH(NH2)2PbI3) are the most used perovskites in the research area. The replacement of bromine with iodine to form the CH3NH3PbI3 perovskite, has improved the efficiency of perovskites solar cells to 3.8% in 2009. In 2012 with replacing of liquid electrolyte to solid-state hole-transporting material spiro-OMeTAD, the efficiency of perovskite solar cells was improved to 9.7%. The typical perovskite solar cells are fabricated using compact and mesoporous TiO2 as electron transporting layer and spiro-OMeTAD as the hole-transport material (ref. 39). During the past five years (figure 2-2), these materials have shown a rapid efficiency (PCE) improvement up to 20% in 2015 (ref. 33). Maximum efficiency of a single junction solar cell could be 33% calculated by Sockly-Queisser. Theoretically this limit is about 31% for perovskite solar cells (ref. 8). Researchers try to push the performance of perovskite solar cells nearer to the Shockly-Queisser limit (ref. 7). The cell efficiencies of perovskite are approaching those of commercialized secondgeneration technologies such as CdTe and CIGS. Other emerging photovoltaic technologies are still struggling with lab cell efficiencies lower than 15%. Because the Perovskite Solar Cells are cheap and simple to fabricate, they could be very attractive for the market commercialization. Some companies already promised modules on the market by 2017 (ref. 9). High and rapidly improved efficiencies, as well as low potential material and processing costs are not the only advantages of perovskite solar cells. Flexibility, semi-transparency, and light-weight are other value propositions of perovskite solar cells. From the past years, many research groups have been investigating the perovskite materials to make them a more stable and efficient light harvesters. Scientists do this for example by choosing the suitable material of A, B and X compounds in the ABX3 structure formula of perovskites. Some of the research groups were working on the morphology of the perovskite layer by developing of some synthesis approaches for known perovskite materials to improve the performance of perovskite solar cells.

 

8  

Figure 2-2: The important evolutions of perovskite solar cells(ref. 52)

Further in this thesis some of most important achievements of scientists in this field will be discussed.

 

9  

3- Most prominent perovskites 3-1 Methylammonium lead triiodide Methylammonium lead triiodide (CH3NH3PbI3 Or MAPbI3) was the first absorber with the perovskite structure that was investigated for potential use in a solar cell (Peplow, 2014). Methylammonium lead triiodide (CH3NH3PbI3) is the most widely used hybrid perovskite solar cell material, where MA is a positively charged organic cation at the center of a lead iodide lattice structure (figure 4-1-1a and b). This perovskite material is employed as a light sensitizer in mesoporous dye cells, as transport layer in a solid-state dye-cell architecture and most recently as the bulk material in a standard planar thin-film solar cell (Park et al.)

3-2 Formamidinium lead triiodide (FAPbI3) Formamidinium lead triiodide (CH(NH2)2PbI3) is another interesting perovskite absorber material for study on the perovskite photovoltaics. The highest power conversion efficiencies (PCE’s) have been obtained with formamidinium lead triiodide FAPbI3. These formamidinium-based perovskite materials are less stable than methylammonium based perovskites. In this thesis the classical synthesis of MAPbI3 will be discussed.

3-3 Synthesis of CH3NH3PbI3 (ref. 61) The synthesized methylamonium iodide CH3NH3I reacts with Pb(CH3COOH)2.3H2O in hydroiodic acid (HI) in air. According to the stoichiometric ratio of Pb(CH3COOH)2.3H2O (37.933 g, 0.1 mol) and CH3NH3I (15.9 g, 0.1 mol), Pb(CH3COOH)2.3H2O we have to dissolve this in 250 ml aqueous HI in a 500 ml round bottom flask, under constant stirring, to form a yellow solution. Then CH3NH3I must to be added to the yellow solution. With the decrease of the solution temperature from 65°C to 40 °C the black and shiny crystals of CH3NH3PbI3 will be formed in the bottom of the flask after several days. The crystals must to be washed and filtered with HI and then acetone. In the case of CH3NH3PbI3; the temperature must not be lowered below 40 °C because there the formation of colorless crystals of CH3NH3PbI3.H2O begins. Color of the perovskite type phases: X = CI: colorless; X = Br: orange; X = I: black. (ref. 62)

3-3-1 Synthesis of CH3NH3I (ref. 61) Methylamine CH3NH2 solution in a slight excess reacts with hydrochloric acid (HI) in an ice bath in ambient atmosphere. The crystallization of methyl-ammonium iodide (CH3NH3I) can be achieved by using a rotary evaporation. A white microcrystal CH3NH3I will be formed, which is washed several times with absolute diethyl ether and finally dried at 60 °C in a vacuum oven overnight.

 

10  

3-4 Properties of MAPbI3 and FAPbI3 Because of the broad absorption of the solar spectrum of formamidinium lead iodide (FAPbI3), this perovskite material can potentially provide better performance than methylammonium lead iodide (MAPbI3). However, it is more difficult and complicated to form stable and high quality perovskite films with FAPbI3 than with MAPbI3 (ref. 32) An important distinction between inorganic (e.g. Cs) and organic hybrid perovskite methylammonium lead halide is the difference symmetry of the A cation, which is reduced from inorganic to hybrid. Methylammonium cation has the C3ν point group and the associated highest-symmetry perovskite structure is pseudocubic. The charge balancing of the perovskite composition formula ABX3 can be achieved only as the sum of the valence number of the two cations of A and B is three. So for CsSnI3, the only viable ternary combination is I-II-X3 (Cs+1 and Sn+2). In hybrid halide perovskite methylammonium lead triiodide, a divalent inorganic cation is present with a monovalent metal replaced by an organic cation of equal charge. In principle any single charged cation could be used, once there is sufficient space to fit it within the inorganic cage. A too large organic cation would cause the expansion of the whole of perovksite lattice (the 3D perovskite structure would be broken), and a too small cation would contract the 3D structure. The effect of cation A on the perovskite 3D structure is further discussed in section 4-1 of this thesis.

 

11  

3-5 Chemical stability of perovskites A big challenge in the topic “perovskites”, is the short-term and long-term stability. Methylammonium lead tri-iodide (MAPbI3) is a proto-typical example of a known perovskite material. The perovskite structure MAPbI3 is a black material and can also exist in white or yellow phases. One weakness of these perovskite materials is their inherent vulnerability to moisture and heat (ref. 46, 51). Perovskites are unstable and highly degradable upon exposure to moisture of air because of the water solubility of their organic constituent (ref. 37). Their color can be changed to yellow and white in higher temperatures. There are some reports of hybrid perovskites reacting irreversible with lewis bases as ammonia (ref. 50). Jarvist M. Frost et al. (ref. 51) reports the simple acid-base reaction shown in figure 3-5-1. In this reaction pathway, a single water molecule is sufficient to degrade the perovskite material. But to dissolve the HI and CH3NH2 byproducts, an excess of water is required. As a result of this reaction in a closed system, traces of water will partial decompose the hybrid perovskite until the HI has saturated the H2O or the vapor pressure of CH3NH2 has reached equilibrium. In exposure of sufficient water, the material can degrades completely to form PbI2. In reversible reaction mechanism, the ionic salt and strong acid HI, would form a separate ionic salt with NH3. This mechanism should cause a proton transfer throughout the material resulting in three following outcomes: (a) evaporation of methylammonia, results the inclusion of NH4 in the perovskite and the simultaneous production of HI; (b) the system decomposes to form HI, PbI2, and volatile organics; (c) the reaction of CH3NH3+/CH3NH2-NH3/NH4+ occurs on the surface causing the material to form the neutral CH3NH2. In this reversible reaction ammonia is more volatile and lighter than methylammonia. However, the reversible formation of hydrates (ref. 62) and ammoniates (ref. 50) has been well described. The exposure of a thin MAPbI3 film to NH3 vapor results in reversible black/white color changes within seconds! Heating above 40˚C, washing with dichloromethane, drying or grinding can remove the water molecule from the hydrate turning the white MAPbI3.H2O material into black MAPbI3.

Frost reported (ref. 51), there have been no reports of hybrid perovskites made of aprotic organic ions, such as tetramethylammonium (CH3)4N+. Such a material would not undergo the degradation pathway depicted in figure 3-5-1 and so may be more chemically stable in presence of water.

Figure 3-5-1: Possible decomposition pathway of hybrid halide perovskites in the presence of water. a) A water molecule: is required to initiate the process of decomposition. b) HI (hydrogen iodide): is soluble in water. c) methylammonia: is volatile and soluble in water. d) PbI2: yellow solid

 

12  

Constantinos C. Stoumpos et al. reported (ref. 68), the Pb-containing perovskite phases are stable in air for months, though they lose their crystalline luster by humidity after a couple of weeks. They reported; that Sn-containing perovskite materials are more sensitive to air and moisture and partially decompose within 2 hour before total decomposition after 1 day. A general trend is that methylammonium-containing materials are considerably more stable than formamidinium-containing ones. This is based on the respective stability of the cations themselves and partially the tendency of CH(NH2)2+ to dissociate to ammonia and sym-triazine (ref. 72). Due to efforts of scientists working on the stability of these perovskite materials, there are some useful methods making these materials more stable. Any modification, which enhances the robustness of the material, will be advantageous. Severin N. Habisreutinger and his collages from the University of Oxford (ref. 37) developed a smart approach for achieving long-term stability of high efficiency perovskite solar cells. They mitigated thermal degradation of perovskite PV’s by replacing the organic hole-transport material with polymer-functionalized single-walled carbon nanotubes (SWNTs) embedded in an insulating polymer matrix, which yields efficient solar cells with greatly enhanced stability to thermal stressing and water exposure. They did this work by protecting the hole-transport material (HTM) that forms the outermost layer of a perovskite solar cell from atmospheric moisture and by “sealing in” the volatile perovskite components, they prevented the thermally induced loss of the organic molecules. The principle of this protection depends on the hydrophobicity, permeability and density of the hole-transporting material. They coated perovskite films with three commonly employed hole-transporting materials, 2,2’,7,7’-tetrakis— (N,N-di-p-methoxyphenylamine)9,9’-spirobiflourene (spiro-OMeTAD), poly(3hexylthiophene) (P3HT) and poly(triarylamine) (PTAA). They used CH3NH3PbI 3-xClx as perovskite layer and heated the perovskite films coated with these hole-transporters to a temperature of 80 °C on a hot plate in ambient air. The first signs of degradation were visible after 24 h. After 96 h, all perovskite films coated with organic hole-transporter spiro-OMeTAD became predominantly yellow, where the degradation had progressed significantly (figure 3-5-2)

Figure 3-5-2: Illustration of the protective effect of various materials on top of perovskite films (CH3NH3PbI3−xClx) under thermal stressing. Photo illustrating the visible degradation of the perovskite layer. The color shifts from almost black to yellow for all organic HTLs except for the films covered with PMMA only or a composite of carbon nanotubes and PMMA. (Picture taken from ref. 37)

 

13  

This color and structural change could be initiated by the forming of the weak hydrogen bonds between the water molecules and the highly hygroscopic methylammonium cations of the crystal structure, leading to a bond dissociation of the crystal constituents. Here as a consequence, the unbound methylammonium iodide can escape the compound structure leaving behind a residual layer of lead iodide. This was corresponding with the results of the X-ray diffraction. The X-ray diffraction (XRD) patterns indicated that the degradation and color change of perovskite layer was due to a loss of methylammonium iodide, because the diffraction patterns had transitioned from perovskite to PbI2 (figure 3-5-3). This observed degradation was faster for a film of spiro-OMeTAD doped with Li-TFSI than for undoped spiro-OMeTAD. This means that hygroscopic nature of the lithiumbased dopant may have an important role in introducing moisture into the perovskite layer, accelerating its degradation.

Figure 3-5-3: X-ray diffraction pattern of devices before (black line) and after (red line) 96 h heat exposure. The spectral changes from the diffraction features characteristic for the CH3NH3PbI3-xClx at 14.28° and 28.57° (black star) to a new diffraction feature at 12.80° associated with PbI2 indicate degradation of the perovskite crystal structure. For the neat PMMA layer and the PMMA-nanotube composite HTL, there are no discernible changes in the diffraction patterns.

The decomposition of the perovskite structure is also clearly visible in the absorption spectra of perovskite devices with Li-spiro-OMeTAD as hole-transport material (figure 3-5-4). Here in these spectra before heat exposure, the sample shows a strong absorption in the visible range of 750 to 450 nm, which is characteristic of CH3NH3PbI3-xClx. This absorption feature is not visible in the spectra taken after heat exposure and degradation. This correlates with a visible color change of the films in figure 3-5-2, which becomes almost transparent in the red region of the spectrum.

 

14  

Figure 3-5-4: Changes in the absorption spectra taken at 24 h intervals. The sample with Li-spiroOMeTAD, being representative of the degrading organic HTLs, loses the perovskite characteristic absorption onset (indicated by the gray arrow).

In X-ray diffractograms taken before and after heat exposure (figure 3-5-3), the discoloration of the perovskite films are confirm by the characteristic strongest diffraction peaks at 14.28° and 28.57°. This correlates with a breakdown of the perovskite crystal structure. The residual films of perovskites exhibits X-ray diffractions of 12.80°, which is characteristic of the neat lead iodide. To protect this perovskite materials from degradation process, insulating poly(methyl methacrylate) (PMMA) is employed, which is hydrophobic and is capable of inhibiting the intrusion of moisture into the perovskite structure. This PMMA protective layer can also simultaneously inhibit the evaporation of the methylammonium iodide. The effectiveness of PMMA is clearly visible in figure 3-52, in comparison to the other three standard hole-transport materials. The color of perovskite sample protected with PMMA is not changed, even after 96 hour at 80° C in air. PMMA is not capable of transporting charge on its own because of the lacking π-conjugation. Thus acts as an insulator and can fully inhibit the perovskite degradation process. To keep the PMMA a good insulator while it can possess selective charge collection properties similar to the organic hole-transporters, they combined PMMA with a highly charge transporting component as single-walled carbon nanotubes (SWCN’s). These materials are good candidates due to their high conductive characteristic as well as their structural and chemical stability (ref. 44). They coated perovskite films with P3HT/SWNT’s and PMMA. Carbon nanotubes are generally insoluble, but when they wrapped with a monolayer of P3HT, they form supramolecular nanohybrids (P3HT/SWNT) that are dispersible in common solvents and can be deposited by spin-coating (ref. 45). The composite structure of P3HT/SWNTs and PMMA appears to be a good conducting material as well as protect the perovskite layer resulting in no traces of degradation of the perovskite layer in the absorption and X-ray diffraction patterns.

Developing non-hygroscopic hole-transporting materials as spiro-OMeTAD for use in perovskite solar cells is very important. None of the three employed hole-transport materials were effective protections against thermal degradation of methylammonium lead halide perovskite films in air without PMMA. With this combination (MAPbI3 with P3HT/SWNT-PMMA) they reported a PCE of up to 15.3% with an average efficiency of 10% ± 2%.

 

15  

Figure 3-5-5: Schematic illustration of the solar cell with a carbon nanotube/polymer composite as hole-transporting structure. Schematic architecture of the investigated device consisting of sequential layers of FTO as transparent electrode, a TiO2 compact layer, a mesostructured layer of Al2O2 coated with CH3NH3PbI3-xClx and the hole transporting structure composed of a P3HT/SWNT layer in-filled with a PMMA matrix.

 

16  

3-6 Photochemical stability of perovskites Beside moisture instability, devices with an embodiment of mesoporous TiO2 layer as electron transporter material, sensitized with perovskite absorber, exhibits a UV-light induced instability. The cause of this instability is linked to the interaction between photo-generated holes inside the TiO2 and oxygen radicals on the surface of TiO2 (figure 3-6-1). To justify the degradation mechanism, Leijtens T. et al. (ref. 63) considered the surface chemistry of TiO2. TiO2 contain many oxygen vacancies (or Ti3+ sites) at the surface, which are electron-donating sites. The electrons of these sites, interact with oxygen molecules in the atmosphere, which adsorbs to the oxygen vacancy sites, resulting a charge transfer complex (O2 --Ti4+). One electron-hole pair is formed, due to the band-gap excitation of TiO2. The hole in the valence band, recombine with the electron at the oxygen adsorption site, resulting a free electron in the conduction band and a positively charged, unfilled oxygen vacancy site at the surface of TiO2. They designed perovskite devices containing p-doped spiroOMeTAD as hole-transfer material, which has excess holes and will readily recombine with the free electrons left behind from the band-gap excitation of the TiO2.

Figure 3-6-1: Proposed mechanism for UV-induced degradation. (a,b) Upon UV light exposure, the photogenerated holes react with the oxygen radicals adsorbed at surface oxygen vacancies. (c) Molecular oxygen is desorbed from these sites, leaving unoccupied, deep surface traps sites and a free electron per site. These electrons will recombine with the excess of holes in the doped-hole transporter. Upon photo excitation of the sensitizer, electrons are injected either (1) into the conduction band from which they become deeply trapped or (2) directly into the deep surface traps. (d) These deeply trapped electrons are not mobile and recombine readily with holes on the spiro-OMeTAD hole transporter.

Leijtens T. et al. reported; this instability does not arise in mesoporous TiO2-free mesosuperstructured solar cells and that their TiO2-free cells deliver stable photocurrent. To prove this and design a perovskite device without a TiO2 layer, they performed a series of aging tests on TiO2-SSC’s (sensitized solar cells), using both dye sensitizers indolene dye (D102) and ruthenium complex dye (C106) and methylammonim lead iodide chloride mixed halide perovskite absorber in different environments. The TiO2-SSC’s encapsulated in nitrogen atmosphere had an extremeem decay in photocurrent and performance under simulated sunlight with the full UV component present. They described the mechanism of this drop in performance and proposed that oxygen serves to pacify deep electronic trap sites at  

17  

oxygen vacancies in the TiO2. This was the most important reason for them to remove the deep trap sites and actually to entirely remove of mesoporous TiO2 and design a planar heterojunction perovskite solar cell without a TiO2 layer (figure 3-6-2b). In this design of solar cell perovskite itself functions as both light absorber and electron transporter material.

Figure 3-6-2: Solar cell architectures. (a) Depiction of a solar cell made by sensitizing mesoporous TiO2 with the mixed halide perovskite absorber. This active layer is then infiltrated with spiroOMeTAD as HTM. (b) Schematic of a solar cell where the perovskite itself is functioning as both light absorber and electron transporter. The Al2O3 is functioning as a scaffold to enhance film formation, while the spiro-OMeTAD HTM is functioning as a hole extraction layer.

To investigate the UV aging effect on the perovskite-based solar cells, Leijtens. T. et al. (ref. 63) have exposed a variety of solar cell configurations to 5h of simulated AM 1.5 100 mWcm-2 sunlight without any UV filtration, as depicted in figure 3-6-2. Figure 3-6-2a, shows a perovskite solar cell composed of 500 nm mesoporous TiO2 layer sensitized by the perovskite as absorber material and infiltrated with the spiroOMeTAD hole-transporter material. In this structure perovskite CH3NH3PbI3-xClx is present only as a thin layer on TiO2 electron-transporter nanoparticles, Figure 3-6-2b, shows the structure of most efficient meso-superstructured solar cell (MSSC) of Leijtens’s group. In this structure a thin inert alumina scaffold is used to allow the formation of a thin solid perovskite-capping layer, forming a planar heterojunction with spiro-OMeTAD hole-transporte material. In this structure electrons and holes are transported within the perovskite layer. They used gold electrodes here rather than silver electrodes, because silver can become corroded in contact with the perovskite film due to the formation of silver halide. The performance of the designed TiO2-based perovskite SSC’s were measured under three difference conditions; in air, encapsulated in a nitrogen-filled glove box, and encapsulated but with the additional of a 435 nm UV cutoff filter. Figure 3-6-3 shows the evolution of solar cell performance parameters of PCE, FF, Jsc and Voc over a

 

18  

time period of 5 hour, under illumination for TiO2-based perovskite SSC’s in above conditions. All devices had initial power conversion efficiencies of at least 5%, with short circuit currents of at least 10 mA cm2, representing typical working solar cells of this type. Surprisingly, the encapsulated sensitized TiO2 cells had a much faster decay than the non-encapsulated solar cells when they were subjected to full spectrum sunlight. They were falling to a performance of less than 10% of their initial performance within 5 hour due to their rapid decrease in photocurrent and a smaller decrease in photovoltage. For the non-encapsulated devices tested in ambient conditions, this was in contrast to only 50% decay over 5 hour period of time. When a UV-filter with a 435 nm cutoff was used with the encapsulated sensitized TiO2, solar cells decayed to only 85% of their initial performance, however the solar cells were more stable. Notably for the 435 nm cutoff UV-filtered devices, de degradation of the photocurrent was diminished, while the photovoltage was constant and the fill factor decreased slightly. The reason of dropping of the fill factor may simply be due to nonoptimum doping of the HTM.

Figure 3-6-3: Aging of TiO2-based solar cells. Evolution of normalized solar cell performance parameters, power conversion efficiency, short circuit current (Jsc), fill factor (FF), and open circuit voltage (Voc) over 5 h of AM1.5 100 mWcm2 solar illumination. The comparing of perovskite sensitized TiO2 solar cells which are non-encapsulated (blue open circles) and encapsulated with (filled black squares) and without a