5Organic-type solar cells
5.1Donor-acceptor type organic solar cells
Carbon is generated by He nuclear fusion in fixed stars, and a huge amount of carbon atoms exist throughout the universe. In life-related activity such as photosynthesis, carbon compounds such as CO2 and C6H12O6 play an important role. Carbon has various hollow-cage nanostructures such as C60, giant fullerenes, nanocapsules, onions, nanopolyhedra, cones, cubes, and nanotubes. These C structures show different physical properties, and there would be great potential in studying these materials at a small scale within an isolated environment. By controlling the size, number of layers, helicity, composition and included clusters, cluster-included C nanocage structures with a band-gap energy of 0–1.7 eV and nonmagnetism are expected to show various electronic, optical, and magnetic properties such as Coulomb blockade, photoluminescence, and superparamagnetism [1]. Recently, C60-based polymer/fullerene solar cells have been investigated and reported on [2–6]. These organic solar cells can be fabricated by the use of printing methods under ordinary atmospheric conditions, and they have a potential for use in lightweight, flexible, inexpensive, and large-scale solar cells [7–12].
The photovoltaic mechanism of organic solar cells is shown in Fig. 5.1. The light is absorbed in the donor (D) layers, such as phthalocyanine and poly[3-hexylthiophene] (P3HT), and electrons are excited to form excitons from the energy levels of the highest occupied molecular orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO). Then, the excitons diffuse to the donor-acceptor (DA) interface, and the charges are separated at the interface. Separated electrons are transported in the acceptor (A) such as fullerene, holes are transported in the donor to the electrodes and the current flows.
Four factors that determine the conversion efficiencies of organic solar cells are exciton generation efficiency η1, exciton transport efficiency η2, charge separation efficiency η3 and carrier transport efficiency, as shown in Fig. 5.2 [8]. Since the total efficiency is calculated by multiplying the four efficiencies, all efficiencies should be high. Although the η1 and η3 values are high for the organic solar cells, the η2 and η4 values are low because of the very short diffusion length of the excitons.
A weak point of organic solar cells is their low conversion efficiency, which results from the recombination of excitons produced by light irradiation. In photovoltaic solar cells, excitons are generated by light irradiation. The excitons are separated into electrons and holes, and they are transported to each electrode to generate potential difference. However, if the electrons and holes of the excitons are recombined before their arrival at the electrode, light is emitted by their recombination, and no electric power is generated.
One of the causes of recombination in organic solar cells is its low carrier mobility. It takes too much time for generated excitons to reach the pn junction for the carrier separation, and the excitons recombine on the way to the pn junction or metal electrodes. In addition, since the sizes of excitons in organic materials are small, they tend to recombine.
5.2Exciton
An exciton is a pair of an excited electron and an excited hole restricted by the electrostatic Coulomb force, and it is an electrically neutral quasiparticle that exists in semiconductors and insulators, as shown in Fig. 5.3. The Coulomb force is expressed as follows (ε: relative permittivity):
The exciton is regarded as an elementary excitation that can transport energy without transporting electric charge. The current of solar cells flows only when the exciton is separated into an electron and a hole. Excitons are introduced physically from excited waves in the wave function of the biding state of electrons on the conduction and holes on the valence band. Frenkel excitons and Mott-Wannier excitons are limit models of the excited waves, and actual excitons have intermediate states between these excitons.
The wave function of Mott-Wannier excitons (weak binding of an electron and a hole) in the excited state is broader than the lattice constants. The excited state is a spread state at a lattice point, and an electron and a hole are in a bound state with weak restriction. In an excited state like the Mott-Wannier, excitons spread in crystals such as various ionic crystals and ionic semiconductors.
Frenkel excitons (comparatively strong binding of an electron and a hole) have a narrower wave function in the excited state compared to the lattice constants. The excited state is similar to the excited states of atoms or ions. An excited state like the Frenkel excitons spread resonantly through lattice points with a certain wavenumber in organic-molecular crystals.
The energy required for exciton generation is lower than the bandgap energy because of the binding energy between an electron and a hole, and the exciton is in a stable state. A sharp reflection peak can be observed at lower energy compared with that of interband transition. Excitons that spread in the hard, non-deformed lattice are called free excitons, and they can be transported freely throughout the crystal. Self-restraint excitons that spread in the vibrating lattice localize at a certain position by interaction with lattice vibration.
5.3Bulk heterojunction
One of the improvements presented by organic solar cells is donor-acceptor (DA) proximity in the devices by using blends of donor-like and acceptor-like molecules or polymers, which are called DA bulk-heterojunction solar cells [12–18], as shown in Fig. 5.4. Previous organic solar cells consisted of a simple pn heterojunction. The bulk-heterojunction is a pin junction which consists of a mixture intrinsic semiconductor layer (i-layer) between p- and n-type semiconductors. For a fullerene-based system, p-type molecular crystals are surrounded by an amorphous fullerene matrix.
When light is irradiated on the junction, excitons are generated around the p-type molecular/fullerene interface, which consists of a bulk-hetero mixture layer, and the excitons can reach the DA or pn junction by transporting several nm. Electrons and holes are separated into an n-layer and p-layer at the interface, respectively. Each carrier transports through connected crystals and the matrix to the electrode, and current flows.
One merit of ordinary heterojunction solar cells is their high efficiency when the carrier mobility and electrical conductivity of the D and A layers are high. However, only the excitons generated near the D/A interface contribute to the photocurrent. On the other hand, the interfacial area of the bulk-heterojunction is so large that the carriers are separated effectively. However, the carrier transport pass is complicated, and the carrier could not be taken away from the cells. After all, an inter-penetrated structure would be effective for carrier generation and carrier transport, and research on new structures such as nanorods [19] and nanotubes is currently in progress.
5.4P3HT:PCBM
C60-based polymer/fullerene solar cells have been investigated, and significant improvements in photovoltaic efficiencies are mandatory for use in future solar power plants. A characterization of polymer/fullerene bulk-heterojunction solar cells using different organic polymers is presented here. Poly[3-hexylthiophene] (P3HT) and poly[2-methoxy-5-(20-ethylhexoxy)-1,4-phenylenevinylene] (MEH-PPV) were used for p-type semiconductors, and 6,6-phenyl C61-butyric acid methyl ester (PCBM) was used for n-type one. Device structures were produced, and efficiencies and spectral responsivity were investigated.
A thin layer of polyethylenedioxythiophen doped with polystyrene–sulfonic acid (PEDOT:PSS) (Sigma Aldrich)was spin-coated on pre-cleaned indium tin oxide (ITO) glass plates. Then, semiconductor layers were prepared on a PEDOT layer by spin-coating using a mixed solution of P3HT, MEH-PPV and PCBM in 1,2-dichlorobenzene. The weight ratios of both P3HT:PCBM and MEH-PPV:PCBMwere 1 : 8. The thickness of the blended device was approximately 150 nm. After annealing at 100 °C for 30 min in N2 atmosphere, aluminum (Al) metal contacts with a thickness of 100 nm were evaporated as a top electrode. A schematic diagram of the presented solar cells is shown in Fig. 5.5(a) [4].
The typical current density–voltage (J–V) characteristics of a P3HT/PCBM structure in the dark and under illumination are shown in Fig. 5.5(a). Although no photocurrent was observed in the dark, a photocurrent over 5 mA cm−2 is observed under illumination. The J–V characteristics of both MEH-PPV/PCBM and P3HT/PCBM solar cell structures are shown in Fig. 5.5(a). Each structure shows characteristic curves for open-circuit voltage and short-circuit current. Measured parameters of these solar cells are summarized in Table 5.1. A solar cell with a P3HT/PCBM structure provided power convergent efficiency of 1.03%, a fill factor of 0.53 and a short circuit current density of 5.18 mA cm−2, which is better than that of a MEH-PPV/PCBM device. On the other hand, the MEH-PPV/PCBM structure showed a higher open-circuit voltage of 0.70 V.
Table 5.1: Measured parameters of solar cells.
Figure 5.5(b) shows the measured spectral photoresponses of the solar cells. The MEH-PPV/PCBM structure shows a high photoresponse in the range of 300–600 nm, while the P3HT/PCBMshows higher spectral responsivity in the range of 400–650 nm, which corresponds to 3.1 and 1.9 eV, respectively. Optimization of the nanocomposite structure with P3HT and MEH-PPV would increase the efficiencies of the solar cells.
The electronic structures of the molecules were calculated, and the energy levels of HOMO of P3HT and MEH-PPV are shown in Fig. 5.6(a) and (b), respectively. HOMO levels are observed around the five and six-membered rings in the main-chain structures of the polymers, which could be due to the charge transfer from the sulfur and oxygen atoms, respectively. The energy levels of the LUMO of PCBM are also shown in Fig. 5.6(c), and the LUMO levels are observed around a C60molecule with high electron negativity. Effective formation and separation of excitons in the P3HT/PCBM system could be due to the nanocomposite structure, and the separated carriers could transfer from P3HT to C60, as has been reported previously [7]. Interdiffusion of PCBM into the P3HT network would lead to the existence of C60 molecules within the exciton diffusion radius of the P3HT network.
An energy-level diagram of P3HT/PCBM solar cells is summarized in Fig. 5.6(d). Previously reported values were used for the energy levels of the figures by adjusting them to the present work [15, 20–22]. An energy gap of 1.9 eV, which is an estimated value from Fig. 5.5(b), is used for the model. The relation between VOC and polymer oxidation potential has been reported as follows:
Where e is the elementary charge [21, 22]. The value of 0.3 V is an empirical factor, and this is enough for efficient charge separation [23]. The present model agrees with this equation, and control of the energy levels is important to increase the efficiency. Combination of the present solar cells and boron nitride nanomaterials with various direct band gaps might be effective for an increase in efficiencies [24]. The performances of these solar cells could also be linked to the nanoscale structures of the polymer materials, and the control of the nanostructure should be investigated further.
5.5Phthalocyanine dimer
Phthalocyanines, which exhibit photovoltaic properties, heat resistance, light stability, chemical stability and a high optical absorption at visible range, are used as an oxidation catalyst, a catalyst in fuel cells and solar cells. Many studies on the metal phthalocyanine (MPc)monomers have been performed [25, 26], and the properties are strongly dependent on central metals or chemical substitutions. Organic–inorganic hybrid device structures were produced, and nanostructure, electronic property and optical absorption were investigated. When the nearest two neighboring phthalocyanines with substituents such as amino group and hydroxy group are connected by a hydrogen bridged substituent, high photoconduction has been observed [27, 28]. However, few phthalocyanine dimers have been reported, and high photoconduction can be expected for the covalently bridged phthalocyanine dimers. The purpose of the study presented here was to fabricate and characterize phthalocyanine dimer/fullerene HJ solar cells. Here, μ-oxo bridged gallium phthalocyanine (GaPc) dimer was used for p-type semiconductors, and fullerene with an excellent electron affinity was used for the n-type ones. The molecular orbital of GaPc dimer and fullerene was investigated as a solar cell material [12].
GaPc monomer with axial Cl ligand was also investigated for the comparison. The measured J–V characteristic of ITO/PEDOT:PSS/GaPc dimer/C60/Al solar cell and ITO/PEDOT:PSS/GaPc/C60/Al showed a characteristic curve for open circuit voltage and short circuit current density. All parameters were improved using GaPc dimer compared with GaPc monomer. Figure 5.7(a) shows a measured optical absorption of GaPc dimer, C60 and GaPc dimer/C60 cells [29]. The solar cells show a wide optical absorption ranging from 320 to 800 nm, which corresponds to 3.8 and 1.5 eV respectively. The absorption spectrum of the GaPc dimer was almost the same as that of the monomer, but a new peak was observed at ~ 450 nm. The fluorescence (FL) spectra of GaPc dimer and GaPc dimer/C60 thin films are shown in Fig. 5.7(b), and the excitation wavelength was 300 nm. A FL peak of GaPc dimer disappeared after formation of GaPc dimer/C60 HJ thin films. It is believed that carriers would be effectively transported from GaPc dimer to C60.
Figure 5.7(c) shows the structure of the μ-oxo-bridged gallium phthalocyanine dimer used here. Two GaPc planes are parallel to one another, and the degree of rotation is 41.35 ° [29, 30]. The plane distance between the GaPc monomer is ~ 0.34 nm. When the nearest two neighboring phthalocyanines are arranged with hydrogen bridged substituent, high photoconduction can be expected for the covalently-bridged phthalocyanine dimer.
The energy level diagram and electronic structures of the solar cell were calculated and summarized as shown in Fig. 5.7(d) and (e). The HOMO and LUMO levels of GaPc, and HOMO and LUMO of two phthalocyanine monomers were stirred and piled up, respectively. The interaction between two phthalocyanine monomers was not able to be confirmed. Carriers could transport from −4.5 eV to −4.3 eV by hopping conduction. Figure 5.7(e) shows HOMO and LUMO energy levels of the GaPc dimer with C60 after structural optimization using DFT/6-31G*. The electronic densities of LUMO, LUMO+1, and LUMO+2 are localized for the fullerene side, while the HOMO is localized for the GaPc-dimer side, which suggests electron transfer between the GaPc dimer and fullerene. The similar localization of frontier orbital was previously reported for the other donor-fullerene systems [31, 32]. A schematic diagram of the energy levels of GaPc dimer, C60 and GaPc with C60 showed that the LUMO levels of the GaPc dimmer with C60 are comparable to the LUMO levels of fullerene, and that the HOMO levels of the GaPc dimer with C60 are close to the HOMO levels of GaPc dimer. However, the symmetry of the GaPc dimer seems to be reduced due to a decrease in degeneracy, which could be due to the interaction with C60.
Although the energy gap and energy level of GaPc dimer were hardly changed by dimerization here, the power conversion efficiency was significantly improved. The improvement of efficiency could be due to the decrease of career recombination in the ordered molecular orientation by dimerization. As a result, open-circuit voltage was greatly improved, which led to a high conversion efficiency.
The X-ray diffraction pattern of the GaPc dimer layer showed a peak of lattice spacing of 1.27 nm. Various crystallizations of μ-oxo bridged GaPc dimer have been reported [27, 28]. When the crystallographic structure is different, the initial surface potential, photosensitivity and residual surface potential are also different. Further investigation of the crystallographic structure should be carried out in the future.
5.6ZnTPP:C60
The fabrication and characterization of porphyrin:C60 BHJ solar cells are presented here. 5,10,15,20-tetraphenyl-21,23H-porphin zinc (ZnTPP) was used for p-type semiconductors [5, 33] and C60 was used for n-type semiconductors. Porphyrin has high optical absorption in the visible spectrum and high hole mobility [34–36] and was expected to form cocrystallites [37, 38] with C60 that would be suitable for the BHJ structure [39, 40]. The second purpose of the investigation was to look at the effects of the electron transport layer (ETL). 3,4,9,10-perylenetetracarboxylic dianhydride (PTCDA) is a perylene derivative with a simple structure, which was reported to be used for solar cells [41]. In the present work, PTCDA was used as the ETL for porphyrin/C60 BHJ solar cells. The ETL prevents hole transfer between an active layer and an electrode, and an improvement in conversion efficiency was expected through the introduction of the ETL. A thin layer of PEDOT:PSS was spin coated on pre-cleaned ITO glass plates. The PEDOT:PSS serves as a hole transport layer (HTL) for an electron blocking layer. Then, semiconductor layers were prepared on a PEDOT layer by spin coating using a mixed solution of C60, ZnTPP in 1 mL odichlorobenzene. The total weight of ZnTPP:C60 was 18 mg, and the weight ratio of ZnTPP to C60 was in the range of 1 : 9 ~ 5 : 5. The thickness of the blended device was ~ 150 nm. To increase efficiencies, PTCDA with a thickness of ~ 20 nm was also added over the active layers. After annealing at 100 °C for 30 min in N2 atmosphere, PTCDA was evaporated between an active layer and a metal layer. Finally, Al metal contacts were evaporated as a top electrode.
Figure 5.8(a) and (b) shows the optical absorption of C60, ZnTPP, ZnTPP:C60 and ZnTPP:C60/PTCDA BHJ solar cells, respectively. The ZnTPP:C60/PTCDA structure provided absorption in the range of 300–800 nm (which corresponds to 4.0 and 1.5 eV respectively), which was higher than that in the ZnTPP:C60 structure.
The J–V characteristics of ZnTPP:C60 BHJ solar cells measured under illumination are shown in Fig. 5.8(c). The BHJ indicates one-layered composite structures with p and n-type semiconductors, denoted as ZnTPP:C60. The effects of the addition of PTCDA to the ZnTPP:C60 BHJ solar cells was also investigated and this is denoted as ZnTPP:C60/PTCDA. Each structure shows a characteristic curve for open circuit voltage and short circuit current. The current density of ZnTPP:C60 increased through the addition of PTCDA, and the best efficiency was obtained for the ZnTPP:C60/PTCDA sample. Exciton migration of C60 can be efficiently suppressed using PTCDA, and excitons would be generated for both ZnTPP/C60 and C60/PTCDA interfaces, which results in an increase in conversion efficiency, as shown in Fig. 5.8(c). A schematic illustration of electron and hole transport is shown in Fig. 5.8(d) and (e).
The X-ray diffraction patterns of the ZnTPP and ZnTPP:C60 BHJ layers are shown in Fig. 5.9 (a) and (b) respectively. In Fig. 5.9 (a), diffraction peaks corresponding to ZnTPP crystal are observed. After the formation of the ZnTPP:C60 BHJ layer, the diffraction peaks corresponding to ZnTPP disappeared, and C60 peaks are observed as shown in Fig. 5.9(b). In addition, a new diffraction peak is observed as indicated by an arrow, which is believed to be porphyrin/C60 cocrystallites [33, 34]. Figure 5.9 (c) is an electron diffraction pattern of the ZnTPP:C60 BHJ layer, taken along the [123] direction of C60. A twin structure with a (112) twin plane is observed in Fig. 5.9(c), as indicated by a dotted line. Diffraction spots, which could correspond to cocrystallites of ZnTPP:C60, are also observed as indicated by arrows.
Since the microstructure of the ZnTPP and C60 BHJ layer is strongly dependent on their weight ratio, it is necessary to control the microstructures to form cocrystallites of ZnTPP:C60. In the present work, higher efficiencies were obtained for the ZnTPP:C60 sample with a weight ratio of 3 : 7, which would be suitable for the formation of cocrystallites, as observed for weak reflections in X-ray and electron diffraction patterns. Recombination of electrons of C60 and holes of ZnTPP could occur in the BHJ layer with intermittent cocrystallite structure. If continuous cocrystallite structures form perpendicular to the thin films, it is believed that the recombination of electrons and holes could be suppressed, which would lead to an improvement in conversion efficiency.
An energy level diagram, and the carrier transfer of electron transport layer (ETL) and hole transport layer (HTL) of a ZnTPP/C60/PTCDA solar cell are summarized as shown in Fig. 5.9(d) and (e), respectively. The incident direction of light is from the ITO side. An energy barrier could exist near the semiconductor/metal interface [42, 43]. Electronic charge-transfer separation was caused by light irradiation from the ITO substrate side. Electrons are transported to an Al electrode, and holes are transported to an ITO substrate. The VOC of organic solar cells is reported to be determined by the energy gap, as indicated by Eq. (5.2). The present experimental data of VOC indicated smaller values compared to the ones calculated from the equation, which might be due to the voltage descent at the metal/semiconductor interface. Control of the energy levels is also important to increase the efficiency.
5.7Diamond:C60
The fabrication and to characterization of C60/phthalocyanine based BHJ and HJ solar cells are presented here. C60 and fullerenol [C60(OH)10-12] were used for n-type semiconductors, and nanodiamoond (ND) and MPc derivatives were used for p-type semiconductors. A schematic diagram of the present C60/phthalocyanine based BHJ and HJ solar cells is shown in Fig. 5.10(a). A thin layer of PEDOT:PSS was spin coated on precleaned ITO glass plates. The PEDOT:PSS plays a role as an electron blocking layer for hole transport. Two types of solution for p-type semiconductors were produced [44, 45]. The first was produced by ND and tetra carboxy phthalocyaninate cobalt (Tc-CoPc) in deionized water. The solution for n-type semiconductors was prepared by dissolving C60 in 1,2-dichlorobenzene. On the thin layer of PEDOT:PSS, p-type semiconductor layers were prepared by spin coating a mixed solution of Tc-CoPc and ND in deionized water. The NDs were dispersed in the Tc-CoPc thin film. The n-type semiconductor layers were deposited on top of the p-type semiconductor layer by spin coating a C60 solution in 1,2-dichlorobenzene.
The second solution was also produced using tetra carboxy phthalocyaninate copper (Tc-CuPc), fullerenol [C60(OH)10-12] and ND in deionized water. On the thin layer of PEDOT:PSS, semiconductor layers were prepared by spin coating using a mixed solution of Tc-CuPc, C60(OH)10-12 and ND in deionized water. The NDs were obtained using the bead milling method in water and were dispersed in the active layer [46, 47].
The parameters measured for diamond based thin films indicated that the thin film structure with ND provided a higher cell performance on the JSC values than that of thin film structure without ND. Figure 5.10(b) and (c) shows the optical absorption spectra of the ND based thin films, and a solid line and a dashed line show thin film structure with ND and thin film structure without ND respectively. These thin films provided photo absorption in the range of 300–800 nm, and thin film structure with ND indicates a higher optical absorption compared with that of thin film structure without ND. The optical absorption properties of the thin film were improved by adding ND to the active layer.
Figure 5.10(d) and (e) shows the X-ray diffraction patterns of diamond powder and the thin films. In Fig. 5.10(d), the diffraction peaks of the diamond powder were confirmed as 111, 220 and 311 of the diamond structure. In Fig. 5.10(d) and (e), diffraction peaks corresponding to diamond are observed for the Tc-CoPc:ND/C60 and Tc-CuPc:ND:C60(OH)10-12 sample. The average particle sizes of the ND were calculated to be 4.5 and 5.5 nm from Scherrer’s formula.
Figure 5.11(a) is a TEM image of a C60 layer, and the lattice image of C60 {111} is observed. Figure 5.11(b) is an electron diffraction pattern of a C60 layer, and the diffraction peaks of C60 are observed. C60 also has an fcc structure with a lattice parameter of a = 1.42 nm. Figure 5.11(c) is an HREM image of the Tc-CoPc:ND composite layer. In Fig. 5.11(c), the lattice image of diamond {111} is observed. Tc-CoPc shows dark contrast in the image. Figure 5.11(d) is an electron diffraction pattern of the Tc-CoPc:ND composite layer, and diffraction peaks of diamond 111, 220, 311 are observed. Diamond powder has an fcc structure with a lattice parameter of a = 0.357 nm. Since no diffraction peak of Tc-CoPc was observed, Tc-CoPc could have an amorphous structure. Figure 5.11(e) is a TEM image of the Tc-CuPc:ND:C60 (OH)10-12 composite layer. The TEM image indicated NDof 4–6 nm as indicated by arrows, which agrees with the XRD results. Figure 5.11(f) is an electron diffraction pattern of the active layer, and the diffraction peaks of diamond 111, 220, 311 are observed. Since no diffraction peaks of Tc-CuPc and C60(OH)10-12 were observed, Tc-CuPc and C60(OH)10-12 could have amorphous structures. The increase in p/n HJ interface presents an advantage for the nanocomposite structure. However, due to the disarray of the donor/acceptor microstructure, electrons and holes could not transport smoothly through carrier recombination at the electronic acceptor/Al interface and at the PEDOT:PSS/electronic donor interface, respectively. To solve these problems, it will be necessary to introduce a layer preventing carrier recombination and improve the crystalline structure to have fewer defects. In the study presented in this chapter, ND based solar cells were fabricated and characterized. In the carbon-based solar cells presented earlier, thin films are fabricated by a chemical vapor deposition method [48, 49]. In the work presented here, solar cells with C60, C60(OH)10-12 and MPc as an organic semiconductor, and diamond particles and ND as an inorganic semiconductor were fabricated using a spin coating method, which is a low cost method.
The JSC values of the cells with ND increased compared to those without ND. In addition, the optical absorption spectra of the cells with ND were higher than those without ND in the range of 600–800 nm. The bandgap energy of diamond is typically ~ 5.5 eV and its carrier mobility is low. However, the ND could have a core shell structure as shown in Fig. 5.12(a), which indicates that the surface of the ND is covered by graphene sheets with sp2 hybridized orbitals, and there is an intermediate layer between the ND core and the graphene sheets. If the ND has such a three layered structure, the ND have various bandgap energies as shown in Fig. 5.12(b), and light with various wavelengths can be absorbed by the ND[12]. The XRD results also showed lattice distances of ~ 3.2A˚, which could be related to the intermediated layers. The energy level diagram of the cell with ND is shown in Fig. 5.12(c), and the carrier can be transported by hopping mechanism.
5.8Ge nanoparticles
The fabrication and characterization of fullerene based solar cells with Ge nanoparticles are presented here. Copper tetrakis (4-cumylphenoxy) phthalocyanine (Tc-CuPc) was used for p-type semiconductors as shown in Fig. 5.13(a), and C60 was used for n-type semiconductors. In addition, Ge(IV) bromide (GeBr4) was added to the solar cells in order to form Ge based quantum dots to increase their photovoltaic efficiencies [50]. Device structures were produced, and efficiencies, optical absorption and nanostructures were investigated.
A thin layer of PEDOT:PSS (Sigma Aldrich Corp.) was spin coated onto precleaned ITO glass plates. Then, semiconductor layers were prepared on a PEDOT layer by spin coating using a mixed solution of C60, Tc-CuPc and GeBr4 in 1 mL o-dichlorobenzene. The weight ratio of Tc-CuPc:C60 was 1 : 8 (2mg : 16mg), and 0.03 mL of GeBr4 was added into the solution [50]. The thickness of the blended device was ~ 150 nm. A schematic diagram of the Tc-CuPc:C60 BHJ and HJ solar cells with a Tc-CuPc/C60 structure is shown in Fig. 5.13(b). To increase efficiencies, GeBr4 was also added in the Tc-CuPclayers for both structures. After annealing at 100 °C for 30 min in N2 atmosphere, Al metal contacts with a thickness of 100 nm were evaporated as a top electrode.
The BHJ indicates one layered composite structures with p and n type semiconductors, which is denoted as Tc- CuPc:C60. The common HJ solar cell that has separated two layers was also investigated for comparison, which is denoted as Tc-CuPc/C60. The open circuit voltages measured for Tc-CuPc:C60 and Tc-CuPc/C60 were increased by several times through the addition of GeBr4, and slight increases were also observed in the short circuit current density of both structures. Figure 5.13(b) shows the optical absorption of Tc-CuPc:Ge:C60 and Tc-CuPc:C60 BHJ solar cells. The Tc-CuPc:Ge:C60 structure provided photo-absorption in the range of 500 to 1200 nm (which corresponds to 2.5 and 1.0 eV. respectively), which was higher than that of the Tc-CuPc:C60 structure. The energy gap between HOMO and LUMO for C60 is 1.7 eV, which corresponds to an absorbance of 730 nm [51].
A TEM image of the Tc-CuPc:Ge:C60 BHJ layer is shown in Fig. 5.13(c). Nanoparticles containing Ge (the element with the largest atomic number in this cell) are observed in the Tc-CuPc layer. An enlarged TEM image is shown in Fig. 5.13(d), and the lattice fringes of Tc-CuPc are observed. The nanoparticle with Ge compounds is denoted as Ge comp. The electron diffraction pattern of the Tc-CuPc:Ge:C60 BHJ layer is shown in Fig. 5.13(e), and many diffraction spots and rings corresponding to C60 111, 220 and 311 were observed, which indicates microcrystalline structures in C60. The dispersion of Ge based nanoparticles is effective for optical absorption in the range of 500 to 1200 nm. Although Ge has a band gap energy of 0.7 eV, optical absorption was observed in the range of 2.5 and 1.0 eV in the present work, which could be due to Ge compound formation and the nanodispersion effect of the nanoparticles, which I have reported on previously [52]. The interpenetrating DA network has a large interfacial area, which could be effective for charge generation. Since the microstructures of Tc-CuPc and C60 were disordered, recombination of electrons of C60 and holes of Tc-CuPc could occur. Therefore, the ordered column-like structure would be suitable for carrier transport. If continuous nanocomposite structures are perpendicular to the thin films, it is believed that the recombination of electrons and holes could be avoided, and the conversion efficiency of the solar cells would increase.
The quantum dot solar cell including intermediate band structures is an important candidates for future high efficiency solar cells [53–58]. In the present work, the efficiencies of the solar cells were increased through the formation of Ge based nanoparticles. The technique presented here as a solution is a very simple and cost effective method for the formation of nanoparticles. To improve the efficiencies, the arrangement of the quantum dots and the control of size distribution are necessary. Combination of solar cells presented here and copper oxide nanomaterials with various direct band gaps or other organic materials might also be effective increasing efficiency. [59, 60] The performances of the solar cells presented here could also be due to the nanoscale structures.
5.9Dye-sensitized solar cells
Although silicon solar cells have high conversion efficiencies and a long lifetime, their production processes are complicated and expensive. Dye-sensitized solar cells, based on the concept of photo-sensitization of wide band-gap mesoporous oxide semiconductors [61], are now in a state of advanced development. This technology has been established as a promising low-cost method [62], and the cells are lightweight and can be colorized. However, dye-sensitized solar cells have a short lifetime due to the leakage and vaporization of electrolytes. Therefore, studies of solidification of dyesensitized solar cells have been performed [63, 64], and organic dyes without noble metals are expected to have potential as low-cost dyes.
An investigation into the electrical and optical properties of dye-sensitized solar cells (DSSC) with an amorphous TiO2 layer to introduce electrons at trap levels in acceptor and donor levels is presented here. The effects caused by the addition of organic dyes such as protoporphyrin IX (PPIX), xylenol orange (XO) and rose Bengal (RB) to dye-sensitized solar cells were investigated through the fabrication and characterization of such cells. The schematic illustrations of the solar cells presented here are shown in Fig. 5.14(a) [65].
Nanocrystalline TiO2 photoelectrodes were prepared, beginning by dispersing TiO2 powder in an aqueous solution (1 mL) in a mixture of acetylacetone (0.02 mL) with Triton X−100 (0.01 mL) with polyethylene glycol (0.2 g) [66, 67]. The TiO2 paste was then coated on pre-cleaned FTO using the squeegee method. After the TiO2 coating, the FTO substrate was sintered for 30 min at 450 °C, and the prepared titanium isopropoxide (TTIP) solution was dropped onto the substrate. After that, the substrate was sintered for 60 min at 180 °C. The TTIP solution was prepared by mixing TTIP (0.3mL), acetilaceton (0.08 mL), ethanol (0.64 mL) and PEG# 20 000 (0.2 g). The TiO2 electrodes were dissolved into the solved organic dyes, which included protoporphyrin, xylenol orange and rose Bengal solutions in distilled water, methanol or ethanol [65, 68]. Bellfine (0.1 g) and Denka black (0.02 g) as carbon was dispersed in the distilled water (0.8 mL) and ethanol (0.4mL) with sodium carboxymethyl cellulose (0.012 g). The carbon paste was applied to the indium tin oxide (ITO) as an opposite electrode using the squeegee method. A heat treatment of the carbon on the ITO substrate was carried out at 180 °C for 30 min. The electrolyte fabricated in a mixture of iodine (0.05 g), lithium iodide (0.09 g), ethylene carbonate (0.41 g), propylene carbonate (0.5 mL) and polyacrylonitrile (0.17 g) was agitated and heated at 80 °C. The dye-sensitized solar cells were assembled by putting the electrolyte between the adsorbed dye materials at the TiO2 layer on the FTO glass substrate and the carbon film on the ITO substrate.
The measured J−V curves of DSSC with or without an amorphous TiO2 layer are shown in Fig. 5.14(b). It can be seen that the current density of DSSC with an amorphous TiO2 layer was higher than that of DSSC without an amorphous TiO2 layer. The current density improved from 0.60 to 0.83 mA cm−2. The other parameters remained almost the same, and the conversion efficiency improved from 0.12% to 0.16%.
Figure 5.15(a) shows XRD patterns of TiO2 layers prepared from the TTIP solution as a function of temperature. No peak of an anatase phase is observed when annealing at 250°C. The small peaks of the anatase phase are observed after annealing at 350°C. The diffraction peaks of TiO2 (P25) are shown for comparison with the amorphous TiO2, which indicates the existence of rutile and anatase structures of TiO2, as shown in Fig. 5.15(b) and (c), respectively. The optical absorption of DSSC with an amorphous TiO2 layer was compared to that without an amorphous TiO2 layer, indicating that the optical absorption peak for TiO2 of around 360 nm was increased through the introduction of an amorphous TiO2 layer.
A TEM image and an electron diffraction pattern of the DSSC with PPIX are shown in Figs. 5.15(d) and (e), respectively. The TEM image indicates the presence of TiO2 nanoparticles with sizes of 20 ~ 60 nm, and the electron diffraction pattern shows 101, 103 and 200 reflections of a tetragonal TiO2 anatase phase. A high-resolution electron microscopy (HREM) image of an interface of TiO2 nanoparticles is shown in Fig. 5.15(f). The interface is directly connected at the atomic scale, which could results in good carrier transport between TiO2 nanoparticles. A HREM image of surface of the TiO2 nanoparticle is shown in Fig. 5.13(g), and the TiO2 nanoparticle is covered by a 2 nm thick PPIX layer with an amorphous structure. The TEM image and an ED pattern also showed the amorphous TiO2 layer [65]. The amorphous TiO2 layer around the TiO2 nanoparticle was observed by TEM, and the ED pattern indicated 101 and 103 reflections of the polycrystalline coagulation of TiO2 with a tetragonal anatase structure. The diffuse ring of an amorphous phase was also observed, which indicates that the structure of the TiO2 layer is a mixture of anatase and amorphous phases.
A schematic diagram of a TiO2 electrode with amorphous TiO2 layer is shown in Fig. 5.16(a). An energy level diagram of DSSC with an amorphous TiO2 layer is shown in Fig. 5.16(b). The values of HOMO and LUMO were calculated by B3LYP/6−31G*. An energy barrier could exist at the semiconductor metal interface. The generation of electronic charge is caused by light irradiation from the FTO substrate side. The TiO2 layer or amorphous TiO2 layer receives the electrons from the dye, and the electrons are trapped by several trap levels of the amorphous TiO2. The electrons are transported to an FTO electrode through the TiO2 layer, and electrons are transferred to the outside circuit, and flow through the carbon electrode. Then, electrons return to the electrolyte through an oxidation-reduction reaction. The improvements seen in the solar cells presented here resulted from the introduction of the amorphous TiO2 electrode, and this structure should be investigated further. Carrier recombination could be a main cause for the low conversion efficiency. It is expected that the conversion efficiency of the cells presented here can be improved through charge separation and electronic charge transfer.
Improvements in conversion efficiency have been reported as an effect of coating materials on TiO2. Insulator oxides such as Al2O3 or SiO2 on TiO2 have been introduced for DSSCs [69]. These oxides suppress carrier recombination by reverse transfer of electrons in the cells. In the present study, DSSCs with amorphous TiO2 layers on TiO2 were fabricated, and the amorphous TiO2 could attract electrons because of the many trap levels in the acceptor and donor levels. In addition, the carrier separation effectiveness is high as a result of the large contact area. Further improvement would be possible by introducing nanoparticles as light-harvesting materials.
In summary, amorphous TiO2 was introduced into DSSC with mixed xylenol orange and rose Bengal. The optical absorption peak of TiO2, around 360 nm, was increased through the introduction of an amorphous TiO2 layer. Diffraction spots of the anatase phase and a diffuse ring from the amorphous phase were observed. The energy diagram shows that the amorphous TiO2 can attract electrons to several trap levels in the TiO2 layers. In addition, the carrier separation is high, which could be due to the large contact area. As a result, the current density and the conversion efficiency were improved.
5.10Polysilane system
Polysilane is a p-type semiconductor, and has been used as an electrically conductive material and in photovoltaic systems [70–76]. Polysilanes are known as σ-conjugate polymers, and their hole mobility is 10−4 cm2 V−1 s−1 [70]. Although polysilanes could be applied to p-type semiconductors on organic thin-filmsolar cells, few studies have previously been carried out on polysilane solar cells [72, 77–79].
The purpose of the study presented here is twofold. First, the aim was to fabricate and characterize bulk heterojunction solar cells with polysilane and fullerenes of C60 and [6,6]-phenyl-C61 -butyric acid methyl ester (PCBM). C60 was selected as a good electronic acceptor material for devices. The second purpose of the study way to fabricate heterojunction solar cells by using a mixture solution of polysilane doped with phosphorus and boron, and to investigate the effects of annealing temperature and doping of boron (B) and phosphorous (P) on their electronic properties and microstructures. It is expected that amorphous silicon doped with boron would function as a p-type semiconductor, and amorphous silicon doped with phosphorus would function as an n-type semiconductor. Spin-coating is a low-cost method, and is essential for the mass production of solar cells. Four types of polysilane were used in the present work: dimethyl-polysilane (DMPS), poly(methyl phenyl silane) (PMPS), poly(phenyl silane) (PPSi) and decaphenyl cyclopentasilane (DPPS). Figure 5.17 shows the solar cell structures and molecular structures of the DMPS, PMPS, PPSi, and DPPS used to fabricate bulk-heterojunction and heterojunction solar cells [79]. Indium tin oxide (ITO) grass plates were cleaned in an ultrasonic bath with acetone and methanol, and then were dried with nitrogen gas. A thin layer of polyethylendioxythiophen doped with poly(3,4-ethylene dioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) was spin-coated onto the ITO substrates. Then, semiconductor layers were prepared on a PEDOT:PSS layer by spin-coating using mixture solutions of DMPS, PMPS, PPSi or DPPS, and C60 or PCBM in 1 mL o-dichlorobenzene. The total weight of C60:PMPS, C60:PPSi, C60:DPPS, or PCBM:DPPSwas 10 mg, the weight ratio of C60:PMPS, C60:PPSi, or C60:DPPS was 8 : 2 and the weight ratio of PCBM:DPPSwas 7 : 3. The total weight of C60:DMPSwas 19 mg, and the weight ratio of C60:DMPSwas 16 : 3. Another type of thin-film solar cells was prepared as described below. Semiconductor layers were prepared on the PEDOT:PSS layer by spin-coating using a mixture solution of DPPS (7 mg) and diethyl-methoxyborane (0.2 mL) in 1mL o-dichlorobenzene. After annealing at 300 °C for 10 min in N2 atmosphere, a subsequent layer was prepared by spin-coating using the mixture solution of DPPS and phosphoric acid (0.2 mL) or phosphorus bromine (0.02mL) in 1 mL of o-dichlorobenzene, and then the samples were annealed at 300 °C for 10 min in N2 atmosphere. Aluminum(Al) metal contacts were evaporated as a top electrode. Finally, the devices were annealed at 140 °C for 30 min in N2 atmosphere.
The J–V characteristics of the solar cells showed open-circuit voltage and short-circuit current. The measured parameters of the present solar cells are summarized in Table 5.2. A solar cell with the DMPS/C60 structure provided a power conversion efficiency of 0.020 %, which was better than in other polysilane-based devices containing C60. A solar cell with the DPPS/PCBM structure also provided a conversion efficiency of 0.032%. The DPPS(B)/DPPS(P) solar cells prepared at 300 °C provided a conversion efficiency of 0.028 %and the highest open-circuit voltage of 0.81 V.
Table 5.2: Measured parameters of the present solar cells.
XRD patterns for DMPS:C60 and DPPS:PCBM thin films are shown in Figs. 5.18(a) and (b), respectively. The diffraction patterns showed several diffraction peaks, which corresponded to DMPS, C60, DPPS, and PCBM, as indicated in the figures. The XRD patterns of the annealed DPPS thin films are also shown in Fig. 5.18(c). Some diffraction peaks were observed for DPPS after annealing at 250 °C, which indicates the presence of microcrystalline structures. After annealing at 300 °C, most of the diffraction peaks decreased and disappeared for DPPS(B) and DPPS(P) thin films, which indicates the formation of an amorphous structure from the microcrystalline structure resulting from the addition of doping elements. The addition of phosphoric acid and diethylmethoxyborane to DPPS could enhance the decomposition of the DPPS to form an amorphous structure.
Figure 5.19(a)–(f) show TEM images and electron diffraction patterns for DMPS: C60, DPPS:PCBM, and DPPS thin films. A TEM image and electron diffraction pattern of the DMPS:C60 composite film are shown in Fig. 5.19(a) and (d), respectively. In Fig. 5.19(a), nanoparticles consisting of the Si element in the C60 matrix were observed. The electron diffraction pattern in Fig. 5.19(d) shows many diffraction spots and Debye-Scherrer rings, which indicates the microcrystalline structures of C60 and DMPS.
Figure 5.19(b) and (e) show a HREM image and an electron diffraction pattern of DPPS:PCBMbulk heterojunction thin films, respectively. In Fig. 5.19(e), the diffraction spots corresponding to DPPS and PCBM were observed. The HREM image in Fig. 5.19(e) indicates that the DPPS:PCBM thin films have a nanocomposite structure, which has a lamella structure with a periodicity of ~ 3 nm. The optimization of the nanocomposite structure of DPPS:PCBM would increase the conversion efficiency of the solar cells.
Figure 5.19(c) shows a HREM image of DPPS annealed at 300 °C, and Fig. 5.19(f) shows an electron diffraction pattern of the DPPS. Contrast in the image corresponding to an amorphous structure is observed in Fig. 5.19(c), and (f) shows a halo-likeintensity signal, which indicates that the DPPS thin films that annealed at 300 °C had an amorphous structure.
The Raman scattering spectra of PMPS were calculated to identify active modes, as shown in Fig. 5.18(d). The Raman scattering spectra of PMPS and doped PMPS thin films were obtained, as also shown in Fig. 5.18(d). The Raman active modes of the film were obtained: 668, 1012, 1644, 2228, 3044, 3116, and 3196 cm−1. The peaks at 668, 1012, 1644, and 2228 cm−1 were identified as the phenyl group vibration modes in PMPS. The strong peaks at 3044, 3116, and 3196 cm−1 were identified as the methyl group vibration modes in PMPS. The Raman peaks observed in non-doped PMPS conformed to the vibration mode of the carbon bonds of the phenyl and methyl groups in PMPS, as shown in Fig. 5.18(d). When phosphorus was doped in the PMPS thin film that was then annealed at 300 °C, the peaks at 2228, 3044, 3116, and 3196 cm−1 disappeared.
The effect of doping phosphorus on polysilane was investigated by Hall effect measurements. The resistivity, carrier concentration, and mobility of PMPS thin films doped with phosphorus are listed in Table 5.3. Although non-doped PMPS films showed a high electrical resistivity (~ 108 Ω cm), the P-doped PMPS thin films showed n-type semiconducting behavior, low resistivity, and high electron mobility. After annealing at 500 °C, electronic properties deteriorated. The phosphorus doping to PMPS provided good transport phenomena after annealing at 300 °C.
Table 5.3: Resistivity, carrier concentration and mobility of PMPS(P) thin films.
The energy level diagrams of the polysilane-based solar cells discussed here are summarized in Fig. 5.20. In this study, the VOC of DPPS was higher than that of PMPS, which could indicate that the HOMO–LUMO levels are different, as indicated in Fig. 5.20. Mixing C60 with polysilanes would suppress the recombination of photo-separated charge carriers by promoting electron transfer from polysilanes. Amorphous silicon doped with boron or phosphorus could function as a p- or n-type semiconductor, respectively, and energy levels would be different between the DPPS with B and DPPS with P.
The DPPS-based solar cells presented here were compared with other silicon-based solar cells such as amorphous silicon solar cells, prepared by inductively coupled plasma chemical vapor deposition (ICP-CVD), and a spin-coating method, as listed in Table 5.4. Solar cells fabricated using a spin-coating method had a simpler fabrication process and better cost performance. The low conversion efficiency of the solar cells presented here could be due to the high electrical resistance and carrier recombination caused by defects, and further improvement of these solar cells will be necessary.
Table 5.4: Comparison of silicon-based solar cells.
In summary, polysilane-based solar cells were fabricated using a mixture solution of DMPS, PMPS, PMSi, DPPS, C60, PCBM, phosphorus, and boron, and characterized using electrical measurements, Raman scattering and microstructural analyses. Bulk heterojunction devices of DMPS:C60 and DPPS:PCBMand heterojunction devices of DPPS(B)/ DPPS(P) exhibited photovoltaic properties. XRD and TEM results indicated that the DMPS:C60 and DPPS:PCBM layers had nanocomposite structures, and that doped DPPS thin films formed an amorphous structure following their annealing at 300 °C. Raman scattering results indicated a desorption of the methyl and phenyl groups owing to phosphorus doping. Energy level diagrams, and carrier transport mechanisms were discussed.
5.11PCBM:P3HT with SiPc or SiNc
Metal phthalocyanines (MPc) and metal naphthalocyanines (MNc) are groups of small molecular materials with Q-band absorption in the red to near-infrared range, and they have high optical and chemical stabilities, and photovoltaic properties. Therefore, they are used as donor materials in organic solar cells. Heterojunction solar cells using copper phthalocyanine and fullerene have been fabricated by an evaporation method, and their power conversion efficiency was ~ 3% [81]. The characteristics of such solar cells with various MPc and MNc, including electronic conductivity, crystalline structure, and absorption range, were investigated by several research groups [82–85]. Organic solar cells such as those using P3HT and PCBM exhibit good incident photon-to-current conversion efficiency and fill factor. The device performance of such polymer solar cells can be affected by the preparation conditions such as the annealing temperature, the concentration of the starting material, and film thickness. The addition of third components such as phthalocyanines, naphthalocyanines, and low-band-gap polymers is expected to help absorb light that the P3HT and PCBM cannot collect. In particular, phthalocyanines absorb near-infrared light, and the effects of adding silicon phthalocyanine, silicon naphthalocyanine, or germanium phthalocyanine to the P3HT:PCBMsystemhave been investigated in previous studies [86–90]. MPc and MNc can be dissolved in organic solvents, and application to the device process using a spin-coating method is possible with solubilization.
The fabrication and characterization of bulk heterojunction polymer solar cells with an inverted structureusing PCBM, P3HT, soluble tetrakis(tert-butyl)[bis(trihexylsiloxy)sillicon phthalocyanine] (SiPc) and tetrakis(tert-butyl)[bis(trihexylsiloxy)sillicon naphthalocyanine] (SiNc) were presented here. SiPc or SiNc was added as the third component for PCBM:P3HT solar cells, as shown in Fig. 5.21(a) [91, 92]. The direction of electron flow in the inverted device structure is opposite to that in conventional devices [93–95], and the inverted structure also provided stable device performances in air. Layered structures of bulk heterojunction solar cells with inverted structures were denoted as ITO/TiOx/PCBM:P3HT(SiPc or SiNc)/PEDOT:PSS/Au, as shown schematically in Fig. 5.21.
The device performance of the solar cell doped with SiPc was better than that of the solar cell with PCBM:P3HT structure. A solar cell with a PCBM:P3HT(SiPc, 3wt.%) structure provided η of 0.768%, which is the best η value of any of the devices presented in this work [91]. The maximumη occured for SiPc concentrations between 3 to 7 wt.%. The stability of inverted-structure solar cells with a PCBM:P3HT(SiPc) active layer was also investigated. J–V characteristics of the device were measured after 1 week of exposure to ambient atmosphere, as listed in Table 5.5, and the devices exhibited stability in air. The conversion efficiency slightly increased upon exposure to air for 1 week, and similar increases in efficiencies have been observed in previous works [96, 97]. The morphology of the solar cells after one week would be improved in comparison to the as-prepared device, leading to improved carrier mobility and a decreased energy barrier of the metal/semiconductor interface. Devices maintained similar conversion efficiencies after 2 weeks of exposure to air.
Table 5.5: Measured parameters of PCBM:P3HT(SiPc, 1wt.%) solar cell.
The EQE spectra of PCBM:P3HT(SiPc) solar cells are shown in Fig. 5.21(b). EQE peaks were observed at ~ 680 nm for the solar cell with SiPc. If SiPc aggregates exist in the P3HT domain, no charge separation occurs, and if the SiPc exists in the PCBM domain, charge transfer does not occur. Therefore, the power conversion efficiency would be improved only when SiPc exists at the PCBM:P3HT interface. Since the peaks of EQE for SiPc were observed at ~ 680 nm, SiPc could exist at the PCBM:P3HT interface. The internal quantum efficiencies of PCBM:P3HT(SiPc) solar cells were calculated from the EQE, as shown in Fig. 5.21(c). A relatively higher IQE is observed for the PCBM:P3HT(SiPc) solar cells in the range of 640–700 nm. EQE spectra of PCBM:P3HT(SiNc) solar cells are also shown in Fig. 5.21(d), which indicates an EQE peak ~ 800 nm. The IQE spectra of the PCBM:P3HT(SiNc) solar cell is shown in Fig. 5.21(e), which also indicates higher IQE in the range of 700-900 nm, which is an effective range for photovoltaic devices.
The XRD patterns of PCBM, P3HT, SiPc, PCBM:P3HT, and PCBM:P3HT(SiPc) thin films are shown in Fig. 5.22(a). Diffraction peaks are observed for P3HT and SiPc, which indicate that they have crystalline structures. No sharp diffraction peak is observed for PCBM, which indicates that the PCBM has an amorphous structure. For the PCBM:P3HT(SiPc) and PCBM:P3HT thin films, diffraction peaks due to P3HT are weaker and broader than that of the single P3HT phase, and no sharp peak due to SiPc is observed. This indicates that P3HT has nanocrystalline structures dispersed in the amorphous PCBM, and that SiPc molecules without a crystalline structure could exist at the PCBM/P3HT interface. No sharp diffraction peak is observed for SiNc, which also indicates the PCBM:P3HT(SiNc) thin films have a similar structure.
A TEMimage and an electron diffraction pattern of the PCBM:P3HT(SiPc) thin film are shown in Figs. 5.22(b) and (c), respectively [91]. In Fig. 5.22(b), P3HT nanowires are observed in the amorphous PCBM matrix. The electron diffraction pattern in Fig. 5.22(c) shows a halo-like intensity, which indicates the presence of amorphous and nanocrystalline structures, which agrees well with the XRD results shown in Fig. 6.15(a).
An interfacial structural model and an energy level diagram of the PCBM: P3HT(SiPc) solar cells are summarized and illustrated in Figs. 5.22(d) and 5.23(a), respectively. In Fig. 5.22(d), P3HT nanowires are dispersed in the PCBM amorphous matrix, which was confirmed by XRD and TEM analyses. Since an increase in photocurrent originating from the SiPc was observed, it is believed that the SiPc moleculeswere located at the PCBM/P3HT interface. The localization of SiPc molecules at the PCBM/P3HT interface could be caused by the insolubility of SiPc both in P3HT nanowires and the PCBM matrix. When the amount of SiPc added increased, the efficiencies decreased, which indicates that the single-layer localization of SiPc at the PCBM/P3HT interface could be effective for the conversion efficiencies. There could be two mechanisms that cause an increase in efficiency upon the addition of SiPc. Additional absorption of SiPc directly contributes to the increase in JSC, which was observed in EQE and IQE spectra at 680 nm in Figs. 5.21(b) and (c), respectively. In addition, EQE and IQE intensities in the range of 400–600 nm were increased by the addition of SiPc, which could be explained by the Förster energy transfer from P3HTto SiPc [98, 99], as shown in Fig. 5.23(b). In the cell with an inverted structure, electrons are transported to an ITO substrate, and holes are transported to the Au electrode. Electrons could be transported only when the value of the LUMO for SiPc (X eV) is −4.2 ≤ X ≤ −3.2 eV, and holes could be transported only when the value of the HOMO for SiPc (Y eV) is Y ≤ −5.2 eV. Control of HOMO and LUMO of the phthalocyanine is required for polymer solar cells. Since a molecule exists in monomeric molecule, a bulky molecular structure is required, and the present SiPc has a suitable structure. The addition of silicon naphthalocyanine to PCBM:P3HT bulk heterojunction solarcells also resulted in a higher IQE, as observed in Fig. 5.21(e), which is due to the different compositions of PCBM:P3HT. [92]
In summary, PCBM:P3HT(SiPc) and PCBM:P3HT(SiNc) bulk heterojunction solar cells with inverted structures were fabricated and characterized. The photovoltaic properties of the solar cells were improved by the addition of SiPc, and they were almost stable after two weeks of exposure in air. Microstructural analysis showed that P3HT nanowires are dispersed in the amorphous PCBM matrix, and it is believed that the SiPc or SiNc molecules are located at the PCBM/P3HT interface. The EQE and IQE spectra of the PCBM:P3HT(SiPc) solar cell showed peaks at ~ 680 nm, and the intensities of EQE and IQE in the range of 400–600 nm were increased by the addition of SiPc. The increase in conversion efficiency caused by the addition of SiPc could be explained in terms of the direct charge transfer from SiPc to PCBM and Förster energy transfer from P3HT to SiPc. PCBM:P3HT(SiNc) solar cells also showed higher values at the near-infrared region (~ 800 nm), which indicated that solar cell performance can be enhanced by the addition of SiNc.
5.12Bibliography
[1]Oku T, Narita I, Nishiwaki A, Koi N, Suganuma K, Hatakeyama R, Hirata T, Tokoro H, Fujii S. Formation, atomic structures and properties of carbon nanocage materials. Topics Appl Phys. 2006; 100: 187–216.
[2]Sariciftci NS, Smilowitz L, Heeger AJ, Wudl F. Photoinduced electron transfer from a conducting polymer to buckminsterfullerene. Science. 1992; 258: 1474–1476.
[3]Hayashi Y, Yamada I, Takagi S, Takasu A, Soga T, Jimbo T. Influence of structure and C60 composition on properties of blends and bilayers of organic donor-acceptor polymer/C60 photovoltaic devices. Jpn J Appl Phys. 2005; 44: 1296–1300.
[4]Oku T, Nagaoka S, Suzuki A, Kikuchi K, Hayashi Y, Inukai H, Sakuragi H, Soga T. Formation and characterization of polymer/fullerene bulk heterojunction solar cells. J Phys Chem Solids. 2008; 69: 1276–1279.
[5]Oku T, Noma T, Suzuki A, Kikuchi K, Kikuchi S. The effects of exciton-diffusion blocking layers on pentacene/C60 bulk heterojunction solar cells. J Phys Chem Solids. 2010; 71: 551–555.
[6]Lin YH, Yang PC, Huang JS, Huang GD, Wang IJ, Wu WH, Lin MY, Su WF, Lin CF. High-efficiency inverted polymer solar cells with solution-processed metal oxides. Sol Energy Mater Sol Cells. 2011; 95: 2511–2515.
[7]Ma W, Yang C, Gong X, Lee K, Heeger AJ. Thermally Stable, Efficient Polymer Solar Cells with Nanoscale Control of the Interpenetrating Network Morphology. Adv Funct Mater. 2005; 15: 1617–1622.
[8]Nikkei Microdevices and Nikkei Electronics Edit. Solar cells 2008/2009 (in Japanese). Nikkei BP. 2008.
[9]Oku T, Takeda A, Nagata A, Noma T, Suzuki A, Kikuchi K. Fabrication and characterization of fullerene-based bulk heterojunction solar cells with porphyrin, CuInS2, diamond and exciton-diffusion blocking layer. Energies. 2010; 3: 671–685.
[10]Manceau M, Angmo D, Jørgensen M, Krebs FC. ITO-free flexible polymer solar cells: From small model devices to roll-to-roll processed large modules. Org Electron. 2011; 12: 566–574.
[11]Dou L, You J, Yang J, Chen CC, He Y, Murase S, Moriarty T, Emery K, Li G, Yang Y. Tandem polymer solar cells featuring a spectrally matched low-bandgap polymer. Nat Photonics. 2012; 6: 180–185.
[12]OkuT, Takeda A, Nagata A, Kidowaki H, Kumada K, Fujimoto K, Suzuki A, Akiyama T, Yamasaki Y, Ōsawa E. Microstructures and photovoltaic properties of C60 based solar cells with copper oxides, CuInS2 , phthalocyanines, porphyrin, PVK, nanodiamond, germanium and exciton diffusion blocking layers. Mater Technol. 2013; 28: 21–39.
[13]Yu G, Heeger AJ. Charge separation and photovoltaic conversion in polymer composites with internal donor/acceptor heterojunctions. J Appl Phys. 1995; 78: 4510–4515.
[14]Padinger F, Rittberger RS, Sariciftci NS. Effects of Postproduction Treatment on Plastic Solar Cells. Adv Funct Mater. 2003; 13: 85–88.
[15]Oku T, Kakuta N, Kawashima A, Nomura K, Motoyoshi R, Suzuki A, Kikuchi K, Kinoshita G. Formation and characterization of bulk hetero-junction solar cells using C60 and perylene. Mater Trans. 2008; 49: 2457–2460.
[16]Nomura K, Oku T, Suzuki A, Kikuchi K, Kinoshita G. The effects of exciton-diffusion blocking layers on pentacene/C60 bulk heterojunction solar cells. J Phys Chem Solids. 2010; 71: 210–213.
[17]Liang Y, Xu Z, Xia J, Tsai ST, Wu Y, Li G, Ray C, Yu L. For the Bright Future—Bulk Heterojunction Polymer Solar Cells with Power Conversion Efficiency of 7.4%. Adv Mater. 2010; 22: E135–138.
[18]Yang L, Zhou H, Price SC, You W. Parallel-like Bulk Heterojunction Polymer Solar Cells. J Am Chem Soc. 2012; 134: 5432–5435.
[19]Shirahata Y, Tanaike K, Akiyama T, Fujimoto K, Suzuki A, Balachandran J, Oku T. Fabrication and photovoltaic properties of ZnO nanorods/perovskite solar cells. AIP Conf Proc. 2016; 1709: 020018-1-9.
[20]Koster LJA, Mihailetchi VD, Blom PWM. Ultimate efficiency of polymer/fullerene bulk hetero-junction solar cells. Appl Phys Lett. 2006; 88: 093511–093513.
[21]Scharber MC, M¨uhlbacher D, Koppe M, Denk P, Waldauf C, Heeger AJ, Brabec CJ. Design rules for donors in bulk-heterojunction solar cells – towards 10% energy-conversion efficiency. Adv Mater. 2006; 18: 789–794.
[22]Brédas JL, Beljonne D, Coropceanu V. Charge-transfer and energy-transfer processes in piconjugated oligomers and polymers: a molecular picture. Chem Rev. 2004; 104: 4971–5004.
[23]Ulbricht R, Lee SB, Jiang X, Inoue K, Zhang M, Fang S, Baughman RH, Zakhid AA. Transparent carbon nanotube sheets as 3-D charge collectors in organic solar cells. Sol Energy Mater Sol Cells. 2007; 91: 416–419.
[24]Oku T, Narita I, Koi N, Nishiwaki A, Suganuma K, Inoue M, Hiraga K, Matsuda T, Hirabayashi M, Tokoro H, Fujii S, Gonda M, Nishijima M, Hirai T, Belosludov RV, Kawazoe Y. Boron nitride nanocage clusters, nanotubes, nanohorns, nanoparticles, and nanocapsules. In: Yap YK ed. B-C-N nanotubes and related nanostructures. Springer. 2009: 149–194.
[25]Yook KS, Chin BD, Lee JY, Lassiter BE, Forrest SR. Vertical orientation of copper phthalocyanine in organic solar cells using a small molecular weight organic templating layer. Appl Phys Lett. 2011; 99: 043308.
[26]Mori S, Nagata M, Nakahata Y, Yasuta K, Goto R, Kimura M, Taya M. Enhancement of incident photon-to-current conversion efficiency for phthalocyanine-sensitized solar cells by 3D molecular structuralization. J Am Chem Soc. 2010; 132: 4054–4055.
[27]Yamasaki Y, Takaki K. Synthesis of μ-oxo-bridged hetero-metal phthalocyanine dimer analogues and application for charge generating material in photoreceptor. Dyes Pigm. 2006; 70: 105–109.
[28]Yamasaki Y, Kuroda K, Yakaki K. Synthesis of new polymorphs of μ-oxo-meta(III) phthalocyanine dimers and their photoconductive properties. J Chem Soc Jpn Chem Indust Chem. 1997; 12: 887–898.
[29]Takeda A, Minowa A, Oku T, Suzuki A, Kikuchi K, Yamasaki Y. Formation and characterization of phthalocyanine dimer/C60 solar cells. Prog Nat Sci. 2011; 21: 27–30.
[30]Takeda A, Oku T, Suzuki A, Yamasaki Y. Theoretical study of gallium phthalocyanine dimerfullerene complex for photovoltaic device. J Mod Phys. 2011; 2: 966–969.
[31]Mizuseki H, Igarashi N, Belosludov RV, Farajian AA, Kawazoe Y. Theoretical study of phthalocyanine–fullerene complex for a high efficiency photovoltaic device using ab initio electronic structure calculation. Synth Met. 2003; 138: 281–283.
[32]Hameed AJ. Theoretical investigation of a phthalocyanine–fulleropyrrolidine adduct and some of its metallic complexes. J Mol Struct Theochem. 2006; 764: 195–199.
[33]Takahashi K, Kuraya N, Yamaguchi T, Komura T, Murata K. Three-layer organic solar cell with high-power conversion efficiency of 3.5%. Sol Energy Mater Sol Cells. 2000; 61: 403–416.
[34]Hasobe T, Imahori H, Kamat PV, Ahn TK, Kim SK, Kim D, Fujimoto A, Hirakawa T, Fukuzumi S. Photovoltaic cells using composite nanoclusters of porphyrins and fullerenes with gold nanoparticles. J Am Chem Soc. 2005; 127: 1216–1228.
[35]Sun Q, Dai L, Zhou X, Li L, Li Q. Bilayer- and bulk-heterojunction solar cells using liquid crystalline porphyrins as donors by solution processing. Appl Phys Lett. 2007; 91: 253505.
[36]Hasobe T, Sandanayaka ASD, Wada T, Araki Y. Fullerene-encapsulated porphyrin hexagonal nanorods. An anisotropic donor–acceptor composite for efficient photoinduced electron transfer and light energy conversion. Chem Commun. 2008: 3372–3374.
[37]Ishii T, Aizawa N, Kanehama R, Yamashita M, Sugiura K, Miyasaka H. Cocrystallites consisting of metal macrocycles with fullerenes. Coord Chem Rev. 2002; 226: 113–124.
[38]Konarev DV, Kovalevsky AY, Li X, Neretin IS, Litvinov AL, Drichko NV, Slovokhotov YL, Coppens P, and Lyubovskaya RN. Synthesis and Structure of Multicomponent Crystals of Fullerenes and Metal Tetraarylporphyrins. Inorg Chem. 2002; 41: 3638–3646.
[39]Belcher WJ, Wagner KI, Dastoor PC. The effect of porphyrin inclusion on the spectral response of ternary P3HT:porphyrin:PCBM bulk heterojunction solar cells. Sol Energy Mater Sol Cells. 2007; 91: 447–452.
[40]Dastoor PC, McNeill CR, Frohne H, Foster CJ, Dean B, Fell CJ, Belcher WJ, Campbell WM, Officer DL, Blake IM, Thordarson P, Crossley MJ, Hush NS, Reimers JR. Understanding and improving solid-state polymer/C60-fullerene bulk-heterojunction solar cells using ternary porphyrin blends. J Phys ChemC. 2007; 111C: 15415–15426.
[41]Berredjem Karst YN, Cattin L, Lakhdar-Toumi A, Godoy A, Soto G, Diaz F, Del Valle MA, Morsli M, Drici A, Boulmokh A, Gheid AH, Khelil A, Bernède JC. The open circuit voltage of encapsulated plastic photovoltaic cells. Dyes Pigm. 2008; 78: 148–156.
[42]Oku T, Wakimoto H, Otsuki A, Murakami M. NiGe-based ohmic contacts to n-type GaAs. I. Effects of In addition. J Appl Phys. 1994; 75: 2522–2529.
[43]Oku T, Furumai M, Uchibori CJ, Murakami M. Formation of WSi-based ohmic contacts to n-type GaAs. Thin Solid Films. 1997; 300: 218–222.
[44]Nagata A, Oku T, Kikuchi K, Suzuki A, Yamasaki Y, Ōsawa E. Fabrication, nanostructures and electronic properties of nanodiamond-based solar cells. Prog Nat Sci. 2010; 20: 38–43.
[45]Nagata A, Oku T, Suzuki A, Kikuchi K, Kikuchi S. Fabrication and photovoltaic property of diamond:fullerene nanocomposite thin films. J Ceram Soc Jpn. 2010; 118: 1006–1008.
[46]Ōsawa E. Recent progress and perspectives in single-digit nanodiamond. Diamond Relat Mater. 2007; 16: 2018–2022.
[47]Korobova MV, Avramenko NV, Bogachev AG, Rozhkova NV, Osawa E. Nanophase of water in nano-diamond gel. J Phys Chem C. 2007; 111: 7330–7334.
[48]Soga T, Kokubu T, Hayashi Y, Jimbo T. Effect of rf power on the photovoltaic properties of borondoped amorphous carbon/n-type silicon junction fabricated by plasma enhanced chemical vapor deposition. Thin Solid Films. 2005; 82: 86–89.
[49]Krishna KM, Umeno M, Nukaya Y, Soga T, Jimbo T. Photovoltaic and spectral photoresponse characteristics of n-C/p-C solar cell on a p-silicon substrate. Appl Phys Lett. 2000; 77: 1472–1474.
[50]Oku T, Kumada K, Suzuki A, Kikuchi K. Effects of germanium addition to copper phthalocyanine/fullerene-based solar cells. Cent Eur J Eng. 2012; 2: 248–252.
[51]Yang J, Nguyen TQ. Effects of thin film processing on pentacene/C60 bilayer solar cell performance. Org Electron. 2007; 8: 566–574.
[52]Oku T, Nakayama T, Kuno M, Nozue Y, Wallenberg LR, Niihara K and Suganuma K. Formation and photoluminescence of Ge and Si nanoparticles with oxide layers. Mater Sci Eng B. 2000; B74: 242–247.
[53]Conibeer G, Green M, Corkish R, Cho Y, Fangsuwannarak T, Pink E, Huang Y, Puzzer T, Trupke T, Richards B, Shalav A, Lin KL. Silicon nanostructures for third generation photovoltaic solar cells. Thin Solid Films. 2006; 511–512: 654–662.
[54]Mart´ı A, L´opez N, Antol´ın E, C´anovas E, Stanley C, Farmer C, Cuadra L, Luque A. Novel semiconductor solar cell structures: The quantum dot intermediate band solar cell. Thin Solid Films. 2006; 511–512: 638–644.
[55]Choi SH, Song H, Park IK, Yum JH. Synthesis of size-controlled CdSe quantum dots and characterization of CdSe–conjugated polymer blends for hybrid solar cells. J Photochem Photobiol A: Chem. 2006; 179: 135–141.
[56]Nozik AJ. Quantum dot solar cells. Physica E. 2002; 14E: 115–120.
[57]Brown P, Kamat PV. Quantum dot solar cells. Electrophoretic deposition of CdSe–C60 composite films and capture of photogenerated electrons with nC60 cluster shell. J Am Chem Soc. 2008; 130: 8890–8891.
[58]Emin S, Singh SP, Han L, Satoh N, Islam A. Colloidal quantum dot solar cells. Sol Energy. 2011; 85: 1264–1282.
[59]Oku T, Motoyoshi R, Fujimoto K, Akiyama T, Jeyadevan B, Cuya J. Structures and photovoltaic properties of copper oxides / fullerene solar cells. J Phys Chem Solids. 2011; 72: 1206–1211.
[60]Oku T, Nomura K, Suzuki A, Kikuchi K. Effect of perylenetetracarboxylic dianhydride layer as a hole blocking layer on photovoltaic performance of poly-vinylcarbazole:C60 bulk heterojunction thin films. Thin Solid Films. 2012; 520: 2545–2548.
[61]Gr¨atzel M. Dye-sensitized solar cells. J Photochem Photobiol C. 2003; 4: 145–153.
[62]Hagfeldt A, Boschloo G, Sun L, Kloo L, Pettersson H. Dye-sensitized solar cells. Chem Rev. 2010; 110: 6595–6663.
[63]Murai S, Mikoshiba S, Hayase S. Influence of alkyl dihalide gelators on solidification of dyesensitized solar cells. Sol Energy Mater Sol Cells. 2007; 91: 1707–1712.
[64]Schmidt-Mende L, Bach U., Humphry-Baker R, Horiuchi T., Miura H, Ito S, Uchida S, Gr¨atzel M. Organic dye for highly efficient solid-state dye-sensitized solar cells. Adv Mater. 2005; 17: 813–815.
[65]Oku T, Kakuta N, Kobayashi K, Suzuki A, Kikuchi K. Fabrication and characterization of TiO2-based dye-sensitized solar cells. Prog Nat Sci. 2011: 21; 122–126.
[66]Lewis LN, Spivack JL, Gasaway S, Williams ED, Gui JY, Manivannan V, Siclovan OP. A novel UV-mediated low-temperature sintering of TiO2 for dye-sensitized solar cells. Sol Energy Mater Sol Cells. 2006; 90: 1041–1051.
[67]Nazeeruddin MK, Kay A, Rodicio I, Humphry-Baker R, Mueller E, Liska P, Vlachopoulos N, Graetzel M. Conversion of light to electricity by cis-X2bis(2,2’-bipyridyl-4,4’-dicarboxylate) ruthenium (II) charge-transfer sensitizers (X = Cl-, Br-, I-, CN-, and SCN-) on nanocrystalline titanium dioxide electrodes. J Am Chem Soc. 1993; 115: 6382–6390.
[68]Kakuta N, Oku T, Suzuki A, Kikuchi K, Kikuchi S. Fabrication and characterization of mixture type dye-sensitized solar cells with organic dyes. J Ceram Soc Jpn. 2009; 117: 964–966.
[69]Palomares E, Clifford JN, Haque SA, Lutz T, Durrant JR. Control of charge recombination dynamics in dye sensitized solar cells by the use of conformally deposited metal oxide blocking layers. J Am Chem Soc. 2003; 125: 475–482.
[70]Silence S, Scott J, Hache F, Ginsbrug E, Jenkner P, Miller R, Twieg R, W. Moerner. Poly(silane)-based high-mobility photorefractive polymers. J Opt Soc Am. 1993; B10: 2306–2312.
[71]Silence S, Scott J, Hache F, Ginsbrug E, Jenkner P, Miller R, Twieg R, Moerner W. Poly(silane)-based high-mobility photorefractive polymers. J Opt Soc Am. 1993; B10: 2306–2312.
[72]Rybak A, Jung J, Ciesielski W, Ulanski J. Photovoltaic effect in novel polysilane with phenothiazine rings and its blends with fullerene. Mater Sci Pol. 2006; 24: 527–534.
[73]Kakimoto M, Kashihara H, Kashiwagi T, Takiguchi T. Visible light photoconduction of poly(disilanyleneoligothienylene)s and doping effect of C60. Macromolecules. 1997, 30, 7816–7820.
[74]Haga Y, Harada Y. Photovoltaic characteristics of phthalocyanine-polysilane composite films. Jpn J Appl Phys. 2001; 40: 855.
[75]Watanabe A, Ito O. Photoinduced electron transfer between C60 and polysilane studied by laser flash photolysis in the near-IR region. J Phys Chem. 1994; 98: 7736–7740.
[76]Kim S, Lee C, Jin MH. Fourier-transform infrared spectroscopic studies of pristine polysilanes as precursor molecules for the solution deposition of amorphous silicon thin-films. Sol Energy Mater Sol Cells. 2012; 100: 61–64.
[77] Shimoda T, Matsuki Y, Furusawa M, Aoki T, Yudasaka I, Tanaka H, Iwasaki H, Wang D, Miyasaka M, Takeuchi Y. Solution-processed silicon films and transistors. Nature. 2006; 440: 783–786.
[78]Lee J, Seoul C, Park J, Youk J. Fullerene/poly(methylphenylsilane) (PMPS) organic photovoltaic cells. Synth Met. 2004; 145: 11–14.
[79]Oku T, Nakagawa J, Iwase M, Kawashima A, Yoshida K, Suzuki A, Akiyama T, Tokumitsu K, Yamada M, Nakamura M. Microstructures and photovoltaic properties of polysilane-based solar cells. Jpn J Appl Phys. 2013; 52: 04CR07.
[80]Huang JY, Lin CY, Shen CH, Shieh JM, Dai BT. Low cost high-efficiency amorphous silicon solar cells with improved light-soaking stability. Sol Energy Mater Sol Cells. 2012; 98: 277–282.
[81]Peumans P, Forrest SR. Very-high-efficiency double-heterostructure copper phthalocyanine/C60 photovoltaic cells. Appl Phys Lett. 2001; 79: 126.
[82]Li L, Tang Q, Li H, Hu W, Yang X, Shuai Z, Liu Y, Zhu D. Organic thin-film transistors of phthalocyanines. Pure Appl Chem. 2008; 80: 2231–2240.
[83]Bamsey NM, Yuen AP, Hor AM, Klenkler R, Preston JS, Loutfy RO. Integration of an M-phthalocyanine layer into solution-processed organic photovoltaic cells for improved spectral coverage. Sol Energy Mater Sol Cells. 2011; 95: 1970–1973.
[84]Bamsey NM, Yuen AP, Hor AM, Klenkler R, Preston JS, Loutfy RO. Heteromorphic chloroindium phthalocyanine films for improved photovoltaic performance. Sol Energy Mater Sol Cells. 2011; 95: 2937–2940.
[85]Kim DY, So F, Gao Y. Aluminum phthalocyanine chloride/C60 organic photovoltaic cells with high open-circuit voltages. Sol Energy Mater Sol Cells. 2009; 93: 1688–1691.
[86]Honda S, Yokoya S, Ohkita H, Benten H, Ito S. Light-harvesting mechanism in polymer/fullerene/dye ternary blends studied by transient absorption spectroscopy. J Phys Chem C. 2011; 115: 11306–11317.
[87]Honda S, Ohkita H, Benten H, Ito S. Selective dye loading at the heterojunction in polymer/fullerene solar cells. Adv Energy Mater. 2011; 1: 588–598.
[88]Honda S, Ohkita H, Benten H, Ito S. Multi-colored dye sensitization of polymer/fullerene bulk heterojunction solar cells. Chem Commun. 2010; 46: 6596–6598.
[89]Oku T, Nose S, Yoshida K, Suzuki A, Akiyama T, Yamasaki Y. Fabrication and characterization of silicon naphthalocyanine, gallium phthalocyanine and fullerene-based organic solar cells with inverted structures. J Phys: ConfSer. 2013; 433: 012025.
[90]Yoshida K, Oku T, Suzuki A, Akiyama T, Yamasaki Y. Fabrication and characterization of PCBM:P3HT bulk heterojunction solar cells doped with germanium phthalocyanine or germanium naphthalocyanine. Mater Sci Appl. 2013; 4: 1–5.
[91]Oku T, Hori S, Suzuki A, Akiyama T, Yamasaki Y. Fabrication and characterization of PCBM:P3HT:silicon phthalocyanine bulk heterojunction solar cells with inverted structures. Jpn J Appl Phys. 2014; 53: 05FJ08.
[92]Oku T, Yoshida K, Suzuki A, Akiyama T, Yamasaki Y. Fabrication and characterization of PCBM:P3HT bulk heterojunction solar cells doped with silicon naphthalocyanine. Phys Status Solidi C. 2013; 10: 1836–1839.
[93]Krebs FC. Air stable polymer photovoltaics based on a process free from vacuum steps and fullerenes. Sol Energy Mater Sol Cells. 2008; 92: 715–726.
[94]Walker B, Kim C, Nguyen TQ. Small molecule solution-processed bulk heterojunction solar cells. Chem Mater. 2011; 23: 470–482.
[95]Lin HY, Huang WC, Chen YC, Chou HH, Hsu CY, Lin JT, Lin HW. BODIPY dyes with β-conjugation and their applications for high-efficiency inverted small molecule solar cells. Chem Commun. 2012; 48: 8913–8915.
[96]Kuwabara T, Nakayama T, Uozumi K, Yamaguchi T, Takahashi K. Highly durable inverted-type organic solar cell using amorphous titanium oxide as electron collection electrode inserted between ITO and organic layer. Sol Energy Mater Sol Cells. 2008; 92: 1476–1482.
[97]Takeda A, Oku T, Suzuki A, Akiyama T, Yamasaki Y. Fabrication and characterization of fullerene-based solar cells containing phthalocyanine and naphthalocyanine dimers. Synth Met. 2013; 177: 48–51.
[98]Scully SR, Armstrong PB, Edder C, Fréchet JM, McGehee MD. Long-range resonant energy transfer for enhanced exciton harvesting for organic solar cells. Adv Mater. 2007; 19: 2961–2966.
[99]Honda S, Nogami T, Ohkita H, Benten H, Ito S. Improvement of the light-harvesting efficiency in polymer/fullerene bulk heterojunction solar cells by interfacial dye modification. ACS Appl Mater Interfaces. 2009; 1: 804–810.