1.IntroductionAs a renewable alternative energy source, solar energy is a potential choice for solving the shortage of fossil energy. In this regard solar cells have drawn much attention towards the realization of efficient conversion of solar energy into electric power. However, the widespread use of solar cells has been seriously hampered by the high production cost of electricity generated from the present silicon-based solar cells. Therefore, it is highly desirable to develop low-cost solar cells exhibiting high cell performance. The prospect of utilizing inexpensive materials that can be mass-produced makes organic solar cells fascinating alternatives for future energy sources.1–4It should be pointed out that they have several unique advantages over inorganic solar cells, other than low-cost, for example, light weight, flexibility and colorfulness.The production of electrical power from sunlight in organic solar cells involves the following processes: (i) sunlight photons are absorbed within a photoactive layer, leading to the formation of locally confined excitons, (ii) the excitons migrate to the interface of a donor–acceptor heterojunction and subsequently they dissociate to form free charges consisting of electrons and holes, and (iii) the charges are moved towards respective electrodes to eventually yield a photocurrent in an external circuit. To develop highly efficient organic solar cells, it is essential to elucidate the controlling factors in the processes and optimize each process in the photovoltaic event based on the fundamental information. In this context, extensive efforts have been made in recent years to select suitable donor and acceptor molecules and organize them on an electrode surface at the nanometer scale towards the realization of cell optimization.1–4It has been well established that fullerenes have small reorganization energies of electron transfer (ET), which leads to remarkable acceleration of photoinduced charge separation (CS) and of charge shift as well as deceleration of charge recombination (CR).5The excellent ET properties of fullerenes as acceptors have prompted many researchers to construct fullerene-based photoelectrochemical devices and photovoltaic cells.5–7For instance, fullerenes and their derivatives have always been employed for bulk heterojunction solar cells exhibiting a high cell performance, together with small donor molecules and p-type conjugated polymers, respectively.1,8,9On the other hand, we have successfully combined fullerenes with porphyrins that are an electron donor with excellent light-harvesting propeties, to develop a novel organic solar cell possessing both characters of dye-sensitized and bulk heterojunction solar cells (i.e., a dye-sensitized bulk heterojunction solar cell).10–13In either case the construction of nanohighways for efficient electron and hole transport in donor–acceptor multilayers on electrodes is highly crucial to attain efficient photocurrent generation. In fact many researchers have suggested the importance of an interpenetrating, bicontinuous electron- and hole-transporting network in the blend films of bulk heterojunction solar cells,1,8,9whereas we have exemplified the importance of such nanostructured electron- and hole-transporting highways using alternate porphyrin–fullerene multilayer structures on semiconducting electrodes.13Recently, the integration of a new class of carbon allotropes (i.e., carbon nanotubes (CNTs)) into organic solar cells has attracted much attention in connection with similar carbon-based structures with fullerenes.14Compared with the spherical shape of fullerenes, however, CNTs reveal unique one-dimensional (1-D), nanowire-like structures. The 1-D structures associate closely with the ideal electron- or hole-transporting highway in the active layer of organic solar cells (see Section 3 in detail), as in the case of one-dimensional materials comprising of semiconductors.15Therefore, CNTs are highly promising for transporting electrons or holes efficiently in the blend films of CNTs with donor or acceptor molecules on electrodes. Although CNTs are still expensive relative to other carbon-based materials such as graphite and fullerenes, low-cost production may be possible considering that CNTs also consist of naturally abundant carbon atoms. Intensive efforts have been devoted to reduce the production cost and the price is now continuously reducing as the demand is increasing.16In this review article we will mainly focus on the application of CNTs, especially single-walled carbon nanotubes (SWNTs), to a photoactive layer of photoelectrochemical devices and photovoltaic cells using liquid electrolytes. The applications of CNTs to polymer-based organic solar cells without liquid electrolytes are also described briefly. Representative examples are highlighted with the results of our SWNT-based photoelectrochemical devices.