(Invited) Lithium-Ion Endohedral Fullerenes on Carbon Nanotube Electrode-Laminated Perovskite Solar Cells As Dopants and Anti-Oxidants

Abstract

Lead halide perovskite solar cells (PSCs) have become an established photovoltaic technology heading towards commercialisation with exceptionally high conversion efficiency (PCE). Although PSCs show high PCEs, the device stability needs to be substantially improved if PSCs are to be commercialized. There are mainly two factors that are responsible for the low stability of PSCs; they are 2,2′,7,7′-tetrakis(N,N-di-p-methoxyphenylamine)-9,9′-spirobi-fluorene (spiro-MeOTAD) hole-transporting layer (HTL) and metal electrode. Use of hygroscopic lithium bis(trifluoromethanesulfonyl)imide (Li+TFSI−) in spiro-MeOTAD results in an uncontrolled oxidation of spiro-MeOTAD as well as moisture-driven degradation. Also, using metal electrodes induces a metal ion-driven degradation, in which metal ions on top migrate into the perovskite layer, lowering the PSC stability. As a solution to the Li+TFSI−-driven degradation, we reported lithium-ion-containing [60]fullerene trifluoromethanesulfonylimide salt ([Li+@C60]TFSI−). [Li+@C60]TFSI−induced an instant oxidation of spiro-MeOTAD producing spiro-MeOTAD•+TFSI− and neutral [Li+@C60]•−(= Li@C60), which functioned as an antioxidant, protecting PSCs from intruding oxygen. With the controlled production of spiro-MeOTAD•+TFSI−and anti-oxidation activity of Li+@C60 •−, the stability of PSCs improved by 10-fold compared with the reference devices. However, the PCE of the PSCs was limited by the lack of spiro-MeOTAD•+TFSI−production and aggregated morphology of HTL, both of which arise from the low solubility of [Li+@C60]TFSI−and Li@C60. Moreover, the stability can be improved even further by removing the metal ion migration from the top metal electrode. Replacing the metal electrode by carbon electrode has been reported to be one of the most effective ways to enhance the device stability of PSCs due to no ion migration, and outstanding encapsulation effect. Among the carbon electrodes, the application of free-standing carbon nanotubes (CNTs) has given the highest PCEs with the use of spiro-MeOTAD HTM. However, it is reported that the use of spiro-MeOTAD limits the full potential of the CNT-PSCs in terms of long-terms stability. Combining the two technologies described above can provide a synergic and ultimate solution to the PSC stability. Therefore, we incorporated the mixture of spiro-MeOTAD and [Li+@C60]TFSI−into the CNT top electrode in PSCs. The HTL solution is typically drop-casted onto the CNT network in CNT-PSCs. When the mixture of spiro-MeOTAD and [Li+@C60]TFSI−was drop-casted, a saturated solution seeped through the CNT network while any undissolved [Li+@C60]TFSI−and Li@C60suspensions stayed on the top. This led to more effective hole extraction by avoiding [Li+@C60]TFSI−in the pathway and more effective anti-oxidation activity by placing Li@C60next to air. Since the drop-casting on CNTs separated the dissolved species from the undissolved species, the oxidation reaction of spiro-MeOTAD came to a stop. Thus, the stirring time of the [Li+@C60]TFSI−-containing spiro-MeOTAD solution was directly linked to the photovoltaic performance and stability of the PSCs. From various analyses and investigation, we found that 2 hour stirring for the HTL solution gave the highest PCE of 17% and the longest operating stability of 1000 hours when unencapsulated. The PCE value of the new devices is close to the PCE of gold electrode-based PSCs (18.5%) while the PCE stability is approximately 100 times greater. Such excellent stability is attributed to no ion-migration and antioxidant activity of Li@C60uniformly covering the CNT electrode. Not only did we demonstrate highly stable and efficiency CNT-PSCs but also discovered a new reaction mechanism within the spiro-MeOTAD and [Li+@C60]TFSI−HTL solution. Figure 1