Kai Brinkmann¹,Tim Becker¹,Florian Zimmermann¹,Jarla Thiesbrummel^2,3,Francisco Pena-Camargo³,Cedric Kreusel¹,Manuel Theisen¹,Tobias Haeger¹,Selina Olthof⁴,Christian Tückmantel¹,Manuel Günster¹,Timo Maschwitz¹,Christine Koch⁴,Dirk Hertel⁴,Pietro Caprioglio²,Lorena Perdigón-Toro³,Amran Al-Ashouri⁵,Lena Merten⁶,Alexander Hinderhofer⁶,Leonie Gomell⁷,Siyuan Zhang⁷,Frank Schreiber⁶,Steve Albrecht⁵,Klaus Meerholz⁴,Henry Snaith²,Dieter Neher³,Martin Stolterfoht³,Thomas Riedl¹

Bergische Universitat Wuppertal¹,University of Oxford²,University of Potsdam³,University of Cologne⁴,Helmholtz-Zentrum Berlin⁵,University of Tübingen⁶,Max Planck Institute for Iron Research⁷

Abstract

The concept of multi-junction solar cells is very promising for the future of the solar cell technology, because it provides the potential to overcome fundamental efficiency limits of single-junction devices. The facile bandgap tunability of metal-halide perovskite solar cells renders them particularly attractive for multi-junction architectures. Besides Pb/Sn based perovskite low band gap materials, organic solar cells are becoming increasingly interesting as tandem partner for a perovskite wide bandgap cell. Especially, since the introduction of non-fullerene acceptors has revived the field and enabled organic solar cells with >18% efficiencies and an absorption spectrum extending well into the infrared region.[1] On the other hand, low gap perovskite materials are almost exclusively based on Sn2+ containing structures, where fundamental stability issues are still a serious topic.[2] However, serious concerns currently exist about photo instability of most non-fullerene organic systems. Now, to alleviate these concerns, we evidenced an outstanding operational stability (~ 95% PCE remaining after > 5000 h continuous operation) of an organic single junction that is based on the PM6:Y6 active system, if excitons are exclusively generated on the acceptor material Y6 (which is the case in a tandem application). Aside from the active materials (all-perovskite or perovskite-organic), two terminal tandem cells critically rely on an interconnect that facilitates the recombination of electrons (holes) from the bottom cell with the holes (electrons) from the top cell. In most cases a direct stacking of both cells without dedicated interconnect results in poorly performing devices with s-shaped J-V characteristics. Thus, thin layers (~ 1 nm) of a metal (Ag or Au) are frequently inserted between top- and bottom-cell.[3] However, even an Ag-layer as thin as 1 nm already introduces significant optical losses that lower the EQE of the back-cell and the overall Jsc of the tandem cell as we could show. Therefore, we developed an ultra-thin, ALD grown, metal-like indium oxide interconnect. We leverage the unique property of ALD to provide utmost control of layer thickness even on the level of Angstroms, which is impossible with conventional deposition techniques. We show, that 32 ALD cycles are enough to achieve an amorphous and pinhole free indium oxide layer of only 1.5 nm thickness that provides a charge carrier density of 1020 cm-3. By this it is able to render the interface between the sub cells ohmic. At the same time the AZO/SnOx/InOx layer stack on top of the bottom cell offers a strong protection against the detrimental impact of follow-up processing, which enables swift integration in both perovskite-perovskite and perovskite-organic tandem solar cells. Moreover, due to the very limited thickness, the indium oxide causes only negligible optical losses while simultaneously offering unprecedentedly high shunt resilience. Finally, by combining the indium oxide interconnect with a multi facet optimization of the wide gap perovskite solar cell, we were able to present high efficiency perovskite-perovskite devices (> 23.5%)[4] and, even more, we were able to set a new world record for perovskite-organic tandem solar cells (PCE = 24 %).[5] This result marks a huge step forward since tandem solar cells are a key topic in order to overcome fundamental efficiency limits of single junction perovskites (realistic PCE potential > 30 %). We could show that ALD grown indium oxide can be utilized for both: high efficiency, tin-perovskite containing all-perovskite as well as world record tin-perovskite free, perovskite-organic tandem solar cells. [1] Liu Q, et al. Science Bulletin 65, 272-275 (2020). [2] Lin R, et al. Nature Energy 4, 864–873 (2019). [3] Chen X, et al. Joule 4, 1594-1606 (2020). [4] Thiesbrummel J, Pena-Carmago F, Brinkmann KO, et al (submitted) [5] Brinkmann KO, et al. Nature, 604, 280 (2022)