Atomic Layer Deposited (ALD) Layers and Interfaces in Perovskite-Based Photovoltaics

Innovation in thin film and interface engineering has played an essential role in pushing the conversion efficiency of the most widespread photovoltaic (PV) technology, i.e. crystalline silicon-based (c-Si), towards its thermodynamic limit. In this respect, ultra-thin, conformal, high-purity Al₂O₃ thin films synthesized by ALD are key in c-Si PV manufacturing industry as they passivate the c-Si surface, thereby suppressing a major channel of electron-hole pair recombination.<br/>In the past years we have explored the ALD synthesis of thin films and interfaces in metal halide perovskite- based photovoltaics. The latter has rapidly reached a conversion efficiency of 26% and, when coupled with c-Si PV in a tandem device, leads to efficiencies already beyond 34%.<br/>In this contribution I will discuss the merits which ALD offers to perovskite-based PV by focusing on NiO-based hole transport layers and SnO₂ buffer layers. Specifically, we have synthesized by ALD thin (&lt; 10 nm) NiO layers which enable homogeneous anchoring of self-assembled monolayers (SAMs) for “lossless” (i.e. in terms of suppression of charge recombination) SAM/perovskite interfaces and narrow spread in conversion efficiency of perovskite [1] and tandem perovskite/c-Si devices [2]. Presently, our studies include the appropriate selection of the ALD Ni-precursor for accurate control over film stoichiometry, Ni oxidation state and layer resistivity, leading to careful engineering of NiO/(SAM)/perovskite interfaces in both wide- and narrow-band gap perovskites for all-perovskite tandem devices.<br/>ALD SnO₂is highly appealing because of its present implementation in perovskite PV R&D and industry as buffer layer, i.e., imparting thermal and environmental stability to the device, while protecting the perovskite absorber and fullerene electron transport layer from the sputtering of the transparent top contact. More recently, ALD SnO₂is explored as solvent barrier layer in the tunnel recombination junction of perovskite/perovskite tandem PV, to prevent the damage of the wide-gap perovskite absorber when processing the narrow- band gap perovskite cell. Although we can conclude that several ALD merits are already extensively acknowledged by the PV community, studies addressing ALD film growth on challenging substrates such as fullerenes and metal halide perovskites are rarely reported in literature. We are convinced that these studies provide the needed rationale to implement more efficiently these layers at device level and promote process upscaling. Therefore, this contribution will also highlight the adoption of <i>in situ</i> diagnostics, namely spectroscopic ellipsometry and IR spectroscopy, to characterize the ALD SnO₂ growth on two commonly adopted fullerenes, C60 and [6,6]-phenyl-C61-butyric acid methyl ester (PCBM). Our studies show that a substrate-inhibited growth occurs in the case of PCBM and, to a minor extent, in the case of C60, with respect to c-Si [3]. Moreover, IR spectroscopy highlights the loss of vibrational features of the ester group in PCBM upon SnO₂ growth, whereas C60 is chemically unaffected. We conclude that the delayed film growth and chemical modifications detected on PCBM are responsible for the consistently lower device performance when ALD SnO₂ is grown on PCBM instead of C60.<br/><br/>[1] Phung et al., ACS Applied Materials & Interfaces 14, 2166 (2021)<br/>[2] Phung et al., Solar Energy Mat. Solar Cells 261, 112498 (2023)<br/>[3] Bracesco et al., under review (2024)