Proceedings of International Conference on Hybrid and Organic Photovoltaics (HOPV19)
Publication date: 6th February 2020
Indoor photovoltaics (PV) is a perspective application for solar cells in domotic applications and low-power devices for the internet of things [1], [2]. The emerging and low-cost solution-processed halide perovskite solar cells (PSC)[3]–[5] is currently demonstrating unprecedented progress for power conversation efficiency (PCE) at 1 sun conditions exceeding 23% [6]. At the same time, PSC are showing competitive output characteristics for indoor (>40 uW/cm2) at 400 lux [7] in comparison to commercially available Si, GaAs and DSSCs devices[8]–[10]. However, the typical architecture for high performing indoor PSCs, that is the n-i-p structure, has inherent hysteresis of the JV characteristics[11] due to parasitic accumulations effects[12], that can negatively contribute to maximum power performances. Thus, optimization of PSC structure for indoor use is still a challenging task. As reported for operation of PSC with inverted planar configuration at standard conditions (1.5 AM G, 100 mW/cm2), the use of p-i-n structure significantly reduces hysteresis effect[13]. However, there are only few reports presented in the literature for p-i-n indoor PSC[14] and no any published results for the device structure with stable inorganic hole transport layer (HTL).
In this work we developed p-i-n indoor PSC with nickel oxide (NiO) HTL deposited from the tris (ethylene diamine) nickel acetate precursor (Tprocess=300 ̊C, 10 nm compact film) and nanoparticles (NP) dispersion (Tprocess=100 ̊, 60 nm nanoparticles film). We provided comparison of device performance fabricated on different transparent conductive electrodes (TCOs - planar FTO and ITO) with the two types of NiO HTL to define the impact of cell architecture on output characteristics. For all types of NiO HTL we observed the absence of hysteresis during JV sweeps, but a strong correlation to Jsc and Voc. In general, at 200 and 400 lux of illumination (white LED), devices with compact NiO HTL showed 20-26 % larger Jsc and, at the same time, 4-10 % lower Voc with respect to NiO NP HTL. Maximum power densities (Pmax) extracted from JV plots at 200 lux achieved at level of 15.9 and 16.8 µW/cm2 for compact NiO HTL devices on FTO and ITO substrates, respectively. On the other hand, cells with NiO NP HTL provided 13.9 µW/cm2 on ITO and 8.8 µW/cm2 on FTO. At 400 lux, compact NiO HTL devices reached a Pmax value of 43.7 µW/cm2 on ITO and 39.6 µW/cm2 on FTO, while NP HTL cells demonstrated 39.2 µW/cm2 on ITO and 27.3 µW/cm2 on FTO (see Fig. 1).
Such difference in Jsc values was investigated with measurement of dark-current characteristics, where devices with NiO NP HTL demonstrated larger shunting leakages on both types of used TCOs. Transient photovoltage study showed that heterojunction with NP has faster dynamics of relaxation due to lower concentration of defects on the interface, that corresponds to higher values of Voc for the devices with NiO NP HTL. We found that achieved values for generated power densities under dim-light illumination are currently the best for inverted planar structures presented in the literature. The stability of indoor solar cells structure was proved during maximum power point tracking under 200-400 lux conditions several hundreds of hours.
The authors gratefully acknowledge the financial support of the Ministry of Education and Science of the Russian Federation in the framework of Megagrant No. 14.Y26.31.0027.