**1. Introduction**

Finding alternatives for the current energy sources (i.e., burning fossil fuels, nuclear materials) has become one of the most important societal challenges for relieving the environmental pollution problem [1]. One of the most promising next-generation energy sources is solar energy, which can be converted to electric power via photovoltaic technology. Currently industrialized photovoltaic panels are based on inorganic materials such as silicon [2–4]. Recently, organic solar cells (OSCs) emerged as an alternative to inorganic photovoltaics devices [5–7]. The merits of OSC technology are: a low-cost solution process, a low temperature process, flexibility, and a tailorable material for further improvement. Recently, the champion single-junction OSC has reached a power conversion efficiency (PCE) of 12.6% [8]. However, it is still necessary for improving PCE and air-stability for large-scale commercialization.

One of the reasons for the low PCE is the narrow light absorption range of organic materials. A tandem solar cell structure, where two or more single-junction cells with complementary absorption spectra are connected in tandem, can be a promising design to overcome the limitations of single cells [9–11]. A tandem structure offers several advantages: (1) a broad absorption spectrum due to the usage of complementary absorbing materials; (2) summation of the open circuit voltage (VOC) of each sub-cell; (3) a reasonable fill factor (FF) due to higher optical density over a wider fraction of solar spectrum than that of single cell without increasing internal resistance. To maximize these advantages, the tandem device requires the qualified charge recombination layer (CRL) to simultaneously act as the anode for one of two adjacent sub-cells, and as the cathode for the other [12]. The CRL should have

low electrical resistance, high optical transparency in the visible range, and a low barrier for charge recombination. Furthermore, the layer should be able to protect the lower layers during the remaining solution fabrication process.

In 2007, the CRL made with a metal oxide material was suggested [13]. The advantage of using a metal oxide as a part of CRL is to minimize the absorption at visible wavelengths. The classical structure of CRL is based on 0.5 nm LiF/1 nm Al/3 nm WoO3, used as multilayers. The first all-solution-processable tandem OSC was reported by Kim et al., where poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) and TiOX were used to form CRL [14]. In 2013, the first inverted tandem OSC that broke 10% efficiency was reported by You et al., with PEDOT:PSS and ZnO being used as CRL [15]. As shown above, new efficient CRLs with a large work function difference between two opposite interfaces have been developed continuously. Therefore, is high VOC of the tandem device without loss-in-sum of the VOC of the component cells is achieved. However, maximizing the short circuit current of tandem device is still under development, due to the difficulties in the current matching between the component cells. The current of tandem device is limited by the cell that produces the lower current. Therefore, it is important to increase the short circuit current of each component cell by enhancing the charge extraction properties of CRL. However, to our knowledge, there has been no reported research on the tandem device with patterned CRL, even though the 3D pattern could enhance the charge extraction capability of the CRL.

In the single junction solar cell, patterning of the charge-collecting layer (CCL) has been developed widely to maximize the charge extraction ratio [16–21]. However, previously proposed fabrication techniques for growing nano-wires, nano-rods, and nanoporous layers are not compatible with tandem structure fabrication. High thermal annealing conditions or vacuum processes should be avoided for all solution-processable tandem devices, to prevent the degradation of the films underneath. Meanwhile, a solution-processable ZnO nano-ripple pattern was firstly shown in a single sub-device by Yang et al. in 2009, by using a ramping thermal annealing method [22]. By using patterned ZnO CCL, the PCE improved by about 25% compared with device-containing planar CCL. However, the high thermal annealing treatment condition of this technique (275–350 ◦C) limited the use of this ripple patterning only in single inverted OSCs [22–24].

Herein, we found the optimal process conditions for nano-ripple patterning of the ZnO film in low-temperature conditions. After confirming the characteristics of ZnO in the single sub-cell device, the layer was introduced as a part of CRL in the tandem device. A newly developed low-temperature process technique actualized the patterning on the CRL without damaging the films underneath. As compared to the non-patterned device, the tandem device with ZnO ripple-patterned CRL showed a PCE improvement of about 30%.
