Organic solar cells (OSCs) show great potential in non-grid energy supply due to its unique properties including lightweight, flexibility, semi-transparency, design flexibility, low cost and so on. Thanks to researchers' tremendous efforts in materials development and device engineering, the power conversion efficiency (PCE) for single-junction OSCs (SJ-OSCs) has exceeded 18%^{[1, 2]}. Further increasing PCE is still demanded in order to make this technology commercially attractive. Several directions have been proposed to push the PCE of SJ-OSCs beyond 20%^[3], e.g., reducing non-radiative voltage loss, increasing both exciton and charge-carrier mobility, reducing optical loss and so on. Most of the requirements rely on next breakthroughs in materials development. By maximizing the utilization of solar energy, theoretically, tandem organic solar cells (T-OSCs) can achieve a higher PCE than the state-of-the-art SJ-OSCs. Here, we highlight high-performance T-OSCs with active layers made by solution-processing.

Before the significant advances in non-fullerene acceptors (NFAs), the PCE of high-performance T-OSCs was 10%–12%^[4]. Most of the active layers adopted donors (D) with complementary absorption and fullerene derivatives as acceptors (A). In 2017, Chen et al. reported a PCE of 12.7% for T-OSCs based on small molecules DR3TSBDT and DPPEZnP-TBO combined with fullerene acceptors, which was among the highest PCEs by then^[5]. Noticeably, the PCE of T-OSCs was rapidly improved over 12% by incorporating NFAs as photoactive materials. In 2017, Hou et al. reported 12.8% T-OSCs combining front cells based on wide-bandgap blend (PBDD4T-2F:PC₇₁BM) and rear cells based on low-bandgap blend (PBDTTT-E-T:IEICO)^[6]. Later, they used NFAs in both sub-cells (PBDB-T:ITCC-M and PBDTTT-E-T:IEICO) and pushed the PCE to 13.8%^[7]. In 2018, Chen et al. reported a PCE of 17.6% for T-OSCs based on PBDB-T:F-M and PTB7-Th:O6T-4F:PC₇₁BM as the active layers for sub-cells. Owing to the excellent near-infrared absorption of O6T-4F (i.e. COi8DFIC^[8], invented by Ding et al.), a broad spectral response (300–1050 nm) was achieved^[9]. Since then, several other T-OSCs with PCEs over 16% were also developed^[10-13].

Recently, Huang et al. have made significant progress in device engineering of T-OSCs with PM7:TfIF-4Cl blend as the front active layer and PCE10:COi8DFIC:PC₇₁BM blend as the rear active layer (Figs. 1(a) and 1(b))^[14]. The short-circuit current density (J_sc) and fill factor (FF) were balanced by simultaneously optimizing active layer thickness of the two sub-cells and the D : A ratio of the front cell with the support of optical simulation (Figs. 1(c) and 1(d)). By increasing the acceptor portion, the absorption coefficient can be increased, which reduces the optimum thickness of the front active layer for photocurrent matching. This strategy is beneficial for suppressing bimolecular recombination in thick active layers, thus leading to a FF improvement. Finally, a record PCE of 18.7% and a high FF of 78% were achieved, with a certified PCE of 18.09% by a third party.

So far, several groups have reported results on storage stability study. Chen et al. demonstrated T-OSCs with only 4% degradation after 166-days storage (Fig. 2(a))^[9]. Huang et al. showed that their T-OSCs with the record PCE retained over 95% of the initial PCE after 500 h storage in nitrogen atmosphere^[14]. The stability under illumination is yet to be studied. Recently, Li et al. developed a T-OSC with a PCE of 16.4%. Promising shelf-life stability (>98% after 500 h) was reported. They performed 500 h stability test under metal halide lamp (Fig. 2(b)). The stability was limited by the front cell, and the T-OSCs showed a medium degradation rate (Fig. 2(b))^[12]. In their previous work, the photostability of the T-OSCs can be significantly improved even though the front sub-cell based on PTQ10:m-DTC-2Cl suffered fast degradation, which hold promising results for the stability of T-OSCs (Fig. 2(c))^[11]. When aged under indoor light, the T-OSCs retained 95% of the initial PCE after one month, demonstrating potential of T-OSCs for indoor application. Despite significant improvement in PCE and initial promising lifetime results, photostability studies on T-OSCs are still rare.

In summary, with the broad choice of advanced donors and NFAs for absorption-spectrum matching, T-OSCs hold great potential to further boost the performance. Although complex tandem structure with multiple functional layers still faces stability challenge, some T-OSCs presented encouraging results even when one sub-cell suffered from strong degradation. T-OSCs may have chance to approach the theoretical PCE limit.

X. Du thanks the Taishan Scholar Foundation of Shandong Province (tsqn202103016) and Qilu Young Scholar Program of Shandong University. L. Ding thanks the National Key Research and Development Program of China (2017YFA0206600) and the National Natural Science Foundation of China (51773045, 21772030, 51922032 and 21961160720) for financial support.