摘要： The photovoltaic efficiency increase of Sb2S3-based solar cells has stagnated for five years since the highest efficiency 7.5% was achieved in 2014 [Adv. Funct. Mater. 24, 3587]. One important bottleneck is the high electrical resistivity of Sb2S3. Our first-principles calculations reveal that the high resistivity results from the compensation between the intrinsic donor VS and acceptors VSb, SbS, SSb which have comparably high concentration (low formation energy). The compensation also limits the improvement of conductivity through direct extrinsic doping. Further calculations of O dopants show that OS has a low formation energy, so the dominant intrinsic donor VS can be passivated by O and thus the p-type doping limit imposed by VS can be overcame. Meanwhile, other p-type limiting and recombination-center donor defects can be suppressed under the S-rich condition, which explains why the highest efficiency was achieved in O-doped Sb2S3 after sulfurization. Given the unexpected beneficial effects of O doping and sulfurization, a two-step doping strategy is proposed for overcoming the efficiency bottleneck: (i) use O to passivate the VS and S-rich condition to suppress other detrimental defects, making p-type doping feasible and minority carrier lifetime long; (ii) introduce other p-type dopants to increase hole carrier concentration.