Donghun Kim1 Seungwoo Song2 Hyun Myung Jang3

1, KIST, Cambridge, Massachusetts, United States
2, Korea Research Institute of Standard and Science, Daejeon, , Korea (the Republic of)
3, POSTECH, Pohang, , Korea (the Republic of)

Despite the potential to exceed the Schockley-Queisser theoretical limit, ferroelectric photovoltaics (FPVs) have performed inefficiently due to their extremely low photocurrents. The reported low photocurrent values are mainly attributed to poor light absorptions. Most ferroelectric materials with sufficiently large polarizations, such as LiNbO3, BaTiO3 (BTO), and PZT, have a very wide band-gap energy of >3.0 eV, which places their absorption onset near the ultraviolet (UV). BiFeO3 is known to have a relatively smaller band gap of 2.7 eV; however, this is still far apart from the optimal band-gap range (1.1-1.5 eV) for PV applications. In this regard, in order to realize the potential of FPVs, it is highly desirable to search for a novel ferroelectric material with both strong polarization and optimal band-gap energy.

We propose a recently discovered material, namely β-CuGaO2 [T. Omata et al., J. Am. Chem. Soc. 2014, 136, 3378] as a strong candidate material for efficient ferroelectric photovoltaics (FPVs). According to first-principles predictions exploiting hybrid density functional, β-CuGaO2 is ferroelectric with a remarkably large remanent polarization of 83.80 μC/cm2, even exceeding that of the prototypic FPV material, BiFeO3. Quantitative theoretical analysis further indicates the asymmetric Ga 3-O 2 hybridization as the origin of the Pna21 ferroelectricity. In addition to the large displacive polarization, unusually small band gap (1.47 eV) and resultantly strong optical absorptions additionally differentiate β-CuGaO2 from conventional ferroelectrics; this material is expected to overcome critical limitations of currently available FPVs.