Nanoscale ferroelectric thin films show promise for solar cells

With the sun providing abundant, renewable clean energy, significant effort is spent on innovatively harvesting light energy through photovoltaic (PV) effects (in which light energy is converted into electricity). Simplified structures to achieve this at reduced cost are critical for practical device applications, including solar cells.

PV effects typically involve two basic processes, including generation of electron-hole pairs as the electrical-charge carriers, and separation of electrons and holes to form the net electric-current flow in a particular direction. Generation and separation of electrons and holes are usually achieved at a material interface. For example, in a conventional semiconductor solar cell, the electric field that exists only in the space-charge region of a p-n junction or Schottky barrier separates the charge carriers (see Figure 1, bottom). By contrast, a ferroelectric thin film could have an internal electric field throughout the bulk region originating from electrical polarization that is not completely canceled out by screening charges. Thus, PV effects are not limited to an interfacial region (see Figure 1, top) and they can be generated without forming complex structures.

Figure 1. Simplified schematics for the bulk photovoltaic (PV) effect in a ferroelectric thin film (top) and the interfacial PV effect in a semiconductor p-n junction (bottom). E: Electric field.

In addition, the photo-induced voltage output (photovoltage) in a ferroelectric thin film is not limited by an energy band gap, as with semiconductor-based PV materials (in which the photovoltage is typically below 1V). Our team previously demonstrated a high photovoltage of 7V in a ferroelectric thin film with a thickness of only 0.72μm, in which the film was in-plane polarized to break the constraint on the photovoltage magnitude imposed by the low film thickness.¹

However, the light-to-electricity conversion efficiency of the bulk PV effect in a ferroelectric thin film is significantly lower than that of the interfacial PV effect. In addition, most ferroelectric thin-film materials in previous studies have had wide energy band gaps, so that they only absorb UV but not visible light.

Through our recent collaboration with the National University of Singapore, we have demonstrated a bulk PV effect in a bismuth ferrite BiFeO₃ (BFO) thin film with large electrical polarization.² Compared to other ferroelectric thin films with a wide energy bandgap, the smaller bandgap of BFO can generate a current in response to visible-light photons. In our experiment, a large portion of the photovoltage and photocurrent (approximately two thirds) is switchable in response to the ferroelectric polarization, with the direction of the photocurrent opposite to that of the polarization vector. In experimental and theoretical work on a different thin film (lead zirconate titanate doped with lanthanum), we showed that nanoscale ferroelectric thin films could significantly improve PV efficiency compared to thicker bulk ferroelectric materials or thin films.^3,4

These results show the potential value of nanoscale ferroelectric thin films for use in solar cells, although the efficiency cannot yet compete with semiconductor materials. We are now working to further clarify the PV mechanisms in ferroelectric thin films, which are much less understood than those operating in semiconductors. We hope to help explain the fundamental limitations and find technical solutions.

The author would like to acknowledge collaboration with the National University of Singapore and contributions by Meng Qin, Wei Ji, Y. C. Liang, Bee Keen Gan, Szu Cheng Lai, Phoi Chin Goh, Yifan Chen, Santiranjan Shannigrahi, and Yee Yuan Tan.