As a transition metal oxide, nickel oxide (NiOx) is theoretically feasible to work as hole selective contacts for silicon solar cells as the valence band offset between NiOx and silicon is much lower than the conduction band offset. However, no promising experimental results have been seen to date.
We tried to optimise the performance of NiOx based hole selective contacts by incorporating Zn into NiOx. Density function theory (DFT) calculations were done to investigate the properties of the Zn-NiO. It was found that the incorporation of Zn can contribute additional electronic states at the top of the valence band, which could be a favourable feature that holes could transport more easily from the silicon substrate to the Zn-NiO. Figure 1 shows the electronic density of states for Ni, Zn and O before and after the incorporation of Zn.
Ultrathin ZnxNi1-xO (ZNO) films with a thickness of ~3.5 nm were synthesized using atomic layer deposition (ALD) using an ALD super-cycle process. Films with different compositions were analysed, i.e., Zn0.09Ni0.91O, Zn0.46Ni0.54O and Zn0.62Ni0.38O which were determined by X-ray photoelectron spectroscopy (XPS). The XPS analysis also revealed that the higher the Zn concentration, the more oxidised the film is. Spectroscopic ellipsometry (SE) measured the optical properties of the films as shown in Figure 2. The refractive index of the films was in a range of 1.6-1.8 which is lower than that of pure NiOx, and Zn0.09Ni0.91O film shows a higher extinction coefficient at low photon energy region which is a sign of sub-bandgap absorption. Although lower than that of pure NiOx, the bandgap of the three films is > 3 eV, which could work well as front contact for coupling light into the silicon absorber. The valence band offset (VBO) between these newly synthesised materials with Si was measured by XPS. It was found that the VBO values are -0.22±0.12 eV, -0.43±0.11 eV and -0.70±0.11 eV for the Zn0.09Ni0.91O, Zn0.46Ni0.54O and Zn0.62Ni0.38O, respectively, and the conduction band offset (CBO) are 2.15±0.14 eV, 1.72±0.13 eV and 1.24±0.13 eV which are all significantly higher than the VBOs. Conceivably, holes can be extracted much easier compared to the electrons.
The contact resistance between the ZNO films with p-Si was measured. All films showed ohmic contact behaviour which indicates that the holes in p-Si can be extracted successfully by the ZNO films. The contact resistivity ρc of the as-deposited films was measured to be 50-60 mΩ∙cm2 as shown in Figure 3. The lowest value was given by the Zn0.09Ni0.91O film, which is consistent with the VBO result. Subsequent rapid thermal annealing in N2 gas atmosphere at 200 oC decreased the contact resistance, and a minimum value of ~21.5 mΩ∙cm2 was obtained from the Zn0.62Ni0.38O sample. Higher temperature annealing revealed good thermal stability of the films as the ρc did not change much until after 500 oC annealing where the ρc increased a bit.
In conclusion, guided by DFT calculations, ALD ZnxNi1-xO films were made to work as hole selective contacts for silicon solar cells. Material analysis indicated that the ZnxNi1-xO can work effectively for this purpose, and was proved by contact measurements. In addition, the ZnxNi1-xO films showed good thermal stability up to temperatures of 500 oC. This work demonstrates that these ALD ZNO films are a promising candidate for application as hole selective contacts for crystalline silicon solar cells.
More details about this work can be found in our recent publication.