Perovskite materials gain a huge interest in the photovoltaic (PV) community due to its unique characteristics, including long carrier diffusion length, widely tunable bandgap, light absorption potential, and low processing cost. Nowadays, most perovskite fabrication methods employ a solution-based process due to its simplicity and production speed. However, this deposition method provides a non-uniform structure and uses highly toxic solvents, posing the risk of contamination and adverse effects on the environment. On the other hand, a solvent-free method like thermal evaporation can produce a uniform and conformal layer. This method can be used to produce not only the perovskite absorber layer but also the contact layers and transport layers. Depending on the deposition parameters, the resulting morphological properties also change. Therefore, it becomes interesting to understand the detailed knowledge of the film growth and the effects of the deposition parameters on the exact kinetics and the optical properties. Hence, the first objective of this study focuses on developing C60 electron transport layers (ETL) for application in all-evaporated perovskite solar cells (PSCs). The C60 thin film was deposited with different thicknesses of 20, 30, 40 nm and deposition rates 0.3, 0.5, and 1 Å/s on top of the silicon wafer substrate. The resulting surface morphology is obtained from scanning electron microscopy (SEM) and atomic force microscopy (AFM). It indicates that C60 with 40 nm thickness and 0.3 Å/s deposition rate shows a pinhole-free layer with an average surface roughness of 1.05 nm and thickness uniformity of more than 94%. The X-ray diffraction (XRD) measurement shows that decrease of peak intensity as the thickness is reduced from 40 to 20 nm. Moreover, with different deposition rates, 1 Å/s of deposition rate exhibits an asymmetric broadening peak which attributes to the small grain size and the presence of a planar defect in the structure of C60.
The optical analysis has also been performed to get the complex refractive index C60 and identify the effect of deposition rates and layer thicknesses on optical constants. A procedure to extract optical constant for the perovskite absorber layer has been developed during this thesis project using a combination approach of b-spline and Tauc-Lorentz dispersion model. The obtained results were found to be in excellent agreement with experimental work and literature data.
Furthermore, the complete solar cells with p-i-n configuration and semi-transparent perovskite solar cells (ST-PSCs) were optically simulated using GenPro4 software. This simulation aims to identify both the photocurrent density of the perovskite absorber layer and the optical losses caused by parasitic absorption in the supporting layers. In the p-i-n structure, ITO and MoOx layer located on the illuminated side contribute to the main portion of optical loss. Simulations suggest that 40-nm-thick ITO and 10-nm-thick MoOx is an ideal layer stack to deliver high implied photocurrent (22.14 mA/cm2). On the other hand, the optical loss in semi-transparent perovskite solar cells is investigated in two different wavelength regions (i) 300 – 800 nm and (ii) 800 – 1200 nm. In this investigation, the metal back contact is replaced with ITO and cells illuminated from the ETL side. The results show that, in the first wavelength range, the main optical losses are due to reflection, parasitic absorption in the C60 and top ITO layer. These losses are reduced by applying 120-nm-thick anti-reflective coating MgF2 and decreasing the thickness of C60 to 10 nm. Moreover, in the wavelength region of 800 – 1200 nm, the optical losses are mainly affected by the top and bottom ITO, MoOx layer, and reflected light. After optimizing top ITO and MgF2 thickness to 50 and 120 nm, respectively, a 17.07 mA/cm2 of photocurrent transmitted through the cells can be achieved. The light transmittance is ~88%, indicating the potential of semi-transparent perovskite solar cells to be applied in perovskite/silicon tandem devices.