On the other hand, remarkable enhancement in overall efficiency (?13%) was also observed from four PSCs of Batch 2 (as shown in FIG. 4 a-d with blue circles with TiO₂+ZrO₂ thickness=2.9 μm) due to an impressive improvement in J_SC and FF after storing them in vacuum (in dark) for three weeks. The champion device upon re-measuring exhibited an improved performance with an overall conversion efficiency of 9.53% (FIG. 7) among the average efficiency (9.4±0.1%) of these four PSCs thus showing tremendous potential of their entire fabrication with non-vacuum based equipment and processes as shown in FIG. 1 (a-h).

The results shown here are very encouraging to adopt inkjet printing as a method of perovskite precursor infiltration in a highly porous structure which offers several advantages over manual infiltration in this type of PSC configuration. For instance there is no additional optimized environment required such as glove box since the precursor ink remain isolated in the cartridge tank and reduces the chances of contamination that cannot be avoided during manual infiltration where the micropipette could touch the carbon electrode and can also damage the active area due to human error. Additionally the automated inkjet infiltration allows homogenous dispensing and distribution of perovskite precursor ink on the active electrode area thus provides high probability of device performance reproducibility which cannot be realized with manual infiltration scheme. This effect was significantly evident among different batches of PSCs fabricated for this experiment where the sample to sample variations were impressively very low and even the enhancements in performance against the variance in ZrO₂ thickness were very systematic.