To unlock the full potential of PSCs, machine learning (ML) was implemented in this research to predict the optimal combination of mesoporous-titanium dioxide (mp-TiO₂) and weight percentage (wt%) of phenyl-C₆₁-butyric acid methyl ester (PCBM), along with the current density (J_sc), open-circuit voltage (V_oc), fill factor (ff), and energy conversion efficiency (ECE). Then, the combination that yielded the highest predicted ECE was selected as a reference to fabricate PCBM-PSCs with nanopatterned TiO₂ layer. Subsequently, the PCBM-PSCs with nanopatterned TiO₂ layers were fabricated and characterized to further understand the effects of nanopatterning depth and wt% of PCBM on PSCs. Experimentally, the highest ECE of 17.338% is achieved at 127 nm nanopatterning depth and 0.10 wt% of PCBM, where the J_sc, V_oc, and ff are 22.877 mA cm⁻², 0.963 V, and 0.787, respectively. The measured J_sc, V_oc, ff, and ECE values show consistencies with the ML prediction. Hence, these findings not only revealed the potential of ML to be used as a preliminary investigation to navigate the research of PSCs but also highlighted that nanopatterning depth has a significant impact on J_sc, and the incorporation of PCBM on perovskite layer influenced the V_oc and ff, which further boosted the performance of PSCs.

Surface-enhanced Raman scattering (SERS) enables rapid detection of single molecules with high specificity. However, quantitative and sensitive SERS analysis has been a challenge due to the lack of reliable SERS-active materials. In this study, we developed a quantitative SERS-based immunoassay using enzyme-guided Ag growth on Raman labeling compound (RLC)-immobilized gold nanoparticle (Au NP)-assembled silica NPs (SiO₂@Au-RLC@Ag). The enzyme amplified Ag⁺ reduction as well as Ag growth on the RLC-immobilized Au NP-assembled silica NPs (SiO₂@Au-RLC), which resulted in a significant increase in SERS signal. In the presence of target antigens such as immunoglobulinG (IgG) or prostate-specific antigen (PSA), Ab₁-Antigen-Ab₂ immune complex with alkaline phosphatase triggered an enzyme-catalyzed reaction to convert 2-phospho-L-ascorbic acid (2-phospho-L-AA) to ascorbic acid (AA). As produced AA reduced Ag⁺ to Ag, forming an Ag hot spot on the surface of SiO₂@Au-RLC, which enhanced the SERS signal of SiO₂@Au-RLC@Ag in a solution with a target antigen concentration. The plasmonic immunoassay for IgG detection showed a high linearity of SERS intensity in the range of 0.6 to 9.0 ng/mL with a detection limit (LOD) of 0.09 ng/mL, while an LOD of 0.006 ng/mL was obtained for PSA. The results indicate that the sensitivity of our novel SERS-based immunoassay is higher than that of conventional enzyme-based colorimetric immunoassays.