Summary Introduction: Phyllanthus niruri L. (PnL) is a herbaceous plant containing flavonoid quercetin and can be used as an immunomodulator to prevent Covid-19. However, the flavonoid content and yield of herbs extract were not maximized. Therefore, PnL herbs were planted in various harvest periods and application of NPK-chitosan nanopolymer fertilizer to estimate these parameters. Objectives: Determine the effect of NPK-chitosan nanopolymer fertilizer and harvesting age on herb extracts also determine the grouping pattern and correlation between responses based on FTIR-ATR spectral pattern using a chemometric approach. Methods: Each group consisted of 50 plants. The formulation of NPK-chitosan nanopolymer fertilizer based on the dose of NPK consisting of the first dose is 15.5 grams/group, the second dose is 31 grams/group, and the third dose is 7.5 grams/group. Grouping of differences in harvesting age for plants consisted of 4, 6, and 8 weeks after the plant (WAP). Extraction used ultrasound-assisted extraction, and data were analyzed using a chemometric approach. Results: Extract with the highest yield was found in second harvest time and third doses of fertilizer (W2D3) which is 9.73 %, and the highest TFC obtained in an extract with second harvest time and first doses of fertilizer (W2D1) is 17.34 mg QE/g. Total flavonoid content and extract yield were influenced by functional groups at wavenumbers 3486.77–3157.12 cm−1 (1); 1740.96–1670.34 cm-1 (3); 1425.02–1272.62 cm-1 (5); 1257,753–1138,81 cm−1 (6); 1131.38–945.53 cm-1 (7); 711.36–529.23 cm-1 (8). Conclusions: The results showed that harvest time and fertilizer dose affected the growth parameters of PnL, total flavonoid content, and yield of extract. Functional groups in IR spectra also have positive and negative correlations with total flavonoid and yield extract responses.

The optimization approach using the simplex lattice design (SCD) has many advantages, including a minimized number of experiments and a good description of the interactions between components. However, modeling with the SCD approach has not evaluated between responses. Therefore, this study aims to apply chemometric analysis to evaluate the response from the optimization stage using the quercetin – self emulsion formulations (quercetin-SEFs) as a model. SEFs were prepared using grapeseed oil, croduret, and PEG 400. The evaluated responses included emulsification time and transmittance. Both responses were developed in endurance test by centrifugation method and stability test using freeze-thaw. Chemometric analysis on CA produced a dendrogram, while PCA produced score plots, loading plots, scree plots, and biplots. Emulsification time has a positive correlation with transmittance value. The quercetin-SEFs formula in SCD was classified into three groups based on the similarity of characters. Chemometric analysis was successfully applied in evaluating the response to the quercetin-SEFs optimization modeling.