Optimization of diameter of PCL/PVP nanofiber loaded with microcrystalline cellulose using response surface methodology and FTIR-chemometric analysis

Polymer-based electrospun nanofibers offer high surface area and tunable porosity, but polycaprolactone (PCL) is hydrophobic and polyvinylpyrrolidone (PVP) is mechanically weak and water-soluble; incorporating microcrystalline cellulose (CSL) is expected to improve hydrophilicity and performance, yet multicomponent interactions and parameter sensitivity require systematic optimization and chemical validation. This study aimed to optimize the diameter of coaxially electrospun PCL/PVP nanofibers loaded with CSL using Response Surface Methodology (RSM) and to verify composite spectral tendencies via ATR-FTIR combined with chemometrics. A Central Composite Design with 30 runs evaluated four variables—PCL concentration (7.50–17.50 wt%), PVP concentration (7.50–17.50 wt%), CSL concentration (2.50–12.50 wt%), and applied voltage (6–22 kV)—while maintaining flow rate (0.6 mL h −1 ), tip-to-collector distance (100 cm), collector rotation (500 rpm), temperature (50 °C), and humidity (45–65%). Fiber morphology and diameters (ImageJ; ≥100 measurements/sample) ranged from 224.53 to 667.68 nm, with center points yielding the largest and most homogeneous, bead-free networks. A quadratic model was highly significant ( p < 0.0001; F = 274.33) with excellent fit (R 2 = 0.9961; adjusted R 2 = 0.9925; predicted R 2 = 0.9831; lack-of-fit p = 0.5171), identifying PCL, CSL, and voltage as significant main effects, while PVP acted primarily through significant interactions. Numerical desirability optimization (D = 1.000) predicted an optimal formulation at PCL 13.65 wt%, PVP 14.69 wt%, CSL 5.09 wt%, and 10.47 kV, targeting 471.03 nm; SEM validated the prediction (487.80 ± 101.40 nm). ATR-FTIR retained constituent bands but exhibited overlap; chemometrics resolved spectral domains (PCA PC1 47.95%, PC2 20.99%; K-means and HCA separating PCL-, PVP-, and CSL-influenced clusters, with strongest discrimination in the fingerprint region), supporting chemical compatibility without asserting definitive interaction mechanisms. Optimization improved performance beyond diameter control: ultimate tensile strength increased to 99.89 ± 1.35 MPa and Young's modulus to 11.77 ± 0.15 MPa; CSL loading increased storage modulus across 30–200 °C with similar tan δ peak (~95–105 °C); wettability and hydration improved (optimized WCA ~60–75°, water content 84.59%, equilibrium water content 89.25%, swelling ratio 291.65%). The integrated RSM–FTIR-chemometric framework links process optimization with molecular-level consistency and yields mechanically stable, hydrophilic nanofibers with potential for biomedical applications.