Fibers and Polymers, 2025 (SCI-Expanded)
This study examines the effects of air plasma pre-treatment on the dyeability behavior of silk fabric. Air plasma pre-treatment was applied to both degummed and raw silk fabric samples with different exposure times to modify the fabric surface and make the dyeing process greener and more sustainable. The silk fabric samples were dyed with the natural dye extracted from tipton weed (hypericum perforatum) using an ecological microwave-assisted method. Due to determining the effect of plasma pre-treatment on silk fabric samples, scanning electron microscope and Fourier-transform infrared analysis was achieved. Furthermore, the effect of plasma exposure and dyeing time on colorimetric and fastness properties was investigated. The etching effect of plasma pre-treatment on silk fabric samples was determined using scanning electron microscopic analysis. The experimental results show that plasma pre-treatment, plasma exposure time, and dyeing time affected fastness and colorimetric characteristics. The color strength of samples increased with the degummed process and plasma treatment. The color change of samples improved from 3–4 to 4 with an increase in plasma exposure time for raw silk fabric samples. Rubbing fastness of raw silk fabric samples rose to 5 with plasma treatment. For degummed silk fabric samples, significant improvements in fastness properties have not been seen after plasma treatment. In this study, an optimizable neural network (ONN) model with a Bayesian optimizer was proposed and implemented for predicting the dyeing characteristics of silk fabrics, which are wet and dry rubbing fastness, color change, L, a, b, and K/S. The R-squared (R2), mean absolute error (MAE), root mean squared error (RMSE), and mean squared error (MSE) metrics were used to evaluate the success of the proposed model in terms of predicting the dyeing characteristics. Experimental results indicate that the proposed ONN model is successful in predicting the dyeing properties of silk fabrics.