RSM-ANN-assisted interfacial engineering of Ni@ag core-shell powders for high-conductivity electronic pastes

Ni@Ag powders combine the cost advantage of Ni with the high electrical conductivity of Ag, making them promising conductive fillers for electronic applications. However, their performance is strongly governed by the integrity of the Ag shell. A discontinuous or nonuniform Ag coating can deteriorate interparticle electrical contact and expose the Ni core to oxidation during thermal treatment, thereby reducing electrical conductivity and stability. In this work, the wet-chemical deposition process for Ni@Ag powders was optimized using a hybrid response surface methodology-artificial neural network (RSM-ANN) strategy. Ascorbic acid concentration, solution pH, and feeding rate were selected as the key process variables, while compacted resistivity was used as the response parameter. The ANN model showed a high fitting accuracy R 2 = 0.9954 and retained strong predictive consistency under leave-one-run-out cross-validation, with a mean cross-validated coefficient of determination of R 2 LOOCV = 0.9728. Under the optimized conditions, the compacted resistivity was reduced markedly from 87.150 to 34.642 μΩ·cm, representing a decrease of 60.25%. Further morphological, structural, and thermal analyses confirmed that the optimized process facilitated the formation of a more compact and continuous Ag coating on the Ni surface, which in turn enhanced the oxidation resistance and thermal stability of the powders. Application-level validation showed that the S4-based conductive film exhibited a 63.94% lower electrical resistivity than the S1-based film and formed a more continuous conductive network. These findings suggest that the RSM-ANN strategy provides an effective approach for controlling Ag deposition behavior and improving the performance of Ni@Ag core-shell conductive fillers.