Optimizing multiscale resin flow kinetics and impregnation uniformity in fiber-reinforced polymers composites via hierarchical carbon nanostructures
Yu Zhang , Muhan Zhang , Helezi Zhou , Rongshan Zhou , Yiu Wing Mai , Huamin Zhou
Microstructures ›› 2026, Vol. 6 ›› Issue (5) : 20260114
Resin flow dynamics and impregnation uniformity are critical factors governing the quality and manufacturing efficiency of carbon fiber-reinforced polymer (CFRP) composites fabricated by liquid composite molding (LCM). To address these challenges, a hierarchical three-dimensional carbonaceous nanostructure was constructed on carbon fiber surfaces. This architecture was designed to enhance resin impregnation across multiple scales: single fibers at the microscale, fiber bundles at the mesoscale, and fiber fabrics at the macroscale. Experimental results showed that the hierarchical structure significantly improved the wettability of the carbon fibers, reducing the contact angle from 50.65° to 34.95°. Meanwhile, the capillary infiltration rate constant of the fiber bundles and the average resin infiltration velocity through the fabric increased by 84.2% and 85.7%, respectively. The enhanced intra-bundle infiltration reduced the velocity mismatch between intra-bundle and inter-bundle flows, thereby improving the overall uniformity of resin impregnation. Moreover, molecular dynamics (MD) simulations were performed to elucidate the underlying infiltration mechanisms. The simulations revealed that the hierarchical structure increased the magnitude of the capillary pressure within the microchannels by 166.4%, leading to enhanced capillary-driven resin transport. Importantly, the nanostructured interface promoted molecular-chain migration within the microchannels by modifying solid-liquid interactions, thereby facilitating resin infiltration within the fiber bundles.
Liquid composite molding / hierarchical carbon nanostructures / multiscale infiltration / impregnation uniformity / carbon fiber-reinforced polymers composites
| [1] |
|
| [2] |
|
| [3] |
|
| [4] |
|
| [5] |
|
| [6] |
|
| [7] |
|
| [8] |
|
| [9] |
|
| [10] |
|
| [11] |
|
| [12] |
|
| [13] |
|
| [14] |
|
| [15] |
|
| [16] |
|
| [17] |
|
| [18] |
|
| [19] |
|
| [20] |
|
| [21] |
|
| [22] |
|
| [23] |
|
| [24] |
|
| [25] |
|
| [26] |
|
| [27] |
|
| [28] |
|
| [29] |
|
| [30] |
|
| [31] |
|
| [32] |
|
| [33] |
|
| [34] |
|
| [35] |
|
| [36] |
|
| [37] |
|
/
| 〈 |
|
〉 |