Experimental characterisation of self-healing composite structures for low and high-velocity impact applications

Sandwich composite structures are widely employed in aerospace applications because of their high specific stiffness and low mass; however, their susceptibility to impact damage remains a major limitation, particularly under repeated loading and in service environments where residual integrity is critical. Conventional structural adhesives provide joining efficiency, but they do not restore local damage once impact has occurred. Self-healing elastomeric systems therefore offer a potentially valuable route to improving post-impact tolerance in composite sandwich structures. This study investigates whether the self-healing elastomer Reverlink™ can improve the impact response of carbon/Dyneema®–Nomex® sandwich structures, relative to the ductile epoxy adhesive Araldite® 2015, under low-velocity impact (LVI) and high-velocity impact (HVI) conditions. Sandwich panels comprising five-ply carbon/Dyneema® face sheets and a 10 mm Nomex® honeycomb core were manufactured and tested experimentally. For LVI, five Reverlink™ specimens were subjected to a 20 J drop-weight impact, followed by 24 h conditioning at 90°C and a second 20 J impact; Araldite® 2015 reference specimens were also tested. HVI tests were conducted using a compressed-air cannon, with damage quantified through area and indentation measurements. Under 20 J LVI, Reverlink™ specimens exhibited a mean damaged area of 242.12 mm², compared with 314.16 mm² for the Araldite® 2015 reference specimen. After conditioning and re-impact, the damaged area remained unchanged, whilst the mean maximum indentation depth increased from 6.87 to 10.26 mm, indicating partial local damage recovery, although full restoration of the original mechanical properties was not demonstrated. In HVI, Reverlink™ specimens absorbed 39.31–40.47 J at approximately 139 m/s and 55.78–57.57 J at 92–95 m/s. Reverlink™ improved LVI damage tolerance relative to the epoxy reference and exhibited evidence of local post-impact recovery after thermal conditioning. However, direct HVI comparison between adhesives was not possible because different projectile conditions were used, and further post-healing mechanical characterisation is required.