Influence of loading frequency on implant failure under cyclic fatigue conditions

摘要

Objectives. Preliminary studies on implant fatigue testing suggested that fractures were more likely to occur at 2 Hz than at 30 Hz (χ~2, p<0.05). This investigation explores frequency and base elastic modulus effects on strain, strain rate and failure.rnMethods. A total of 66 implants were mounted in different base materials (acrylic, glass-filled epoxy, aluminum) and loaded up to 10~6 cycles per ISO 14801 (20 N to 420-500 N) at frequencies of 2 Hz and 30 Hz (chosen to accelerate frequency as the stressor). Absolute strain magnitudes and strain rates under varying loading conditions and with different base materials were measured using one strain-gauged implant. Failure probability distributions were analyzed by both Weibull and life data analysis. Measured strain was used to validate an FEA model. Fracture surfaces were examined by SEM.rnResults. Number of failures and failure-rates-per-cycle differed significantly between implants tested at 2 Hz versus 30 Hz (p<0.05). Strain magnitude was independent of frequency. Strain rates were highly correlated with frequency (linear r~2>0.99) and differed significantly under failure conditions (420 N): 2 Hz = 8292 (μstrain/s; (500 N): 30 Hz = 80,840 μstrain/s. Measured and FEA-calculated strains were similar. Fracture surfaces were indistinguishable (2 Hz versus 30 Hz).rnSignificance. Fatigue failure was significantly more likely at 2 Hz than 30 Hz whereas base material and loading magnitude seemed to have only minor influence. Absolute strain was identical at these frequencies suggesting strain rate sensitivity for this commercially pure titanium implant. Both the Weibull and SEM analyses support an identical failure mechanism with damage accumulation more severe at lower frequencies, an interpretation consistent with strain rate sensitivity.