Investigations on the elastic and plastic dynamic pulse buckling response of a slender beam, with geometric imperfections, subject to an axial impulse is presented. The current study is concerned with high order events that can be defined as an intense transient loading condition (i.e. large amplitude, short time period), where considerations of the stress wave front propagation are important. The dynamic buckling instability was primarily viewed as a modal perturbation due to an unbounded growth of transverse deflections due to exploitation of the geometric imperfections by the applied impulse. Experimental investigations and numerical analyses, employing the finite difference and finite element methods, were conducted.

For a slender beam subject to an axial impulse, the elastic studies substantiated the validity and utility of the "preferred" wavelength theory. The peak buckle amplitude and dominant waveform could be effectively defined in terms of a single "preferred" mode. The most significant conclusion was that to accurately model the pulse buckling event, random geometric imperfections were required to develop the hyperbolic growth of transverse displacements for both the finite difference and finite element models. Other modelling considerations were also required for the development of an accurate pulse buckling numerical model with respect to theory and experimental observations. These factors include adopting a finely discretised mesh with aspect ratios on the order of 1:​1 and employing plane strain elements. The superior performance of the quadratic plane strain elements was related to inherent curvature restriction of the beam element formulation. For plane strain element models, however, nonlinear geometric and material analysis was required. A threshold elastic pulse buckling criterion was substantiated by the finite difference study and further supported by finite element analyses. The influence of more complex structures, boundary constraints or loading conditions on the accuracy of numerical models investigating elastic pulse buckling events should be examined.

The experimental investigations and finite element analyses demonstrated that, for the parameters investigated, the dynamic plastic pulse buckling events could also be defined by a characteristic modal response. The peak buckle response was assessed in terms of a normalised modal parameter as a function of the effective slenderness ratio. The modal parameter accounted for the longitudinal position of the peak buckle crest with respect to local stiffness terms and natural frequency characteristics defined with respect to the axial and flexural response. The analyses demonstrated the need for further parametric investigations to examine the influence of element formulation and imperfection type on the pulse buckling response. The importance of slenderness ratio, element aspect ratio and constitutive relationships (e.g. dynamic yield flow, rate-sensitive behaviour, viscoplasticity) should also be considered. Furthermore, based on these recommendations, future work should address the development of dynamic pulse buckling threshold limits.