We analyse the statistical distribution function for the height fluctuations of brittle fracture surfaces using extensive experimental data sampled on widely different materials and geometries. We compare a direct measurement of the distribution to a new analysis based on the structure functions. For length scales δ larger than a characteristic scale δ * , we find that the distribution of the height increments ∆h = h(x + δ) − h(x) is Gaussian. Self-affinity enters through the scaling of the standard deviation σ, which is proportional to δ ζ with a unique roughness exponent. Below the scale δ * we observe an effective multi-affine behavior of the height fluctuations and a deviation from a Gaussian distribution which is related to the discreteness of the measurement or of the material.

We study experimentally the propagation of an in-plane fracture into a transparent and heterogeneous Plexiglas block. A stable crack propagation in mode I is monitored by an imposed displacement. The experimental setup allows a high resolution observation of the crack front in situ. Self-affine properties of the crack front are described over more than three decades using several techniques: variable bandwidth, return probability, Fourier spectrum, and wavelet analysis. The different methods lead to a roughness exponent of 0.63+/-0.03, consistent with a previous work.

No abstract