A numerical investigation and experimental verification of size effects in loaded bovine cortical bone

Tools

Frame, J.C. and Wheel, M.A. and Riches, P.E. (2018) A numerical investigation and experimental verification of size effects in loaded bovine cortical bone. International Journal for Numerical Methods in Biomedical Engineering, 34 (1). e2903. ISSN 2040-7939 (https://doi.org/10.1002/cnm.2903)

[thumbnail of Frame-etal-IJNMBE2017-experimental-verification-of-size-effects-in-loaded-bovine-cortical]
Preview
 Text. Filename: Frame_etal_IJNMBE2017_experimental_verification_of_size_effects_in_loaded_bovine_cortical.pdf
Accepted Author Manuscript
Download (1MB)| Preview

Abstract

In this paper we present two and three dimensional finite element based numerical models of loaded bovine cortical bone that explicitly incorporate the dominant microstructural feature: the vascular channel or Haversian canal system. The finite element models along with the representation of the microstructure within them are relatively simple: two dimensional models, consisting of a structured mesh of linear elastic planar elements punctuated by a periodic distribution of circular voids, are used to represent beam samples of cortical bone in which the channels are orientated perpendicular to the sample major axis, while three dimensional models, employing a corresponding mesh of equivalent solid elements, represent those samples in which the canals are aligned with the axis. However, these models are exploited in an entirely novel approach involving the representation of material samples of different sizes and surface morphology. The numerical results obtained for the virtual material samples when loaded in bending indicate that they exhibit size effects not forecast by either classical (Cauchy) or more generalized elasticity theories. However, these effects are qualitatively consistent with those that we observed in a series of carefully conducted experiments involving the flexural testing of bone samples of different sizes. Encouraged by this qualitative agreement we have identified appropriate model parameters, primarily void volume fraction but also void separation and matrix modulus by matching the computed size effects to those we observed experimentally. Interestingly, the parameter choices that provide the most suitable match of these effects broadly concur with those we actually observed in cortical bone.

ORCID iDs

Frame, J.C., Wheel, M.A. ORCID logoORCID: https://orcid.org/0000-0002-1372-6324 and Riches, P.E. ORCID logoORCID: https://orcid.org/0000-0002-7708-4607;
CORE (COnnecting REpositories)