Session: 15-01-01: ASME International Undergraduate Research and Design Exposition

Paper Number: 99683

99683 - High Strain Characterization of Simulated Brain Tissue Using Dic 

When a head, with or without protective gear such as a helmet, is impacted by a traumatic force, the resulting pressure waves transmit to the interior of the brain. A major part of the brain’s interior is built on roughly 100 billion neuron cells, 85 billion glial cells, and intra/extra-cellular matrices. To understand the helmet to head load transfer mechanism and the connection between TBI with neuron damage and neurodegeneration, it is first necessary to understand and analyze the structural behavior of brain tissue as a function of different strains and strain rates. This is because brain tissue is made of rate-dependent viscoelastic materials. Biologically equivalent brain matter, for example, is a soft, high deformation material with low mechanical impedance that exhibits a viscoelastic response when subjected to various strain conditions. High-strain-rate characterization of large-deformation materials is commonly conducted with a modified Split-Hopkinson pressure bar (SHPB). Theory associated with this technique requires the assumption of a uniaxial, homogeneous stress state within the material. Soft or mechanically low impedance materials such as simulated brain matter experience more complex stress states under dynamic loading. Strain results thus have inherent inaccuracies due to this assumption. Digital image correlation (DIC) is a no-contact method of quantifying strains and object motion. DIC’s capability to assess full-field deformations may be used to relax the uniaxial, homogeneous strain assumption used in SHPB calculations. Application of local, subset-based algorithms to highdeformation materials, however, is complicated by decorrelation and subsequent loss of data points due to drastic changes in subset shape. Incremental correlation that tracks subsets between each consecutive image may be used post-decorrelation onset to compensate, but this method is generally undesirable due to the high accumulative error associated with it. A custom manufactured Split-Hopkinson pressure bar with hollow aluminum bars is used to apply compressive impact loadings on simulated gray and white brain matter. Strains are measured with a half-Wheatstone bridge circuit that receives inputs from strain gauges located on the incident and transmission bars. Two Shimadzu HPV-X2 ultra-high-speed cameras are used to record the dynamic response of the material. The images are imported into VIC-3D software to perform DIC and calculate all three Cartesian components of linear strain, forming a three-dimensional contour that displays the non-homogeneity of the soft material’s behavior. Axial strain from the SHPB and DIC are compared to assess the pressure bar method’s accuracy, using only the data prior to decorrelation. Additionally, a parametric study is conducted to form a method of reducing error in subset based DIC results for soft materials. SHPB specimens are prepared with speckle patterns of various shapes, distributions, and application methods then analyzed via subset reference image tracking. Start of decorrelation is compared between each pattern and application to determine the optimal speckle technique.

Presenting Author: Lauren Hutchison The University of Texas at Arlington

Presenting Author Biography: Lauren Hutchison is an undergraduate research assistant at the Multiscale Mechanics and Physics Laboratory in the Mechanical and Aerospace Engineering Department of the University of Texas at Arlington. Her research is related to low and high-strain-rate characterization of biologically equivalent and additively manufactured advanced materials using digital image correlation.

Authors:

Lauren Hutchison The University of Texas at Arlington
Jacob Navarro The University of Texas at Arlington
Ashfaq Adnan The University of Texas at Arlington