Finger Pad Viscoelastic Response to Shear Load

Uniaxial human skin viscoelasticity has been demonstrated in vitro (Pan et al., 1998). Although some have experimentally measured in vivo finger pad viscoleasticity under normal compression (e.g. Jindirch et al., 2003), none have measured its response to shear load. Knowledge of the viscoelastic properties of the finger pad is important for understanding dynamic finger force coordination during manipulation. While finite element models (FEM) of the finger pad have been developed for dynamic loading studies (e.g. Wu et al., 2002; 2003), these models have not been validated using experimental data. The purpose of the current study was to measure the viscoelastic response of the finger pad to tangential shear load, and to compare the data with results of FEM simulations. The index, middle, ring, and little fingers of the right hand of eight subjects (age: 26.0 ± 2.3 years, height: 175.1 ± 9.5 cm, body mass: 69.3 ± 8.3 kg) were individually clamped at their distal interphalangeal joints in a custom-built device that allowed for compression of the finger pad against a multi-axis force transducer (ATI, North Carolina, USA). The transducer was topped with 100-grit sandpaper to prevent slip; the coefficient of static friction between the finger and the sandpaper was measured to be approximately 1.4. Three different levels of compressive normal force (ranging from 1 to 5 N) were applied to each finger of each subject. Subsequent tangential displacements in both the medial and lateral directions were applied in steps of 0.6 mm (to an accuracy of 0.01 mm) to the force transducer by a micrometer positioning slide (Techno, Inc., NY, USA). Since the micrometer slide was adjusted manually, the loading rate was not precisely controlled (the loading rate was estimated to be 0.6 mm/s). Thus only force relaxation was analyzed (using nonlinear regression techniques) — this was considered sufficient to compare to FEM results. The force response after full relaxation was also considered as a long-term ‘stiffness’ response. The experimental results were compared with two FEM from the literature: Wu et al. (2002) and Wu et al. (2003) that were reconstructed using ABAQUS 6.2 (ABAQUS Inc.; Pawtucket, RI, USA). Both models were 2-D plain strain models with hard normal and rough tangential contact. Both incorporated linearly elastic bone and nail components and had geometry of the average male index finger. The soft tissue of the former FEM was modeled en masse as hyperelastic skin. The soft tissue of latter model incorporated a thin skin layer with biphasic subcutaneous tissue (see the original articles for material parameters, constitutive equations, etc.). The experimental data showed tangential force relaxation on the order of 40% over an average time period of 11.2 seconds. A logarithmic function applied to the rate of change of the force relaxation successfully reproduced the relaxation curves. The long-term ‘stiffness’ was found to be linearly related to the applied shearing displacement magnitude. ANOVA found that both stiffness and the relaxation parameters were different for each finger (p<0.01). These data were also dependent on the direction of the shear load (p<0.01). While the ABAQUS models have been constructed and qualitative agreement has been found between the modeled and experimental results, a quantitative comparison has not yet been performed. The substantial relaxation and inter-finger differences may have important implications to studies of force coordination among redundant fingers. The agreement between experimental data and predictions of FEM confirm the usefulness of the FEM for soft tissue biomechanics studies.