Measurement of the mechanical properties of the skin using the suction test.

INTRODUCTION

The suction test is commonly used to study the mechanical properties of human skin in vivo. The unevenness of the stress fields complicates obtaining the intrinsic mechanical parameters of the skin in vivo because the values of the local stresses and deformations cannot be calculated directly from the displacements and forces applied by the test apparatus. In general, users only take into account the negative pressure applied and the elevation of the dome of skin drawn up in order to deduce the properties of the skin. This method has the major disadvantage of being dependent on the experimental conditions used: in particular, the size of the suction cup and the negative pressure applied. Here, we propose a full mechanical study of the test to provide rigorous results. We compare the frequently used geometric method (making the thin plate hypothesis), Timoshenko's method (which can take greater plate thicknesses into account) and finally various results obtained by the finite elements (FE) technique.

METHODS

The suction test was modelled by FE with large displacements and large deformations both for orthotropic and isotropic plates. The results obtained in the elastic domain for various values of Young's modulus and of applied negative pressure were used as references and were compared with methods using analytical relationships.

RESULTS

The geometric method generally used in the interpretation of suction tests gives results, which in certain configurations, are very different from those obtained by FE. The method of Timoshenko is suited to thick plates 'in contact' or embedded round the edge, the elevation of the dome and the tension and flexion stresss are analytically accessible through relationships involving four constants that are dependent on the limit conditions. Comparison with the FE results enabled the optimisation of the coefficients to adapt the relationships to the particular conditions of the suction trials.

CONCLUSION

We showed the limits of the geometrical method and proposed a solution, which while remaining simple to use, gives results that are closer to reality both for the calculation of the modulus and for the determination of the state of the stresses obtained.