Comparison of elastic and tactile behavior of human skin and elastomeric materials through tribological tests.

INTRODUCTION

Ageing is a physiological mechanism associated with changes in skin morphology and mechanical properties, including dry state, and anisotropy of the skin relief. Moreover, cosmetic products induce tactile changes of the skin surface [1]. Quality of the skin such as suppleness, smoothness, roughness, greasiness, or dryness is the combined result of the mechanical and frictional properties of the skin. A variety of techniques have been employed to evaluate skin friction [2, 3].

Since the human skin has viscoelastic behavior [4] and to avoid technical difficulties encountered when determining in vivo human skin behavior, it will be interesting to simulate its elastic and tactile properties through elastomeric materials.

In fact, the elastomeric materials have attractive properties. In particular, they have shown viscoelastic behavior [5]. They can be easily deformed under a relatively weak load and often had a weak residual deformation [6, 7].

The aim of our study was to determine quantitatively the elastic and tactile properties of the human skin, and then compare these tribological properties with those obtained for two different elastomeric materials.

EXPERIMENTAL

Materials

Human Skin. The study group consisted of three women (24, 25, and 26 years old). All participants were in good health. They were instructed not to apply any skin care product a day before testing. All measurements were performed after the subjects have been physically inactive for at least 15 min. Measurements were made on the volar aspect of their forearm.

Thermoplastic Polyurethane Elastomer. Thermoplastic polyurethane elastomer (TPU) belongs to thermoplastic elastomers. The studied material is (PEARLTHANE D11T92EM), which is a polycaprolactone-copolyester based TPU, supplied in form of translucent, colorless or slightly yellowish pellets. This material can be extruded and injection-molded.

Silastic. The second selected elastomer is called silastic "SIL." It is used as replica material in many applications such as the practice of dentistry. This material was also chosen for its homogenous roughness. Moreover, its elastic property tended to simulate the physical in vivo skin properties with an indentation in the material during the test in the same order as in the skin.

Methods

Indentation Test. To characterize surface material's mechanical properties, the equipment used in this study employs a fully automated tribological device (Fig. 1) based on a static contact between a spherical indenter (100Cr6 Steel) of 6.35 mm radius and the flat specimen (skin, TPU, or SIL). The tested elastomeric samples have a cylindrical shape of a 20-mm diameter and 18-mm in depth. Prior to each test, the indenter was washed by acetone and isopropyl alcohol. The indentation test allows the following of the normal load ([F.sub.n]) evolution versus the penetration depth in real time. The load domain of [F.sub.n] is ranged from 20 to 80 mN. For skin indentation, a specific device was used to immobilize the human forearm.

[FIGURE 1 OMITTED]

Friction Test. The same tribometer was used to carry out friction tests. Spherical indenter (100Cr6 Steel) of 6.35 mm radius, tangents the specimen surface, after indentation, it carries out a scratch at a constant normal load. The sensor system measures the normal and tangential forces in real time. For these tests, we can control the normal force [F.sub.n], the friction speed (V), the loading speed ([V.sub.p]), and the sliding distance (L). Friction tests were carried out at a loading speed ranging from 50 to 200 mN/s and the friction speed V = 50 [micro]m/s under normal load ([F.sub.n]) ranging from 20 to 80 mN. The sliding distance was equal to L = 5000 [micro]m.

RESULTS AND DISCUSSION

Indentation Tests

Influence of the Applied Normal Load. In Fig. 2, indentation curves (variation of normal load as function of penetration depth) of human skin, TPU, and SIL are illustrated using three maximal normal loads [F.sub.n] = 20, 40, and 80 mN and a fixed loading speed ([V.sub.p] = 50 mN/s).

[FIGURE 2 OMITTED]

The analysis of the indentation curves allows the determination of the mechanical properties of the material's surface such as Young's modulus (E), viscoelastic energy ([W.sub.v]), adhesion force ([F.sub.ad]),... [8, 9].

Variation of these mechanical properties versus the maximal normal load is proposed in Fig. 3. It was found that The Young's modulus is constant as function of normal load for all tested materials. However, TPU shows higher stiffness than the SIL and the skin (Fig. 3a).

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

Furthermore, viscoelastic energy increases as function of the normal load. However, the ratio of viscoelatic energy relative to the total energy is almost constant for all tested materials. Indeed, TPU has shown more important viscoelastic energy compared with the one presented by SIL and skin (Fig. 3b).

It was also noted that adhesion force is more developed in SIL case compared with TPU and skin cases under indentation. For TPU, [F.sub.ad] is almost equal to zero. Whereas, SIL has shown significant adhesive force, which is independent of the normal load (Fig. 3c).

Influence of the Loading Speed [V.sub.p]. The influence of the loading speed ([V.sub.p] = 50, 100, and 200 mN/s) is investigated in Fig. 4 from which the mechanical properties of the skin, TPU, and SIL surface as Vp varies are estimated.

Variation of these mechanical properties versus loading speed is proposed in Fig. 5, we noticed that under constant normal load, Young's modulus of the TPU and SIL increases as function of loading speed [V.sub.p]. However, no significant variation was noted for skin case. Such result can be explained firstly through the viscoelastic behavior of these materials and secondly referring to lipid layer in the human skin [10]. In fact, viscoelastic energy decreases versus loading speed [V.sub.p] in all cases. This fact is more obvious in TPU case.

However, only SIL has shown significant adhesion force. Both TPU and skin have demonstrated absence of adhesion phenomenon. Moreover, no influence of loading speed was found on this phenomenon.

As first conclusion we note that all materials have shown viscoelastic behavior under indentation. Under the considered test conditions, elastic behavior prevails for the skin case. The same remarks can be made for the elastomeric material but with less obviousness depending on the nature of the elastomer. So, under the considered indentation tests, SIL has comparable rigidity as the skin; whereas, TPU has comparable adhesive response as the skin.

[FIGURE 5 OMITTED]

Numerical Analysis

In this section we try to quantify the critical normal load corresponding to the limit of elastic model when describing indentation test of the skin, SIL, and TPU. Finite element method (ABAQUS) was used to simulate indentation tests (only loading step).

Figure 6a shows the used geometry and boundary conditions. Bilinear axisymmetric quadratic meshes were used (2601 nodes, 2451 elements) as illustrated in Fig. 6b. Furthermore, an elastic behavior was considered for the flat specimen, so only Young's modulus and Poisson coefficient variables have been considered. The spherical indenter is defined as a rigid material.

The predicted indentation curve describes only the loading step (ABAQUS). It fits well the experimental one until a critical normal load "[P.sub.p]" (Fig. 7). This critical value is considered to be the limit of the validation of the elastic model.

For the same loading speed ([V.sub.p] = 50 mN/s), this critical value is lower in the TPU case compared to SIL and skin cases. In fact, Fig. 7a shows that skin has an elastic behavior until [P.sub.p] = 25 mN, whereas SIL has an elastic behavior until [P.sub.p] = 15 mN (Fig. 7b) and TPU has an elastic behavior until [P.sub.p] = 10 mN (Fig. 7c).

Validation of such critical forces is possible by performing indentation tests at different normal loads. For [F.sub.n] < [P.sub.p] there is no hysterisis and so elastic behavior is confirmed. An example was illustrated in Fig. 2a, corresponding to skin case (no hysterisis for 20 mN and there is hysterisis apparition for 40 mN).

[FIGURE 6 OMITTED]

Friction Tests

Influence of the Applied Normal Load. Friction tests are performed with a constant sliding speed (V = 50 [micor]m/s) and a total sliding distance L = 5 mm.

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

Figure 8 shows skin, SIL, and TPU friction behavior under loading speed [V.sub.p] = 50 mN/s.

It was found that tangential forces increase when normal force increases. Indeed, three distinct friction phases were distinguished:

  Tangential force increases linearly to reach a maximum, which
    corresponds to the static limit friction force. The slope of the
    linear curve corresponds to the lateral contact stiffness.
  The slope change in the transition regime between the static or
    partial slip and dynamic friction or gross slip corresponds to the
    second phase.
  During the dynamic friction phase (third phase), an intermittent
    "stick-slip" motion can be more or less important depending on the
    loading conditions. Table 1 summarizes the corresponding friction
    coefficient for each material as function of normal load. It was
    found that friction coefficient decreased when normal load
    increased. Moreover, the measured values were particularly high.
    These results can be explained by the adhesive behavior of this
    material. Hence, SIL shows the higher value of the friction
    coefficient and the sharp decrease of this coefficient when the
    normal load increases [11, 12]. Since the adhesive component of the
    contact is extremely important, influence of loading speed [V.sub.p]
    will be studied in the following.

Influence of the Loading Speed [V.sub.p]. The second friction campaign was performed under a constant normal load [F.sub.n] = 20 mN and a constant sliding speed V = 50 [mu]m/s.

The same tendency was observed in Fig. 9. However, there is no significant variation of tangential force when the loading speed [V.sub.p] increases, except for SIL material when rubbed using [V.sub.p] = 200 mN/s.

[FIGURE 9 OMITTED]

It seems that the elasticity prevails and no viscoelastic effect was noted in the friction tests. It means that contact area has not changed or does not have a significant influence on the tangential force.

Tactile properties of skin are more similar to those of TPU rather than the SIL ones (under the considered conditions).

Comments and Discussion. The skin seems to be not sensitive to the loading speed either in indentation or in friction. These results proved that the elastic behavior prevails under the considered test conditions.

However, SIL and TPU behaviors were very sensitive to normal load variation: when [F.sub.n] increases, [F.sub.t] increases but friction coefficient ([F.sub.t]/[F.sub.n]) decreases.

Furthermore, for all tested materials, transition distance in the friction curves is not influenced by the [V.sub.p] variation. Whereas, contact stiffness decreases when the loading speed increases. Such results can be confirmed with indentation tests. In fact, the penetration depth decreases when [V.sub.p] increases, hence contact area decreases and so contact stiffness decreases [13].

Moreover, TPU and SIL have shown similar contact stiffness (Fig. 10) higher than the skin contact stiffness. In the same way, transition-sliding distance (PS/GS) in the skin case is higher than the TPU and the SIL ones.

All results prove that skin has a dominant elastic behavior either under indentation or in friction tests. However, SIL has lower viscoelastic behavior than the TPU behavior.

CONCLUSION

The current article describes static and frictional contact tests carried out with spherical indenter on both human skin surface and elastomeric materials. This tribological study is performed to analyze the elasticity and tactile properties of the human skin, and therefore to evaluate reliability of human skin behavior simulation through the considered elastomeric materials.

[FIGURE 10 OMITTED]

For indentation tests, the following conclusions can be made:

i. Indentation tests have confirmed the viscoelastic behavior of skin, SIL, and TPU materials. Such behavior is more remarkable in the case of the TPU.

ii. Skin behavior depends on normal load because of the presence of lipid layer. However, loading speed has a weak influence on tribological properties of the skin (E, [F.sub.ad], [W.sub.v],...). This is a proof that elastic behavior is dominant under the studied conditions.

iii. Numerical analyses (FEM) of indentation test (loading step) have been performed. It was found that numerical and experimental curves coincided until reaching a critical normal load. This value limits validation of the elastic model. Under the same loading speed, critical value relative to TPU is lower than the SIL and the skin values. Hence, dominance of elastic behavior of the skin was confirmed compared to dominance of viscoelastic behavior of the TPU.

For friction tests, using steel indenter, the following conclusion can be made:

i. High friction values of SIL were explained by the high adhesive component in the contact.

ii. Concerning contact stiffness TPU and SIL have same values. However it's too much higher than the one relative to skin.

iii. PS/GS transition relative to skin was occurred for greater sliding distance comparing to SIL and TPU.

Both friction and indentation tests have proved that elastic behavior of the skin prevails (under the studied conditions). However, SIL has shown a lower viscoelastic behavior than the TPU. For this reason it will be more interesting to simulate elastic and tactile behavior of the human skin under the studied conditions.

REFERENCES

1. J. Asserin, H. Zahouani, Ph. Humbert, V. Couturaud, and D. Mougin, Colloids Surf. B: Biointerfaces, 19, 1 (2000).

2. S. Cosmaish, Acta. Derm. Venereol., 53, 455 (1973).

3. W.A. Gerrard, Bio. Eng. Skin, 3, 123 (1987).

4. P. Agach, Physiologie de la peau et exploitations fonctionnelles cutanees, Editions medicales internationales (2000).

5. P. Saad, F. Thouverez, J.P. Laine, and L. Jezequel, Mec. Indus., 4(2), 133 (2003).

6. P. Brocard, J.P. Chereul, and M. Durond, Les 4 pages des statiques industrielles, 176 (2003).

7. P. Martinon, Techniques de l'ingenieur, K380 (1996).

8. W.C. Oliver and G.M. Pharr, J. Mater. Res., 7, 6 (1992).

9. J.L. Loubet, M. Bauer, A. Tonck, S. Bee, and B. Gauthier-Manuel, in Mechanical Properties and Deformation Behavior of Materials Having Ultra-Fine Microstructures, M. Nastasi, editor, Kluwer, Dordrecht, MA, 1 (1993).

10. C. Pailler-Mattei and H. Zahouani, Tribol. Int., 39, 1 (2006).

11. B.J. Briscoe and D. Tabor, Br. Polym. J., 3 (1978).

12. B.J. Briscoe, Philos Mag A, 43, 3 (1981).

13. K.L. Johnson, Contact Mechanics, University of Cambridge, Cambridge (2001).

Khaled Elleuch, Riadh Elleuch

Laboratoire des Systemes Electro-mecanique (LASEM), Ecole Nationale d'lngenieurs de Sfax, Tunisie

Hassan Zahouani

Laboratoire de Tribologie et Dynamique des Systemes (LTDS), UMR CNRS 5513, Ecole Centrale de Lyon, France

Correspondence to: Khaled Elleuch; e-mail: [e-mail]

TABLE 1. Dynamic friction coefficient.

[F.sub.n] (mN)  Skin  SIL   TPU
20              1.62  3.25  1.90
40              1.70  2.62  1.47
80              1.56  2.07  1.42