A mechanics model is developed for the encapsulated piezoelectric thin-film actuators/sensors system imperfectly bonded to the human skin to simultaneously determine the Young's moduli of the epidermis and dermis as well as the thickness of epidermis.

## Introduction

The overall mechanical properties of the human skin depend mainly on the nature and organization of the dermal collagen and elastic fiber network, water, and proteins [1]. Studies of the mechanical properties of skin, such as Young's modulus, can provide an assessment in diagnosis and rehabilitation of dermal diseases. The complex stratified structure of the human skin adds many restrictions in the experiments for measuring the Young's modulus. Conventional methods including suction [1,2], indentation [3], traction [4], torsion [5], and wave propagation [6] provide useful insight into the averaged mechanical behavior of human skin but are problematic in terms of extracting the Young's moduli of the epidermis and dermis.

Stretchable and flexible electronics have been developed to measure the electrophysiological signals and mechanical properties of the human body [7–15]. Recently, Dagdeviren et al. [16] presented a microscale, conformal piezoelectric system for measuring the modulus of the human epidermis, which provides soft and reversible contact with the underlying human skin surface. A mechanics model is developed in this note to extend the experimental design of Dagdeviren et al. [16] for simultaneous determination of the Young's moduli of the epidermis and dermis when the system is not perfectly bonded to the human skin.

### Model Description and Analytical Solutions.

As shown in Fig. 1, the human skin is modeled as a two-layered structure composed of an epidermis layer and a dermis layer. Each layer is assumed to be linear elastic and isotropic. The dermis layer (thickness ∼ 1 mm [17]) is much thicker than the epidermis layer (thickness ∼ 0.1 mm [17]) and is therefore modeled as a semi-infinite solid. The thickness of the epidermis layer is denoted as *H*. The Young's moduli of the epidermis and dermis layers are *E*_{epidermis} and *E*_{dermis}, respectively, and their Poisson's ratios are 0.5 because of their incompressibility.

The encapsulated piezoelectric sensors and actuators, if not perfectly bonded to the human skin, may slip along the skin surface, which may significantly reduce the interfacial shear stress. The deformation of the actuators/sensors is then dominated by the normal stress, not shear stress, at the interface. This leads to the change in the actuator/sensor thickness, which can be modeled by a pair of edge dislocations embedded in the surrounding media for each actuator/sensor.

The piezoelectric, dielectric, and elastic constants of the actuators and sensors normal to the skin are *e*_{33}, *k*_{33}, and *c*_{33}, respectively. The thick encapsulation layer is also modeled as a semi-infinite, linear-elastic solid with the Young's modulus *E*_{encap} and Poisson's ratio of 0.5 [10,16]. Figure 1 shows an actuator and a sensor (made of identical materials), with length 2*b* and spacing *l*. A two-dimensional model for plane-strain deformation is adopted for Fig. 1.

where $A=(c33/e33)\u22c5(c11e33\u2212c13e31)/(c11c33\u2212c132)$ is given in terms of the piezoelectric constants *e _{ij}* and elastic constants

*c*. The thickness increase induces deformation in the surrounding encapsulation, epidermis, and dermis, though their Young's moduli (∼100 kPa [16]) are many orders of magnitude smaller than that of the actuator (∼100 GPa [16]). Consequently, the decrease of the actuator thickness due to the constraint of the surrounding media is negligible as compared to $\Delta $ [10,16].

_{ij}The thickness of the actuators and sensors (∼5 *μ*m [16]) is much smaller than that of the epidermis (and dermis and encapsulation). Therefore, Eq. (1) can be modeled as a pair of edge dislocations, with the Burgers vector $\Delta $ and $\u2212\Delta $, separated by the length 2*b* of the actuator on the encapsulation/epidermis interface for a finite-thickness epidermis sandwiched by semi-infinite encapsulation and dermis. The deformation and stress in the encapsulation, epidermis, and dermis are obtained analytically, similar to the studies of dislocations in layered media [18,19].

*b*of the sensor (at the spacing

*l*from the actuator) is obtained analytically by [18–20]

$f1$ and $f2$ are shown versus *l*/*b* in Fig. 2 for different values of *H*/*b*; they approach zero for $H/b\u226b1$ (small sensor 2*b* as compared to the thickness *H* of epidermis).

where $S=1+(e31/e33)\u22c5(c33e31\u2212c13e33)/(c11e33\u2212c13e31)+(k33/e33)\u22c5(c11c33\u2212c132)/(c11e33\u2212c13e31)$ is given in terms of the piezoelectric constants *e _{ij}*, elastic constants

*c*, and dielectric constants

_{ij}*k*, and

_{ij}*h*

_{piezo}is the thickness of the piezoelectric layer in the sensor.

This ratio for a system poorly bonded to the skin is expected to be much smaller than that for a perfectly bonded case. Equation (6) is explored in the following to develop a strategy for determining *E*_{epidermis}, *E*_{dermis}, and *H* in experiments.

### Determination of the Young's Modulus of the Epidermis.

For a known Young's modulus *E*_{encap} of the encapsulation, *E*_{epidermis} can be determined from the experiment for small actuators and sensors, independent of *E*_{dermis} and *H*.

### Determination of the Young's Modulus of the Dermis.

*b*but different actuator–sensor spacing

*l*

_{1}and

*l*

_{2}, Eq. (6) becomes

where $(Uoutput/Uinput)i$ is obtained from the experiments for sensor *i*, $(f1)i$ and $(f2)i$ are $f1$ and $f2$ in Eq. (3) for spacing *l _{i}*, and

*E*

_{epidermis}obtained from Eq. (7) also gives $\alpha encap$. Equation (8) constitutes two equations for $\alpha dermis$ and

*H*, which can be solved numerically, and $\alpha dermis$ then gives

*E*

_{dermis}.

The Young's moduli of the epidermis and dermis can be obtained simultaneously by applying a system consisting of actuators and sensors with small lengths and spacing (2*b*_{small}, *l*_{small}) and ones (2*b*, *l*) comparable to the thickness of epidermis.

## Acknowledgment

J.Y. acknowledges the support from the National Natural Science Foundation of China (Grant No. 11402133). X.F. acknowledges the support from the National Basic Research Program of China (Grant No. 2015CB351900) and the National Natural Science Foundation of China (Grant No. 11320101001). Y.H. acknowledges the support from the U.S. National Science Foundation (CMMI-1400169 and CMMI-1300846) and from the U.S. National Institutes of Health (Grant No. R01EB019337).