Colloids and Surfaces B: Biointerfaces 116 (2014) 734–738
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Tribological evaluation of porcine skin Huaping Xiao a , Nethika Ariyasinghe b , Xingliang He a , Hong Liang a,∗ a b
Department of Mechanical Engineering, Texas A&M University, TX 77840, United States Department of Biomedical Engineering, University of Southern California, CA 90089, United States
a r t i c l e
i n f o
Article history: Received 15 June 2013 Received in revised form 11 November 2013 Accepted 3 December 2013 Available online 11 December 2013 Keywords: Coefficient of friction Pig skin Tribological properties Bearing ratio
a b s t r a c t This research studies the effects of external parameters on the friction of porcine skin. A tribometer was used to evaluate the frictional behavior of the same. The effects of DI water and body oil on porcine skin against steel and glass balls were evaluated in terms of coefficient of friction (COF). The COF dropped rapidly when DI water/body oil was introduced into the sliding system and remained stable when the volume of the liquid exceeded a certain value. The COF increased with increasing sliding speed under dry conditions and decreased in wet. Under an increasing normal force, the COF decreased regardless of the presence of liquid. The ratio of the real contact area to the nominal contact area of the skin with the steel/glass ball was found to increase with a power law as the applied force was increased. These results reveal basic tribological properties of the skin in contact with a hard slider. These properties could be used as reference for the design and development of artificial skin in prosthetic applications. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Prosthetics help people who have lost part of their body to lead a more normal life. Since the middle of last century, research in prosthetics has led the creation of various devices [1]. The interface between the human body and prosthetics has been studied for proper integration. Steege et al. established a finite-element (FE) model to describe the interface between a residual lower-limb and a prosthetic device [2]. The FE method was then widely adopted to model the residual limb–prosthesis interface [3,4]. It was found that in order to improve the efficiency of load transfer between the residual limb and the prosthetic device, the geometry of the socket should be modified according to the residual limb rather than using a replica. In addition to the modification of the interface, spherical and cylindrical wrapping algorithms [5] were developed to improve the geometry of prosthetics. Proper algorithm for the prosthetics is another key factor for sophisticated design of flexible prosthetics. Even though current prosthetics can visually replace a missing limb and may be able to replicate some functions of a human hand, they cannot replicate all the functions of a human hand. One reason is that current prosthetics are not integrated with artificial skins. The skin plays a crucial role in the hand’s ability to grasp and manipulate objects [6]. For better mimicking human’s behavior, artificial skin coated metal framework is a promising option. Because a person’s grip on an object is related to the frictional force between the hand and the object, the tribological properties of human skin
∗ Corresponding author. Tel.: +1 9798622623. E-mail address: [email protected] (H. Liang). 0927-7765/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.colsurfb.2013.12.006
must be studied to improve the prosthetic implants and develop artificial skin. Seo et al. [7] showed that surface friction affects the force that a person needs to grip an object. Under the same normal force, higher friction leads to greater friction force and a tighter grip. Any factor that changes the frictional properties between the object and the skin should affect the force of the grip. Due to its complex structure, the skin’s contact behavior is affected by various factors. The impacts of the surface roughness, the superficial sebum, the skin moisture, the relative velocity, and the applied normal force on tribological properties of skin have been studied [8–20]. Pailler-Mattei et al. [14] showed that the influence of the skin lipid film on the adhesion between the human skin and the steel indenter could be attributed to the capillary effect. The capillary effect is caused by the liquid capillary bridge between two surfaces without direct contact. Most of the studies have been done in vivo on human skin. However, the artificial skins in real application are separate systems from the human body and are tested ex vivo. Result for artificial skin is different from real human skin and there is neither clear understanding nor effective evaluation of artificial ones. To close this gap, further understanding of skin’s function, ex vivo evaluation of the real skin is needed. In the present research we propose an ex vivo approach in investigating a real skin, i.e., pig skin, in order to understand its behavior. In the ideal case, the ex vivo study needs to be done on human skin. The reason we used pig skins is that it is very difficult to obtain the specimen of human skin. The aim of this study is to understand the frictional behavior of porcine skin in various environments and conditions. The effects of deionized (DI) water and body oil (Neutrogena light sesame formula) on the coefficient of friction (COF) was evaluated. A contact model was developed in order to explain the principles of friction.
H. Xiao et al. / Colloids and Surfaces B: Biointerfaces 116 (2014) 734–738
Fig. 1. Schematic expression of the tribo-test.
2. Materials and methods
735
Fig. 2. Values of the COF change with rotating speed at the load of 0.1 N. (The hollow shapes stand for the steel-skin pair and the solid shapes are for the glass–skin pair. The circles, triangles and diamonds represent the conditions of dry contact, contact lubricated by water and contact lubricated by body oil, respectively. The presented COF is the mean value of 3 measurements in the glass–oil case, 10 measurements in the steel-dry case and 6 measurements in all the other cases.)
2.1. Materials Pig legs were bought from a local grocery store (HEB). The porcine skins were peeled off from the foreleg of the pig. The skins were washed with liquid soap and then ultrasonically cleaned in DI water for 5 min. The samples were left in open air to dry until they did not leave traces of water on the Kimwipes placed on top. Steel balls and glass balls with a diameter of 6.5 mm were used as the counter contact. Glass and steel were chosen because they are the most common rigid materials people will touch in daily life. Before each test, the rigid balls were ultrasonically cleaned in acetone and then in ethanol for 5 min each. 2.2. Tribological experiments The skins were cut into rectangles with a length of 60 mm and a width of 20 mm for tribotests. The measured thickness is from ∼1.5 mm to ∼2 mm. The surface roughness of the steel ball and glass ball were measured using a Newview 600 interferometer (Zygo Corp., USA). Plasticine was pressed against the pig skin to make a reversed replica of the skin. After drying up, the surface roughness of the replica was measured using a TR200 (Time Group Inc., USA) roughness tester. 5 measurements were taken for each sample. The measured surface roughness (Ra) is 5.814 ± 0.699 m for pig skin, 0.286 ± 0.035 m for glass ball, and 0.088 ± 0.009 m for steel ball. The COF between the skin and the steel ball/glass ball was measured using a pin-on-disk tribometer (CSM Instruments, Switzerland). The skins were clipped onto a microscope slide using binder clips. This slide was super glued to the rotational stage of the tribometer. The steel or glass ball was fixed to the steel rod of the tribometer as shown in Fig. 1. There is a piezoelectric sensor inside the supporting arm. When the normal load was applied, the resulting frictional force produced a torque to the supporting arm. This torque was recorded by the sensor and was converted to frictional force and COF. The COF under different conditions was measured. The sliding speed was increased in increments of 10 rpm from 10 rpm to 60 rpm with a radius of 3.5 mm, and the applied load was increased from 0.05 N to 0.2 N. To determine the impact of liquid on the COF, 1–15 L DI water and 1–6 L body oil were introduced to the tested area on skin. The change of COF with variation of sliding speeds and applied loads was measured under dry conditions first. The COF at different liquid volumes was measured while the sliding speed was set at 40 rpm under a load of 0.1 N. After the liquid reached to the maximum volume, the same range of sliding speeds and applied loads were tested again as under dry
conditions. The number of rotation cycles for each test is 10 and each test was repeated two to three times with the same sample under the same conditions. 2.3. Measurement of the contact area of the skin with glass The skins were cut into squares with a length and width of 10 mm. A glass slide was placed on top of the skin. An image of the contact area between the skin and glass slide was recorded using a KEYENCE VHX optical microscope (KEYENCE Corp., Japan) through the glass slide. 4 spots were measured in this study. The real contact area of the skin to the glass can be distinguished because of the contrast between the areas of the skin in contact with the glass and otherwise. The real contact area was then calculated using ImageJ program. 3. Results and discussion 3.1. Effects of sliding speed on COF The values of the COF at different sliding speeds for both dry and wet conditions are shown in Fig. 2. The presented COFs are the mean values of 3 measurements in the glass–oil case, 10 measurements in the steel-dry case and 6 measurements in all the other cases. The error bars show the standard deviation of those measurements. Under dry conditions there is a notable trend of going up for the COF when the sliding speeds are changed from 10 rpm to 60 rpm. After DI water or body oil is introduced into the contact, different trend is observed. The COF slightly decreases as the sliding speeds are increased. The plots shown in Fig. 2 reveal that the COF under dry conditions is always higher than that in wet. These results demonstrate that both DI water and body oil lubricate the skin–steel/glass contacts. In addition, lower COF with lubrication by body oil than by DI water indicates that body oil has better lubricating performance in this study. No conclusive remarks can be made to determine whether steel or glass has lower COF on the pig skin. Overall, the change of the COF under dry conditions is greater than that under wet conditions. This means that the speed plays a more significant role on COF in dry contacts. 3.2. Effects of applied force on COF The change of COF with variation of applied loads is displayed in Fig. 3. The COF consistently reduces with increasing load in all cases.
736
H. Xiao et al. / Colloids and Surfaces B: Biointerfaces 116 (2014) 734–738
Fig. 3. Change of the COF with applied load at the sliding speed of 40 rpm. (The shapes have the same meaning as in Fig. 2. The presented COF is the mean value of 3 measurements in the glass–oil case, 10 measurements in the steel-dry case and 6 measurements in all the other cases.)
Under small loads, there is a significant decrease in COF. The trend of decreasing COF becomes less notable as the applied load increases. Fig. 3 shows that the COF decreases faster under dry conditions than in wet. Curve fitting was done using power law relation: Y = mXn . The results show that the values of n are the smallest under dry condition. The highest value of n is obtained under oil lubrication. Since n is less than 0, the COF drops faster with a smaller n when the load is increased. These results demonstrate that the COF is less sensitive to applied load under wet conditions. Similar to the phenomenon observed in Fig. 2, the results from Fig. 3 indicate that body oil lubricates the contacts better than DI water. But the difference fades as the loads increases.
Fig. 4. Change in the COF as (a) body oil (b) DI water is introduced into the area of contact (The presented COF is the mean value of 3 measurements in the glass–oil case and 6 measurements in all the other cases.)
3.3. Effects of the amount of liquid on friction Fig. 4 shows how the introduction of body oil and DI water affects the COF. The presented COF is the mean value of 3 measurements in the glass–oil case and the mean value of 6 measurements in all the other cases. The error bars show the standard deviation of those results. From Fig. 4(a) it can be seen that the COF greatly reduces when only 1 l of body oil is added to the contact. However, the COF does not change much as the amount of body oil is >1 l. At the glass–skin contact, the COF increases slightly as more body oil is introduced. A similar situation is observed when DI water is added into the contact, as shown in Fig. 4(b). The COF drops rapidly when DI water is added, but the reducing trend of COF gradually diminishes with the presence of more DI water. When the volume of water is >4 l, the COF becomes stable in both contacts. When the liquid is firstly introduced, effective lubrication between the skin surface and the other surface is established, and so the COF drops fast when a small amount of DI water or body oil is introduced into the contact. However, once the volume of liquid is higher than a certain number, full lubrication is established. The COF will not change much with addition of more lubricant. Thus, the COF remains stable when the volume of liquid is over a certain amount. All the contact surfaces involved in this study, pig skin, steel and glass, are more oleophilic than hydrophilic, which means that body oil has lower contact angle on these materials than DI water. Less volume of oil is required to cover a certain area. In this study, 1 l of body oil is enough to cover the sliding track to establish full lubrication whereas 4 l of DI water is needed.
contact area was measure under different loads. The photos of the contact area are displayed in Fig. 5. The dark spots are the real contact area. It is obvious that the dark spots gradually enlarge as the load is increased. The ratio of the dark area to the whole area (the nominal contact area) was calculated using ImageJ program. The
3.4. Variation of bearing ratio with load To investigate why load affects the COF in the manner shown above, the bearing ratio of the real contact area to the nominal
Fig. 5. Photos of contact area at spot 4 under different loads.
H. Xiao et al. / Colloids and Surfaces B: Biointerfaces 116 (2014) 734–738
737
Fig. 6. Increase of bearing ratio with applied load. Table 1 The calculated values of a and b and the standard error.
Spot 1 Spot 2 Spot 3 Spot 4
Value of a
Value of b
Standard error
1.044 0.293 0.146 0.933
0.305 0.981 0.685 0.806
a 0.6 ± 0.45 b 0.69 ± 0.29
variation of the bearing ratio with the applied load is represented in Fig. 6. The lines in Fig. 6 are fitted based on the power law: Y = aX
b
3.5. Contact model Skins are usually characterized as elastomers which have viscoelastic and nonlinear properties [21]. When a rigid ball is sliding on the pig skin with a normal force, there is a noticeable deformation of the skin at the front surface as shown in Fig. 7(a). At the back surface, the deformation reduces and recovers when the surfaces lose contact with each other. This scenario is consistent with the findings of Kwiatkowska et al.’s study [22]. Based on these effects, a two-term model is employed to describe the frictional force between the skin surface and the counter-surface. The total effective frictional force is a combination of the adhesion friction force, which is the dominant component, and the deformation friction force. Therefore, the effective frictional force can be expressed as [23]: (2)
where Feff is the effective frictional force, Fadh is the adhesion friction force, and Fdef is the friction force due to the deformation of the skin. The frictional coefficient can be calculated as: =
Fadh + Fdef N
where N is the normal force. Since the adhesion friction force governs the frictional force, the frictional coefficient can be approximated as: ≈
(1)
where Y is the contact area, X is the applied load, and a and b are the fitted parameters (Table 1). The difference between the fitted value and the measured value is small, indicating that the fitting curves are good approximations of the real trends. One possible explanation for the high variation is that the skin surface is not perfectly flat. When a slide is forced against the skin, some region of the skin sustains more loads and other region sustains less. As shown in the Fig. 5(a), the contact regions are not uniformly distributed. The bearing ratio inside the dashed line is definitely higher than that outside. When the external load is increased, the increase of real contact area for different region could be different. Thus, there is high variation between the four measurements. Measurement uncertainty is another reason for the observed variation.
Feff = Fadh + Fdef
Fig. 7. Schematic expression of skin-rigid ball contact (a) before lubrication (b) after lubrication. The small particles in (b) represent the liquid molecules. The contact between the skin and the rigid ball is partly separated by the liquid.
(3)
Fadh N
(4)
The adhesion friction force is generated from the repulsive and attractive forces that operate in a short range. It is proportional to the real contact area, Ar (Fadh ∝ Ar ). Based on the forces shown in Fig. 7, the relationship between the bearing ratio and the normal force is: R=
Ar = aN b A
(5)
A is the nominal contact area; Ar = AaN b
(6)
The frictional coefficient can be calculated by combining all the relationships shown above: ≈
Fadh Ar AaN b ∝ = = AaN b−1 N N N
(7)
Eq. (7) shows that the frictional coefficient is proportional to N b−1 . From Table 1, all the estimated values of b are less than 1. Therefore, the frictional coefficients are inversely related to the applied normal force, which is a possible explanation for the results shown in Fig. 3. The calculated values of (b − 1) are comparable with the values of m in Fig. 3. When the load is kept constant, the COF increases as the sliding speed increases in most of the dry contact cases shown in Fig. 2. Assuming that the real contact area stays constant under the same load, the adhesion friction force remains constant as the speed changes as well. The energy dissipation caused by the deformation and recovery of the skin, however, increases as the sliding speed increases. At higher speed, the sliding distance in the same time period is longer. This means more deformation and recovery of the skin happen in the same time period. The energy dissipation due to the deformation and recovery of skin increases as well. The increased energy consumption could lead to a higher COF. Under wet conditions, the lubrication effect of the liquid plays a remarkable role in contact area between the skin and the steel/glass ball as shown in Fig. 7(b). Portion of the contact region is separated by the lubricant. In these regions, the adhesion friction force greatly decreases leading to the decrease of COF. Assume that most of the
738
H. Xiao et al. / Colloids and Surfaces B: Biointerfaces 116 (2014) 734–738
contact points between the skin and the rigid ball are separated in the presence of adequate lubricant under high speed. Hydrodynamic lubrication could be established under these conditions. The influence of hydrodynamic lubrication is enhanced as the relative speed between the skin surface and the counter-surface is raised. Thus, the COF decreases gradually with increasing sliding speed under wet conditions as shown in Fig. 2. Most of the prosthetic devices in application are metal based. Previous study [24] showed that specially designed alternative materials (artificial skin) have better approximation to real skin than the metal based material. Artificial skin coated metal frameworks is an option for better prostheses. Two aspects involving the artificial skin need remarkable notice. First, wear of the artificial skin should be avoided. Second, the artificial skin could provide sufficient grabbing force for the user to handle or hold objects. From the point view of tribology, these two aspects are contradictory. To avoid the wear the COF needs to be reduced whereas greater COF is preferable to provide higher grabbing force. Based on the findings of this study, introduction of water and oil is an effective way to reduce the COF. So in the design of artificial skin, the material which is capable of capturing the moisture from the environment has advantage if lower COF is desired. On the contrary, the material with the ability of repelling liquid and forming larger real contact area is preferable in the cases that higher grabbing force has priority. 4. Conclusion The COF between pig skin and steel/glass ball decreases when DI water or body oil is introduced. However, the amount of reduction diminishes as more liquid is added to the contact, and then COF becomes stable. The COF is found to decrease with applied load following a power law. Under dry conditions, the COF increases with the speed, which attributes to the higher energy dissipation at higher speed. Under wet conditions, however, the COF decreases with speed because the lubrication is improved as speed is increased. The bearing ratio between the pig skin and glass slide is found to increase with loads following a power law. The curve fitting shows that the experimental results are consistent with the calculated value. This research could be beneficial in design and evaluation of artificial skin, as material surfaces and/or as a coating for prosthetic applications. Acknowledgement NA was supported by Dr. Hsien through REU program at TAMU.
References [1] G. Musser, Making them whole, Sci. Am. 298 (2008) 52–53. [2] J. Steege, D. Schnur, R. Vorhis, J. Rovick, Finite element analysis as a method of pressure prediction at the below-knee socket interface, in: Proceedings of RESNA 10th Annual Conference, California, 1987, pp. 814–816. [3] J. Brennan, D. Childress, Finite element and experimental investigation of above-knee amputee limb/prosthesis systems: a comparative study, ASME Adv. Bioeng. 20 (1991) 547–550. [4] M. Zhang, A.F.T. Mak, V. Roberts, Finite element modelling of a residual lowerlimb in a prosthetic socket: a survey of the development in the first decade, Med. Eng. Phys. 20 (1998) 360–373. [5] I.W. Charlton, G.R. Johnson, Application of spherical and cylindrical wrapping algorithms in a musculoskeletal model of the upper limb, J. Biomech. 34 (2001) 1209–1216. [6] C.A. Brohem, L.B. da Silva Cardeal, M. Tiago, M.S. Soengas, S.B. de Moraes Barros, S.S. Maria-Engler, Artificial skin in perspective: concepts and applications, Pigment Cell Melanoma Res. 24 (2011) 35–50. [7] N.J. Seo, J.K. Shim, A.K. Engel, L.R. Enders, Grip surface affects maximum pinch force, human factors, J. Human Factors Ergon. Soc. 53 (2011) 740–748. [8] T. André, V. Lévesque, V. Hayward, P. Lefèvre, J.L. Thonnard, Effect of skin hydration on the dynamics of fingertip gripping contact, J. R. Soc. Interface 8 (2011) 1574–1583. [9] M. Changizi, R. Weber, R. Kotecha, J. Palazzo, Are wet-induced wrinkled fingers primate rain treads, Brain Behav. Evol. 77 (2011) 286–290. [10] S. Derler, L.C. Gerhardt, Tribology of skin: review and analysis of experimental results for the friction coefficient of human skin, Tribol. Lett. (2012) 1–27. [11] S. Derler, L.C. Gerhardt, A. Lenz, E. Bertaux, M. Hadad, Friction of human skin against smooth and rough glass as a function of the contact pressure, Tribol. Int. 42 (2009) 1565–1574. [12] C. Hendriks, S. Franklin, Influence of surface roughness, material and climate conditions on the friction of human skin, Tribol. Lett. 37 (2010) 361–373. [13] R. Lewis, C. Menardi, A. Yoxall, J. Langley, Finger friction: grip and opening packaging, Wear 263 (2007) 1124–1132. [14] C. Pailler-Mattei, S. Nicoli, F. Pirot, R. Vargiolu, H. Zahouani, A new approach to describe the skin surface physical properties in vivo, Colloids Surf. B: Biointerfaces 68 (2009) 200–206. [15] C. Pailler-Mattei, S. Pavan, R. Vargiolu, F. Pirot, F. Falson, H. Zahouani, Contribution of stratum corneum in determining bio-tribological properties of the human skin, Wear 263 (2007) 1038–1043. [16] S.M. Pasumarty, S.A. Johnson, S.A. Watson, M.J. Adams, Friction of the human finger pad: influence of moisture, occlusion and velocity, Tribol. Lett. 44 (2011) 117–137. [17] L. Skedung, K. Danerlöv, U. Olofsson, M. Aikala, K. Niemi, J. Kettle, M.W. Rutland, Finger friction measurements on coated and uncoated printing papers, Tribol. Lett. 37 (2010) 389–399. [18] S. Tomlinson, R. Lewis, M. Carré, The effect of normal force and roughness on friction in human finger contact, Wear 267 (2009) 1311–1318. [19] M.A. Darden, C.J. Schwartz, Investigation of skin tribology and its effects on the tactile attributes of polymer fabrics, Wear 267 (2009) 1289–1294. [20] W. Li, M. Kong, X. Liu, Z. Zhou, Tribological behavior of scar skin and prosthetic skin in vivo, Tribol. Int. 41 (2008) 640–647. [21] K.P. Wilhelm, Bioengineering of the Skin: Skin Surface Imaging and Analysis, CRC Press, Boca Raton, FL, USA, 1997. [22] M. Kwiatkowska, S. Franklin, C. Hendriks, K. Kwiatkowski, Friction and deformation behaviour of human skin, Wear 267 (2009) 1264–1273. [23] F.P. Bowden, D. Tabor, The Friction and Lubrication of Solids, Clarendon Press, Oxford, UK, 1986, pp. 374. [24] S. Derler, U. Schrade, L.-C. Gerhardt, Tribology of human skin and mechanical skin equivalents in contact with textiles, Wear 263 (2007) 1112–1116.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Tribological evaluation of porcine skin Huaping Xiao a , Nethika Ariyasinghe b , Xingliang He a , Hong Liang a,∗ a b
Department of Mechanical Engineering, Texas A&M University, TX 77840, United States Department of Biomedical Engineering, University of Southern California, CA 90089, United States
a r t i c l e
i n f o
Article history: Received 15 June 2013 Received in revised form 11 November 2013 Accepted 3 December 2013 Available online 11 December 2013 Keywords: Coefficient of friction Pig skin Tribological properties Bearing ratio
a b s t r a c t This research studies the effects of external parameters on the friction of porcine skin. A tribometer was used to evaluate the frictional behavior of the same. The effects of DI water and body oil on porcine skin against steel and glass balls were evaluated in terms of coefficient of friction (COF). The COF dropped rapidly when DI water/body oil was introduced into the sliding system and remained stable when the volume of the liquid exceeded a certain value. The COF increased with increasing sliding speed under dry conditions and decreased in wet. Under an increasing normal force, the COF decreased regardless of the presence of liquid. The ratio of the real contact area to the nominal contact area of the skin with the steel/glass ball was found to increase with a power law as the applied force was increased. These results reveal basic tribological properties of the skin in contact with a hard slider. These properties could be used as reference for the design and development of artificial skin in prosthetic applications. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Prosthetics help people who have lost part of their body to lead a more normal life. Since the middle of last century, research in prosthetics has led the creation of various devices [1]. The interface between the human body and prosthetics has been studied for proper integration. Steege et al. established a finite-element (FE) model to describe the interface between a residual lower-limb and a prosthetic device [2]. The FE method was then widely adopted to model the residual limb–prosthesis interface [3,4]. It was found that in order to improve the efficiency of load transfer between the residual limb and the prosthetic device, the geometry of the socket should be modified according to the residual limb rather than using a replica. In addition to the modification of the interface, spherical and cylindrical wrapping algorithms [5] were developed to improve the geometry of prosthetics. Proper algorithm for the prosthetics is another key factor for sophisticated design of flexible prosthetics. Even though current prosthetics can visually replace a missing limb and may be able to replicate some functions of a human hand, they cannot replicate all the functions of a human hand. One reason is that current prosthetics are not integrated with artificial skins. The skin plays a crucial role in the hand’s ability to grasp and manipulate objects [6]. For better mimicking human’s behavior, artificial skin coated metal framework is a promising option. Because a person’s grip on an object is related to the frictional force between the hand and the object, the tribological properties of human skin
∗ Corresponding author. Tel.: +1 9798622623. E-mail address: [email protected] (H. Liang). 0927-7765/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.colsurfb.2013.12.006
must be studied to improve the prosthetic implants and develop artificial skin. Seo et al. [7] showed that surface friction affects the force that a person needs to grip an object. Under the same normal force, higher friction leads to greater friction force and a tighter grip. Any factor that changes the frictional properties between the object and the skin should affect the force of the grip. Due to its complex structure, the skin’s contact behavior is affected by various factors. The impacts of the surface roughness, the superficial sebum, the skin moisture, the relative velocity, and the applied normal force on tribological properties of skin have been studied [8–20]. Pailler-Mattei et al. [14] showed that the influence of the skin lipid film on the adhesion between the human skin and the steel indenter could be attributed to the capillary effect. The capillary effect is caused by the liquid capillary bridge between two surfaces without direct contact. Most of the studies have been done in vivo on human skin. However, the artificial skins in real application are separate systems from the human body and are tested ex vivo. Result for artificial skin is different from real human skin and there is neither clear understanding nor effective evaluation of artificial ones. To close this gap, further understanding of skin’s function, ex vivo evaluation of the real skin is needed. In the present research we propose an ex vivo approach in investigating a real skin, i.e., pig skin, in order to understand its behavior. In the ideal case, the ex vivo study needs to be done on human skin. The reason we used pig skins is that it is very difficult to obtain the specimen of human skin. The aim of this study is to understand the frictional behavior of porcine skin in various environments and conditions. The effects of deionized (DI) water and body oil (Neutrogena light sesame formula) on the coefficient of friction (COF) was evaluated. A contact model was developed in order to explain the principles of friction.
H. Xiao et al. / Colloids and Surfaces B: Biointerfaces 116 (2014) 734–738
Fig. 1. Schematic expression of the tribo-test.
2. Materials and methods
735
Fig. 2. Values of the COF change with rotating speed at the load of 0.1 N. (The hollow shapes stand for the steel-skin pair and the solid shapes are for the glass–skin pair. The circles, triangles and diamonds represent the conditions of dry contact, contact lubricated by water and contact lubricated by body oil, respectively. The presented COF is the mean value of 3 measurements in the glass–oil case, 10 measurements in the steel-dry case and 6 measurements in all the other cases.)
2.1. Materials Pig legs were bought from a local grocery store (HEB). The porcine skins were peeled off from the foreleg of the pig. The skins were washed with liquid soap and then ultrasonically cleaned in DI water for 5 min. The samples were left in open air to dry until they did not leave traces of water on the Kimwipes placed on top. Steel balls and glass balls with a diameter of 6.5 mm were used as the counter contact. Glass and steel were chosen because they are the most common rigid materials people will touch in daily life. Before each test, the rigid balls were ultrasonically cleaned in acetone and then in ethanol for 5 min each. 2.2. Tribological experiments The skins were cut into rectangles with a length of 60 mm and a width of 20 mm for tribotests. The measured thickness is from ∼1.5 mm to ∼2 mm. The surface roughness of the steel ball and glass ball were measured using a Newview 600 interferometer (Zygo Corp., USA). Plasticine was pressed against the pig skin to make a reversed replica of the skin. After drying up, the surface roughness of the replica was measured using a TR200 (Time Group Inc., USA) roughness tester. 5 measurements were taken for each sample. The measured surface roughness (Ra) is 5.814 ± 0.699 m for pig skin, 0.286 ± 0.035 m for glass ball, and 0.088 ± 0.009 m for steel ball. The COF between the skin and the steel ball/glass ball was measured using a pin-on-disk tribometer (CSM Instruments, Switzerland). The skins were clipped onto a microscope slide using binder clips. This slide was super glued to the rotational stage of the tribometer. The steel or glass ball was fixed to the steel rod of the tribometer as shown in Fig. 1. There is a piezoelectric sensor inside the supporting arm. When the normal load was applied, the resulting frictional force produced a torque to the supporting arm. This torque was recorded by the sensor and was converted to frictional force and COF. The COF under different conditions was measured. The sliding speed was increased in increments of 10 rpm from 10 rpm to 60 rpm with a radius of 3.5 mm, and the applied load was increased from 0.05 N to 0.2 N. To determine the impact of liquid on the COF, 1–15 L DI water and 1–6 L body oil were introduced to the tested area on skin. The change of COF with variation of sliding speeds and applied loads was measured under dry conditions first. The COF at different liquid volumes was measured while the sliding speed was set at 40 rpm under a load of 0.1 N. After the liquid reached to the maximum volume, the same range of sliding speeds and applied loads were tested again as under dry
conditions. The number of rotation cycles for each test is 10 and each test was repeated two to three times with the same sample under the same conditions. 2.3. Measurement of the contact area of the skin with glass The skins were cut into squares with a length and width of 10 mm. A glass slide was placed on top of the skin. An image of the contact area between the skin and glass slide was recorded using a KEYENCE VHX optical microscope (KEYENCE Corp., Japan) through the glass slide. 4 spots were measured in this study. The real contact area of the skin to the glass can be distinguished because of the contrast between the areas of the skin in contact with the glass and otherwise. The real contact area was then calculated using ImageJ program. 3. Results and discussion 3.1. Effects of sliding speed on COF The values of the COF at different sliding speeds for both dry and wet conditions are shown in Fig. 2. The presented COFs are the mean values of 3 measurements in the glass–oil case, 10 measurements in the steel-dry case and 6 measurements in all the other cases. The error bars show the standard deviation of those measurements. Under dry conditions there is a notable trend of going up for the COF when the sliding speeds are changed from 10 rpm to 60 rpm. After DI water or body oil is introduced into the contact, different trend is observed. The COF slightly decreases as the sliding speeds are increased. The plots shown in Fig. 2 reveal that the COF under dry conditions is always higher than that in wet. These results demonstrate that both DI water and body oil lubricate the skin–steel/glass contacts. In addition, lower COF with lubrication by body oil than by DI water indicates that body oil has better lubricating performance in this study. No conclusive remarks can be made to determine whether steel or glass has lower COF on the pig skin. Overall, the change of the COF under dry conditions is greater than that under wet conditions. This means that the speed plays a more significant role on COF in dry contacts. 3.2. Effects of applied force on COF The change of COF with variation of applied loads is displayed in Fig. 3. The COF consistently reduces with increasing load in all cases.
736
H. Xiao et al. / Colloids and Surfaces B: Biointerfaces 116 (2014) 734–738
Fig. 3. Change of the COF with applied load at the sliding speed of 40 rpm. (The shapes have the same meaning as in Fig. 2. The presented COF is the mean value of 3 measurements in the glass–oil case, 10 measurements in the steel-dry case and 6 measurements in all the other cases.)
Under small loads, there is a significant decrease in COF. The trend of decreasing COF becomes less notable as the applied load increases. Fig. 3 shows that the COF decreases faster under dry conditions than in wet. Curve fitting was done using power law relation: Y = mXn . The results show that the values of n are the smallest under dry condition. The highest value of n is obtained under oil lubrication. Since n is less than 0, the COF drops faster with a smaller n when the load is increased. These results demonstrate that the COF is less sensitive to applied load under wet conditions. Similar to the phenomenon observed in Fig. 2, the results from Fig. 3 indicate that body oil lubricates the contacts better than DI water. But the difference fades as the loads increases.
Fig. 4. Change in the COF as (a) body oil (b) DI water is introduced into the area of contact (The presented COF is the mean value of 3 measurements in the glass–oil case and 6 measurements in all the other cases.)
3.3. Effects of the amount of liquid on friction Fig. 4 shows how the introduction of body oil and DI water affects the COF. The presented COF is the mean value of 3 measurements in the glass–oil case and the mean value of 6 measurements in all the other cases. The error bars show the standard deviation of those results. From Fig. 4(a) it can be seen that the COF greatly reduces when only 1 l of body oil is added to the contact. However, the COF does not change much as the amount of body oil is >1 l. At the glass–skin contact, the COF increases slightly as more body oil is introduced. A similar situation is observed when DI water is added into the contact, as shown in Fig. 4(b). The COF drops rapidly when DI water is added, but the reducing trend of COF gradually diminishes with the presence of more DI water. When the volume of water is >4 l, the COF becomes stable in both contacts. When the liquid is firstly introduced, effective lubrication between the skin surface and the other surface is established, and so the COF drops fast when a small amount of DI water or body oil is introduced into the contact. However, once the volume of liquid is higher than a certain number, full lubrication is established. The COF will not change much with addition of more lubricant. Thus, the COF remains stable when the volume of liquid is over a certain amount. All the contact surfaces involved in this study, pig skin, steel and glass, are more oleophilic than hydrophilic, which means that body oil has lower contact angle on these materials than DI water. Less volume of oil is required to cover a certain area. In this study, 1 l of body oil is enough to cover the sliding track to establish full lubrication whereas 4 l of DI water is needed.
contact area was measure under different loads. The photos of the contact area are displayed in Fig. 5. The dark spots are the real contact area. It is obvious that the dark spots gradually enlarge as the load is increased. The ratio of the dark area to the whole area (the nominal contact area) was calculated using ImageJ program. The
3.4. Variation of bearing ratio with load To investigate why load affects the COF in the manner shown above, the bearing ratio of the real contact area to the nominal
Fig. 5. Photos of contact area at spot 4 under different loads.
H. Xiao et al. / Colloids and Surfaces B: Biointerfaces 116 (2014) 734–738
737
Fig. 6. Increase of bearing ratio with applied load. Table 1 The calculated values of a and b and the standard error.
Spot 1 Spot 2 Spot 3 Spot 4
Value of a
Value of b
Standard error
1.044 0.293 0.146 0.933
0.305 0.981 0.685 0.806
a 0.6 ± 0.45 b 0.69 ± 0.29
variation of the bearing ratio with the applied load is represented in Fig. 6. The lines in Fig. 6 are fitted based on the power law: Y = aX
b
3.5. Contact model Skins are usually characterized as elastomers which have viscoelastic and nonlinear properties [21]. When a rigid ball is sliding on the pig skin with a normal force, there is a noticeable deformation of the skin at the front surface as shown in Fig. 7(a). At the back surface, the deformation reduces and recovers when the surfaces lose contact with each other. This scenario is consistent with the findings of Kwiatkowska et al.’s study [22]. Based on these effects, a two-term model is employed to describe the frictional force between the skin surface and the counter-surface. The total effective frictional force is a combination of the adhesion friction force, which is the dominant component, and the deformation friction force. Therefore, the effective frictional force can be expressed as [23]: (2)
where Feff is the effective frictional force, Fadh is the adhesion friction force, and Fdef is the friction force due to the deformation of the skin. The frictional coefficient can be calculated as: =
Fadh + Fdef N
where N is the normal force. Since the adhesion friction force governs the frictional force, the frictional coefficient can be approximated as: ≈
(1)
where Y is the contact area, X is the applied load, and a and b are the fitted parameters (Table 1). The difference between the fitted value and the measured value is small, indicating that the fitting curves are good approximations of the real trends. One possible explanation for the high variation is that the skin surface is not perfectly flat. When a slide is forced against the skin, some region of the skin sustains more loads and other region sustains less. As shown in the Fig. 5(a), the contact regions are not uniformly distributed. The bearing ratio inside the dashed line is definitely higher than that outside. When the external load is increased, the increase of real contact area for different region could be different. Thus, there is high variation between the four measurements. Measurement uncertainty is another reason for the observed variation.
Feff = Fadh + Fdef
Fig. 7. Schematic expression of skin-rigid ball contact (a) before lubrication (b) after lubrication. The small particles in (b) represent the liquid molecules. The contact between the skin and the rigid ball is partly separated by the liquid.
(3)
Fadh N
(4)
The adhesion friction force is generated from the repulsive and attractive forces that operate in a short range. It is proportional to the real contact area, Ar (Fadh ∝ Ar ). Based on the forces shown in Fig. 7, the relationship between the bearing ratio and the normal force is: R=
Ar = aN b A
(5)
A is the nominal contact area; Ar = AaN b
(6)
The frictional coefficient can be calculated by combining all the relationships shown above: ≈
Fadh Ar AaN b ∝ = = AaN b−1 N N N
(7)
Eq. (7) shows that the frictional coefficient is proportional to N b−1 . From Table 1, all the estimated values of b are less than 1. Therefore, the frictional coefficients are inversely related to the applied normal force, which is a possible explanation for the results shown in Fig. 3. The calculated values of (b − 1) are comparable with the values of m in Fig. 3. When the load is kept constant, the COF increases as the sliding speed increases in most of the dry contact cases shown in Fig. 2. Assuming that the real contact area stays constant under the same load, the adhesion friction force remains constant as the speed changes as well. The energy dissipation caused by the deformation and recovery of the skin, however, increases as the sliding speed increases. At higher speed, the sliding distance in the same time period is longer. This means more deformation and recovery of the skin happen in the same time period. The energy dissipation due to the deformation and recovery of skin increases as well. The increased energy consumption could lead to a higher COF. Under wet conditions, the lubrication effect of the liquid plays a remarkable role in contact area between the skin and the steel/glass ball as shown in Fig. 7(b). Portion of the contact region is separated by the lubricant. In these regions, the adhesion friction force greatly decreases leading to the decrease of COF. Assume that most of the
738
H. Xiao et al. / Colloids and Surfaces B: Biointerfaces 116 (2014) 734–738
contact points between the skin and the rigid ball are separated in the presence of adequate lubricant under high speed. Hydrodynamic lubrication could be established under these conditions. The influence of hydrodynamic lubrication is enhanced as the relative speed between the skin surface and the counter-surface is raised. Thus, the COF decreases gradually with increasing sliding speed under wet conditions as shown in Fig. 2. Most of the prosthetic devices in application are metal based. Previous study [24] showed that specially designed alternative materials (artificial skin) have better approximation to real skin than the metal based material. Artificial skin coated metal frameworks is an option for better prostheses. Two aspects involving the artificial skin need remarkable notice. First, wear of the artificial skin should be avoided. Second, the artificial skin could provide sufficient grabbing force for the user to handle or hold objects. From the point view of tribology, these two aspects are contradictory. To avoid the wear the COF needs to be reduced whereas greater COF is preferable to provide higher grabbing force. Based on the findings of this study, introduction of water and oil is an effective way to reduce the COF. So in the design of artificial skin, the material which is capable of capturing the moisture from the environment has advantage if lower COF is desired. On the contrary, the material with the ability of repelling liquid and forming larger real contact area is preferable in the cases that higher grabbing force has priority. 4. Conclusion The COF between pig skin and steel/glass ball decreases when DI water or body oil is introduced. However, the amount of reduction diminishes as more liquid is added to the contact, and then COF becomes stable. The COF is found to decrease with applied load following a power law. Under dry conditions, the COF increases with the speed, which attributes to the higher energy dissipation at higher speed. Under wet conditions, however, the COF decreases with speed because the lubrication is improved as speed is increased. The bearing ratio between the pig skin and glass slide is found to increase with loads following a power law. The curve fitting shows that the experimental results are consistent with the calculated value. This research could be beneficial in design and evaluation of artificial skin, as material surfaces and/or as a coating for prosthetic applications. Acknowledgement NA was supported by Dr. Hsien through REU program at TAMU.
References [1] G. Musser, Making them whole, Sci. Am. 298 (2008) 52–53. [2] J. Steege, D. Schnur, R. Vorhis, J. Rovick, Finite element analysis as a method of pressure prediction at the below-knee socket interface, in: Proceedings of RESNA 10th Annual Conference, California, 1987, pp. 814–816. [3] J. Brennan, D. Childress, Finite element and experimental investigation of above-knee amputee limb/prosthesis systems: a comparative study, ASME Adv. Bioeng. 20 (1991) 547–550. [4] M. Zhang, A.F.T. Mak, V. Roberts, Finite element modelling of a residual lowerlimb in a prosthetic socket: a survey of the development in the first decade, Med. Eng. Phys. 20 (1998) 360–373. [5] I.W. Charlton, G.R. Johnson, Application of spherical and cylindrical wrapping algorithms in a musculoskeletal model of the upper limb, J. Biomech. 34 (2001) 1209–1216. [6] C.A. Brohem, L.B. da Silva Cardeal, M. Tiago, M.S. Soengas, S.B. de Moraes Barros, S.S. Maria-Engler, Artificial skin in perspective: concepts and applications, Pigment Cell Melanoma Res. 24 (2011) 35–50. [7] N.J. Seo, J.K. Shim, A.K. Engel, L.R. Enders, Grip surface affects maximum pinch force, human factors, J. Human Factors Ergon. Soc. 53 (2011) 740–748. [8] T. André, V. Lévesque, V. Hayward, P. Lefèvre, J.L. Thonnard, Effect of skin hydration on the dynamics of fingertip gripping contact, J. R. Soc. Interface 8 (2011) 1574–1583. [9] M. Changizi, R. Weber, R. Kotecha, J. Palazzo, Are wet-induced wrinkled fingers primate rain treads, Brain Behav. Evol. 77 (2011) 286–290. [10] S. Derler, L.C. Gerhardt, Tribology of skin: review and analysis of experimental results for the friction coefficient of human skin, Tribol. Lett. (2012) 1–27. [11] S. Derler, L.C. Gerhardt, A. Lenz, E. Bertaux, M. Hadad, Friction of human skin against smooth and rough glass as a function of the contact pressure, Tribol. Int. 42 (2009) 1565–1574. [12] C. Hendriks, S. Franklin, Influence of surface roughness, material and climate conditions on the friction of human skin, Tribol. Lett. 37 (2010) 361–373. [13] R. Lewis, C. Menardi, A. Yoxall, J. Langley, Finger friction: grip and opening packaging, Wear 263 (2007) 1124–1132. [14] C. Pailler-Mattei, S. Nicoli, F. Pirot, R. Vargiolu, H. Zahouani, A new approach to describe the skin surface physical properties in vivo, Colloids Surf. B: Biointerfaces 68 (2009) 200–206. [15] C. Pailler-Mattei, S. Pavan, R. Vargiolu, F. Pirot, F. Falson, H. Zahouani, Contribution of stratum corneum in determining bio-tribological properties of the human skin, Wear 263 (2007) 1038–1043. [16] S.M. Pasumarty, S.A. Johnson, S.A. Watson, M.J. Adams, Friction of the human finger pad: influence of moisture, occlusion and velocity, Tribol. Lett. 44 (2011) 117–137. [17] L. Skedung, K. Danerlöv, U. Olofsson, M. Aikala, K. Niemi, J. Kettle, M.W. Rutland, Finger friction measurements on coated and uncoated printing papers, Tribol. Lett. 37 (2010) 389–399. [18] S. Tomlinson, R. Lewis, M. Carré, The effect of normal force and roughness on friction in human finger contact, Wear 267 (2009) 1311–1318. [19] M.A. Darden, C.J. Schwartz, Investigation of skin tribology and its effects on the tactile attributes of polymer fabrics, Wear 267 (2009) 1289–1294. [20] W. Li, M. Kong, X. Liu, Z. Zhou, Tribological behavior of scar skin and prosthetic skin in vivo, Tribol. Int. 41 (2008) 640–647. [21] K.P. Wilhelm, Bioengineering of the Skin: Skin Surface Imaging and Analysis, CRC Press, Boca Raton, FL, USA, 1997. [22] M. Kwiatkowska, S. Franklin, C. Hendriks, K. Kwiatkowski, Friction and deformation behaviour of human skin, Wear 267 (2009) 1264–1273. [23] F.P. Bowden, D. Tabor, The Friction and Lubrication of Solids, Clarendon Press, Oxford, UK, 1986, pp. 374. [24] S. Derler, U. Schrade, L.-C. Gerhardt, Tribology of human skin and mechanical skin equivalents in contact with textiles, Wear 263 (2007) 1112–1116.
