Skin Research and Technology 2003; 9: 227-234 Printed in Denmark. All rights reserved

Copyright Blackwdl Munksgaard 2003

Skin Research and Technology

ISSN 0909-752X

Review

Coefficient of friction: tribological studies in man

an overview

Raja K. Sivamani¹'², Jack Goodman¹, Norm V. Gitis¹ and Howard I. Maibach²

¹ Center for Tribology, Inc., Campbell, CA, USA and Department of Dermatology, School of Medicine, University of California, San Francisco, CA, USA

Background/purpose: Compared to other studies of skin, relatively few studies have focused on the friction of skin. This work reviews existing skin friction, emphasizing test apparatuses and parameters that have added to information regarding the friction coefficient. This review also outlines what factors are important to consider in future friction studies.

Methods: Past studies have utilized numerous designs for a test apparatus, including probe geometry and material, as well as various probe motions (rotational vs. linear). Most tests were performed in vivo; a few were performed in vitro and on porcine skin.

Results: Differences in probe material, geometry and smoothness affect friction coefficient measurements. An increase in skin hydration, either through water or through moisturizer application, increases its friction coefficient;

a decrease in skin hydration, either through clinical dermatitis or through alcohol addition, decreases the coefficient. Differences are present between anatomical sites. Conflicting results are found regarding age and no differences are apparent as a result of gender or race.

Conclusion: Skin friction appears to be dependent on several factors-such as age, anatomical site and skin hydration. The choice of the probe and the test apparatus also influence the measurement.

Key words: coefficient - friction - review - skin - tribology

© Blackwell Munksgaard, 2003 Accepted for publication 12 August 2002

PHYSIOLOGICALLY, THE skin is the first line of defense against the environment and it is repeatedly subjected to physical and chemical damage. The skin's mechanical properties - such as its friction characteristics - can alter under this repeated damage. Mechanically, friction allows us to keep from slipping as we step out of the shower, hold the Styrofoam cups of coffee, or turn the steering wheel in our cars. Because the skin is a surface itself, it is convenient to analyze and describe it in terms of a surface phenomenon - such as friction; friction studies on skin provide valuable insight into how the skin interacts with other surfaces. Friction also provides information about the skin under various conditions - for example, age and gender - and under various chemical treatments - for example, lotions and moisturizers. Studying the friction of skin supplements other mechanical tests. An advantage of friction studies

is that they can be performed with non-invasive methods and give a measure of the skin's health -for example, skin hydration. Naylor (1) showed that moistened skin has an elevated friction response and El-Shimi (2) demonstrated that drier skin has a lowered friction response. Friction provides a quantitative measurement to assess skin condition.

The friction parameter generally measured is the coefficient of friction. In order to measure the friction coefficient, one surface is brought into contact with another and moved relative to it. When the two surfaces contact, the perpendicular force is defined as the normal force N. The tangential friction force F is that force which opposes relative movement between the two surfaces. From Amonton's law, the coefficient of friction u is defined as the ratio of the friction force to the normal force:

The friction coefficient can be measured in two ways: (i) the static friction coefficient and (ii) the dynamic or kinetic friction coefficient u_k. The static friction coefficient is defined as the ratio of the force required to initiate relative movement to the normal force between the surfaces; the dynamic or kinetic friction coefficient is defined as the ratio of the friction force to the normal force when the two surfaces are moving relative to each other. Much of the research has been focused on the dynamic friction coefficient wherein the two surfaces move at a relative constant velocity. Most of the friction studies on skin have dealt with the dynamic friction coefficient and the subscript k is usually dropped. This overview references the dynamic coefficient of friction unless otherwise noted.

According to Amonton's law, the dynamic friction coefficient remains unchanged regardless of the probe velocity or applied normal load in making the measurement. Amonton's laws hold true in the case of solids with limited elastic properties. Although Naylor (1) concluded Amonton's law to be true, later studies by El-Shimi (2), Comaish and Bottoms (3) and Koudine et al. (4) found that skin deviates from Amonton's law, because their studies found the friction coefficient to be inversely proportional to load. El-Shimi (2) and Comaish and Bottoms (3) reasoned that the rise in friction coefficient with decreasing load resulted from the viscoelastic nature of the skin allowing for a non-linear deformation of the skin.

Materials and Methods

Experimental design

Various experimental designs have been devised in order to measure the friction on skin. They focus on measuring friction by pressing a probe onto the skin with a known normal force, and then detecting the skin's frictional resistance to movement of the probe. The designs fall into two categories:

• A probe moved across the skin in a linear fashion.

• A rotating probe in contact with the skin surface.

In the linear designs, the probe movement is accomplished in several ways. Comaish and

Bottoms (3) utilized one of the simplest linear designs; they moved the probe across the skin by attaching it to a pan of weights by means of a pulley. Their design is illustrated schematically in Fig. 1. Weights are placed in the pan such that the probe slides over the skin at a constant velocity. This allows for the calculation of the dynamic friction coefficient by dividing the total weight in the pan by the normal load on the probe. However, there are many inaccuracies involved with this method as there is no monitor or control of probe speed or normal force.

More sophisticated linear designs followed the design used by Comaish and Bottoms (3), but provided motorized unidirectional movement of the probe or the use of a reciprocating motor in order to move the probe back and forth. In both designs the motorization afforded greater control in maintaining the velocity of the probe. Strain gauges were used in order to measure the friction force as the probe moved along the skin surface.

The second design category measures friction with a rotating wheel pressed onto the surface of the skin with a known normal force. Highley et al. (5) measured the frictional resistance by determining the angular recoil of the instrument as the wheel contacted the skin. They measured this angular recoil by recording the proportion of light that hit a dual element photocell. An electrical signal was then generated in proportion to the frictional resistance. Comaish et al. (6) developed a portable, hand-held device (Newcastle Friction Meter) that relied on a torsion spring in order to measure the skin's frictional resistance. The devices are surveyed in Table 1.

An important part of designing a friction measurement apparatus is the choice of probe size, shape and material. Because friction is an interaction between two surfaces, the probe geometry and material can affect the values calculated for the friction coefficient of the other surface. Several shapes and materials have been used as outlined

Normal force

TABLE 1. Probe and apparatus used in order to measure the dynamic friction coefficient u of untreated 'normal' skin in vivo

in Table 1. Also, results will be more accurate when the probe's normal force is maintained at a constant value or continuously monitored; previous methods used to maintain the normal force include spring mechanisms or static weights to weigh down the probe (Table 1). These parameters are revisited critically later.

Much effort has been made in understanding how skin friction changes with differing biological conditions and upon the application of various products to the skin surface. These studies are of interest to various companies that manufacture products meant as skin topical agents, because friction measurements can provide clues regarding the effectiveness of their products. Previous studies are outlined in Table 2.

Hydration

Hydration is a complex phenomenon influenced by intrinsic - that is, age, anatomical site - and extrinsic - that is, ambient humidity, chemical exposure - factors. These factors can affect the mechanical properties of skin and research has been performed in order to correlate hydration levels with the skin's friction coefficient. Hydration studies have investigated how increases and decreases in skin hydration correlated with the friction coefficient. In past studies, researchers generally induced increases in skin hydration through water exposure. However, decreases in skin hydration were not experimentally induced

All studies were performed in vivo except Comaish and Bottoms (3) who performed some in vitro tests on human skin and Hills et al. (15) utilized porcine skin in their in vitro tests.

and dehydration studies were performed between subjects with 'normal' skin and subjects that had clinically 'dry' skin (2,12).

Lubricants/emollients/moisturizers Much of the reviewed research has been devoted to ascertaining how the application of certain ingredients influences the skin surface, which is of interest to the cosmetic/moisturizer and lubricant industry. The studies focused on the effects of talcum powder (2,3), oils (2,3,5,14) and skin creams/moisturizers (7,14). Hills et al. (15) analyzed how changes in the friction coefficient, following emollient application, differed with temperature.

Probes

As mentioned earlier, the probe geometry and material influence the measured value of the friction coefficient, because friction is a probe-skin interaction phenomenon. Few studies have examined probe effects; El-Shimi (2) studied probe roughness and Comaish and Bottoms (3) probe roughness and material.

skin's mechanical properties change under various conditions.

Previous studies report a range of values for the skin's friction coefficient. Dynamic friction coefficient measurements (Table 3) fall in the range 0.12-0.7; however, most fall in a narrower range of 0.2-0.5 (Fig. 2). Besides natural variations in skin, the wide range in results may be as a result of differences in probe movement, geometry and material, and controlled monitoring of the normal force. In the reviewed friction measurement apparatuses, the two types of probe movements utilized were rotational and linear (Table 1). The linear probe constantly moves over 'untested' skin and the rotational probe spins over 'tested' skin. The different movements can lead to discrepancies in reported values for the skin friction

Fig. 2. Outline of the ranges in the dynamic coefficient of friction. These ranges reflect measurement of untreated 'normal' skin friction in vivo.

coefficient. Another important source of variation may be in the ability to control the normal force while the probe is moving over the skin surface. The skin friction instruments are designed in order to measure the frictional resistance of the skin and it is assumed that the normal force is constant. During a test the normal force may not remain constant as a result of many factors - for example, uneven skin surface, inaccurate spring and/or a non-uniform distribution of static weights placed above the probe. Therefore, the assumption of a constant normal force may be incorrect and can lead to inaccuracy and variation in the calculated friction coefficient. A third source for variation is the choice of the probe material. Because friction is a surface phenomenon between two materials, the choice of the probe will influence the numerical value obtained for the friction coefficient.

Hydration

Hydration studies reveal that drier skin has lowered friction while hydrated skin has an increased amount of friction (Table 4). However, the skin response is more complex, because very wet skin also has a lowered friction coefficient much like the characteristics of dry skin (16). Most studies focus on an intermediate zone of hydration where the skin has been moistened without an appreciable 'slippery' layer of water on the skin. Results in Table 4 show that the increases in friction are varied and this possibly results from the various probes used. Although the addition of water increases the friction coefficient, this effect lasts for a period of minutes before the skin returns to its 'normal' state (2, 5,

14, 17). The water has an effect of softening the skin and this in turn allows for greater contact area between the probe and the skin. Also, water results in adhesive forces between the water and the probe. Thus, there is more frictional resistance between the skin and the probe and results in a higher friction coefficient (18). Because the water evaporates over an order of minutes, the skin returns to its 'normal' state in the same time frame. For dry skin, the skin becomes less supple and the probe does not achieve as much contact area and this allows the probe to glide more easily over the skin surface. This results in a lowered friction coefficient as seen in the iso-propyl study (17) and in prior studies involving subjects with clinically dry skin (2,12). The agreement between the experimentally induced dry skin and clinical dry skin is expected (19).

Lubricants/emollients/moisturizers The studies on lubricants, emollients and moisturizers are important for cosmetics and products developed in order to make the skin look and feel healthier. The literature reports that the important qualitative characteristics in skin topical agents are skin smoothness, greasiness and moist-urization (18,20). Previous research has tried to describe these subjective, qualitative descriptions in a quantitative fashion by correlating them against the friction coefficient. Prall (7) was unable to make a direct correlation of skin smoothness with friction coefficient until he added skin topography and hardness to the analysis. Nacht et al. (14) found a linear correlation between perceived greasiness and the friction coefficient (Fig. 4).

*Comaish and Bottoms (3) studied the change in the static friction coefficient in their hydration study.

Talcum powder

El-Shimi (2) and Comaish and Bottoms (3) showed that the friction coefficient decreased after the application of powder. El-Shimi (2) found that the friction coefficient decreased by 50% after application; Comaish and Bottoms (3), in analyzing the static friction coefficient, observed an insignificant change for a wool probe and a 30% decrease in friction with a polyethylene probe. However, they also found that wetting the talcum powder caused an increase in the measured friction.

Lubricant oils

A lowering in the friction coefficient is the initial effect after the application of oils and oil-based lubricants (2,5,14). Nacht et al. (14) and Highley et al. (5) also showed that after the initial decrease in friction, the oils eventually raised the skin's friction coefficient. The results of the lubricant cosmetic studies by Nacht et al. (14) are shown in Fig. 3.

Emollients and moisturizers Prall (7) and Nacht et al. (14) found that the friction coefficient rises with the addition of emollients and creams in a fashion similar to water. However, the effects of the creams lasted for hours, whereas the water effects lasted for about 5-20 min (7,17). Hills et al. (15) also studied emollients, but they examined how various emollients compared against one another and how changes in temperature changed the friction coefficient. At a higher temperature (45 °C), most emollients

lowered the friction coefficient to a greater degree than at a lower temperature (18 °C).

When lubricant/moisturizers are applied to the skin, the skin friction is affected in three general ways (14,18).

• A large, immediate increase in the friction coefficient, similar to water application, that follows with a slow decrease in the friction coefficient. These agents can be interpreted to act by immediate hydration of the skin through some aqueous means in order to give the immediate increase in friction. In Fig. 5, cream B falls into this category and in Fig. 4, creams A, B and C represent this type of lubricant/moisturizer.

• An initial decrease in the friction coefficient that is followed by an overall increase in the friction

Fig. 4. Correlation between changes in the friction coefficient and the sensory perception of greasiness. A, B, C, D, E and F represent different creams that were applied to the skin. The reported percent change in the friction coefficient is immediately after application and the greasiness scores were subjective evaluations (From Nacht et al. (14)).

Time = -1 is immediately prior to application; Time = 0 is immediately after application

Fig. 3. Effect of lubricant cosmetic ingredient on skin friction coefficient. Amount applied of each material: approx. 2 mg cm^-2. Reproduced from Nacht et al. (14) (mean of five subjects but P-value was not published). Time— —1 is immediately prior to application; Time — 0 is immediately after application.

Elapsed time after application

0 min is immediately before application

Fig. 5. Effects on the dynamic friction coefficient after applying moisturizing creams. The cream was applied to the back of the finger and then monitored for 4h as shown above. Cream A was Loreal® Plentitude Hydra-Renewal Cream, a slow-acting, long-duration moisturizer. Cream B was Loreaf Plentitude Excell-A³ Alpha Hydroxy Cream, a fast-acting, short-duration moisturizer; 0 min is immediately before application. Each data point represents the average of four measurements; (n = 2; P< 0.05 for 2,3, and 4h marks).

coefficient over time. These agents are fairly greasy products (Fig. 3) and this greasiness causes the immediate decrease in the friction coefficient. The eventual rise in the friction coefficient is probably because of the increase in skin hydration through the occlusive effects of these agents (21). Representations of a few ingredients that elicit this response are in Fig. 3 and represented as cream F in Fig. 4. • A small, immediate increase in the friction coefficient that then increases slowly with time. These agents are interpreted to act as a combination of effects seen in the previous two cases. These lubricants/moisturizers have ingredients and agents that serve to both hydrate the skin through some aqueous method and prevent water loss through some occlusive mechanism. Because of the presence of these occlusive agents, which tend to be more slippery, the immediate rise in the friction coefficient is lower than in products that fall into the first category listed above. In Fig. 5, this is seen in cream A and in Fig. 4, this is seen in creams D and E.

Probes

El-Shimi (2) and Comaish and Bottoms (3) compared probes (Tables 3 and 4) and found that smoother probes gave higher friction coefficient measurements. El-Shimi (2) noted that higher friction coefficient measurements were made with a smoother stainless steel probe as opposed to a roughened stainless steel probe. Comaish and Bottoms (3) found a similar result with two types of nylon probes: a sheet probe and a knitted probe. The sheet probe (the smoother of the two) gave a

higher friction coefficient measurement. El-Shimi (2) postulates that the smoother probe forms more contact points with the skin and has a greater skin contact area than the rougher probe, resulting in more resistance from the skin and a larger measurement for the friction coefficient.

Anatomic region, age, gender and race Few studies address the effects of anatomic region, age, gender, or race as they pertain to the friction coefficient. To date, no significant differences have been found with regard to gender (8,22) or race (23). Age-related studies have been contradictory where some authors found no difference (8,22) and others found differences (10,11).

The friction coefficient varies with anatomical site. Cua et al. (8,22) found that friction coefficients varied from 0.12 on the abdomen to 0.34 on the forehead. Eisner et al. (11) measured the vulvar friction coefficient at 0.66, whereas the forearm friction coefficient was 0.48. Manuskiatti et al. (23) studied skin roughness and found significant differences in skin roughness at various anatomical sites. Differences in environmental influences - that is, sun exposure - and hydration may account for this. Eisner et al. (11) showed that the more-hydrated vulvar skin had a 35% higher friction coefficient than the forearm, in agreement with hydration studies that contend that skin has an increased friction coefficient under increased hydration.

With respect to age, friction measurement results are contradictory. Cua et al. (22) showed no differences in friction with respect to age except for friction measurements on the ankle. Eisner et al. (11) also performed age-related tests and found no differences in the vulvar friction coefficient, but observed a higher forearm friction coefficient in younger subjects. They postulate that the skin on parts of the body that become exposed to sunlight can undergo photoaging and thus, forearm skin shows evidence of age-related differences while the light-protected vulvar skin does not (11). Asserin et al. (10) concluded that younger subjects had a higher forearm friction coefficient than older subjects.

There are few gender-related friction studies. Cua et al. (8,22) found no significant friction differences between the genders. There are no studies addressing race as it pertains to friction, but Manuskiatti et al. (23) looked for racial (black and white skin) differences in skin roughness and

Conclusion

Although there have been limited studies dealing with the measurement of the skin friction coefficient, past studies and our study (17) show that differences in skin, because of various factors -such as age and hydration - can be correlated with the friction coefficient. Friction coefficient studies can serve as a quantitative method to investigate how skin differs on various parts of the body and how it differs between different people. It is also a useful method for tracking the changes resulting from the environmental and chemical treatments - such as sunlight - and when various chemicals are applied to the skin - such as soaps, lubricants and skin creams. The reviewed studies show that friction is an important parameter for understanding the skin's mechanical state. The reviewed studies also indicate that the design of the test apparatus is an extremely important factor, because test design parameters can also have an influence on friction measurements. A better appreciation of the importance of the friction coefficient will become clearer as measurement methods improve and allow for greater accuracy.

References

1. Naylor PFD. The skin surface and friction. Br J Dermatol 1955; 67: 239-248.

2. El-Shimi AF. In vivo skin friction measurements. J Soc Cosmet Chem 1977; 28: 37-51.

3. Comaish S, Bottoms E. The skin and friction: deviations from Amonton's laws, and the effects if hydration and lubrication. Br J Dermatol 1971; 84: 37-43.

4. Koudine AA, Barquins M, Anthoine PH, Auberst L, Leveque J-L. Frictional properties of skin: proposal of a new approach. Int J Cosmet Sci 2000; 22:11-20.

5. Highley DR, Coomey M, DenBeste M, Wolfram LJ. Frictional properties of skin. J Invest Dermatol 1977; 69: 303-305.

6. Comaish JS, Harborow PRH, Hofman DA. A hand-held friction meter. Br J Dermatol 1973; 89: 33-35.

7. Prall JK. Instrumental evaluation of the effects of cosmetic products on skin surfaces with particular reference to smoothness. J Soc Cosmet Chem 1973; 24: 693-707.

8. Cua A, Wilheim KP, Maibach HI. Frictional properties of human skin: relation to age, sex and anatomical region, stratum corneum hydration and transepidermal water loss. Br J Dermatol 1990; 123: 473-479.

9. Johnson SA, Gorman DM, Adams MJ, Briscoe BJ. The friction and lubrication of human stratum corneum.

In: Dowson D, et al., eds. Thin Films in Tribology. Proceedings of the 19th Leeds-Lyon Symposium on Tribology. Elsevier; 1993. p. 663-672.

10. Asserin J, Zahouani H, Humbert PH, Couturaud V, Mougin D. Measurement of the friction coefficient of the human skin in vivo. Quantification of the cutaneous smoothness. Colloids Surf B: Biointerfaces 2000; 19:1-12.

11. Eisner P, Wilheim D, Maibach HI. Frictional properties of human forearm and vulvar skin: influence of age and correlation with transepidermal water loss and capacitance. Dermatologica 1990; 181: 88-91.

12. Loden M, Olsson H, Axell T, Linde YW. Friction, capacitance and transepidermal water loss (TEWL) in dry ato-pic and normal skin. Br J Dermatol 1992; 126:137-141.

13. Sulzberger MB, Cortese TA, Fishman L Jr, Wiley H. Studies on blisters produced by friction. J Invest Dermatol 1966; 47: 456-465.

14. Nacht S, Close J, Yeung D, Cans EH. Skin friction coefficient: changes induced by skin hydration and emollient application and correlation with perceived skin feel. J Soc Cosmet Chem 1981; 32: 55-65.

15. Hills RJ, Unsworth A, Ive FA. A comparative study of the frictional properties of emollient bath additives using porcine skin. Br J Dermatol 1994; 130: 37-41.

16. Dawson, D. In: Wilheim K-P, Eisner P, Berardesca E, Maibach H, eds. Bioengineering of the skin: skin surface imaging and analysis. Boca Raton: CRC Press; 1997. p. 159-179.

17. Sivamani RK, Goodman J, Gitis NV, Maibach HI. Friction coefficient of skin in real-time. In press.

18. Wolfram LJ. Friction of skin. J Soc Cosmet Chem 1983; 34: 465-476.

19. Denda M. In: Loden M, Maibach H, eds. Dry skin and moisturizers: chemistry and function. Boca Raton: CRC Press; 2000. p. 147-153.

20. Wolfram LJ. In: Leveque J-L, ed. Cutaneous investigation in health and disease: noninvasive methods and instrumentation, Chapter 3, New York, NY: Marcel Dekker 1989.

21. Zhai H, Maibach HI. Effects of skin occlusion on percutaneous absorption: an overview. Skin Pharmacol Appl Skin Physiol 2001; 14:1-10.

22. Cua AB, Wilheim K-P, Maibach HI. Skin surface lipid and skin friction: relation to age, sex, and anatomical region. Skin Pharm 1995; 8: 246-251.

23. Manuskiatti W, Schwindt DA, Maibach HI. Influence of age, anatomic site and race on skin roughness and scaliness. Dermatology 1998; 196: 401-407.

Address:

Howard 1. Maibach

Department of Dermatology

School of Medicine

University of California

San Francisco

Box 0989

Surge 110

San Francisco

CA, 94143-0989

USA

Tel: 415 476 2468 Fax: 415 753 5304 e-mail: himjlm@itsa.ucsf.edu