4O JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Boundary Lubrication Mixed Lubrication• Fluid or Hydrodynamic Lubrication I I I I I I I I I I I I I I I So•erfeld Number (•p•) Figure 1. Stribeck curve showing the relationship between friction coefficient and the Sommerfeld number relationship is represented schematically in Fig. 1. It has been shown that the value of/x in boundary lubrication depends greatly on the state of the adsorbed film and that, generally speaking, the film must be in a condensed state to give a low coefficient of friction. A number of models have been suggested to explain the mechanism of boundary lubrication (1, 4, 8, 9). PKEVIOUS FRICTION MEASUREMENTS ON SKIN Few papers have been published on the frictional characteristics of skin. Naylot (10) measured the friction of a polyethylene ball rubbing against the skin. He found that the friction was higher when the skin was damp than when it was either very wet or dry. Appeldoorn and Barnett (11) have observed other distinctive frictional characteristics of skin as follow: (1) skin friction is "relatively high " (2) a small amount of talcum powder greatly reduces skin friction. This is well known, but it is not a characteristic of the friction of other systems such as steel-against-steel, where talc increases friction and (3) the skin friction is higher on a smooth surface than it is on a rough surface. This behavior is just the opposite to that normally encountered, but it can be verified by rubbing one's finger on a microscope slide. The friction is much greater on the clear glass (smooth) part than on the ground glass (rough) part. Although Appeldoorn and Barnett have not conducted any in vivo work, they found that an in vitro model combining a rubber ball and a rotating stainless steel cylinder cot-

SKIN FRICTION MEASUREMENTS 41 relates well with their observations on the behavior of skin friction. They concluded that the property of skin that gives it the unusual and characteristic behavior is not its roughness nor chemical composition, but its "flexibility." Like rubber, the skin can flex to conform to the shape of another surface. This gives it a relatively large area of contact and, therefore, a high coefficient of friction as compared to the relatively unflexible metal or plastic materials. These results have been confirmed by Prall (12) and by Comaish and Bottoms (13) on skin under in vivo conditions. EXPERIMENTAL TECHNIQUE Earlier experimental techniques for skin friction measurements have been reviewed by Prall (12). Traditionally, friction measurements involve the sliding of a probe over the skin and the force is determined as a function of load. Prall describes a friction dyna- mometer which features a constant-thrust friction head whereby a standard ground glass disc was pressed against the skin with a force of 200 g/cm 2. The friction head was attached to the shaft of an ac motor, which was energized by a variable transformer. In use, the friction head was presented to the skin, and the power to the motor gradually increased until the friction head just started to rotate. In this work, we employ a modified Haake viscometer (RV-1)* to measure the friction behavior of skin in vivo with the help of a rotating stainless steel probe in contact with the arm (or any other part) surface. Preliminary experiments have shown the need for controlling the •ontact pressure between the skin surface and the probe. To this end, a special probe assembly was designed such that a constant load was maintained in contact with the skin surface in the course of the experiment. The assembly is depicted schematically in Fig. 2. The assembly features an adapter which fits tightly onto the shaft of the Haake measuring head. The part carrying the load and the probe is precisely machined so that it slides smoothly over the cylindrical adapter. The extent of vertical movement of the probe attachment is controlled by the size of the slit and a protruding knob on the adapter body. Loads can be added to the assembly by screwing on metal discs of known weight. The load is, thus, suspended and floats freely between the two ends of the slit. This design offers a convenient means to ensure a constant load contacting the skin. The panelist is only required to maintain the knob approximately in the middle of the slit during the experiment. The measuring principle is as follows. The control console of the Haake RV-1 houses the operating controls, synchronous motor, electrical circuitry, and indicating meters. It drives the measuring head and the meter reading indicates only the torque induced by the frictional resistance to the rotating probe, and not the friction in the trans- mission. Torque is measured by the angular displacement of a creep-resistant torsion spring, mounted between two concentric conical shafts. The displacement angle is converted into an electrical signal by means of a high-precision potentiometer. The voltage output of the potentiometer is linear to the angular displacement of the spring. Thus, the torque exerted on the probe is proportional to the signal registered on the console meter. *Haake Inc., Saddle Brook, N.J.