JOURNAL OF MATERIALS SCIENCE 32 (1997) 4493 — 4500 Mechanical property characterization of a number of polymers using uniaxial compression and spherical tipped indentation tests G. KOURTESIS, G. M. RENWICK, A. C. FISCHER-CRIPPS * Departments of Materials Science and * Physics, University of Technology, Sydney Broadway, NSW 2007, Australia M. V. SWAIN CSIRO Division of Applied Physics, Lindfield NSW 2070 and Department of Mechanical Engineering, University of Sydney, NSW 2006, Australia A number of thermoplastic polymers have been tested in uniaxial compression and by monitoring the force—displacement response of small spherical indenters loaded into the surface. The compression tests were performed using a conventional universal testing machine while the very small diamond indenters, of nominal 20 and 50 m radius, were impressed only a few micrometres into the surface with a high precision micromechanical probe. Results from the two approaches are compared and discussed in terms of the proposed relationships between hardness or contact pressure and yield stress for the considered polymers. 1. Introduction Indentation has long been used as a simple means for estimating the mechanical properties of metals. Hard- ness or contact pressure, about which international standards have been established, is often used as a means of defining the plastic penetration resistance or yield stress of a metallic material. In recent years there have been significant advances in the measuring instruments so that with modern micromechanical probe systems, contact pressure or hardness as well as elastic modulus measurements may be made at sub- micrometer or even nanometer dimensions [1, 2]. Although hardness standards exist for rubbers and polymers, they are not a widely reported property. This arises because of the difficulty of accurately measuring the residual impression, although some test procedures such as the Durometer (Shore A, etc.) and Barcol [3] do monitor the change in depth of penetra- tion of an indenter for an increment of force, from which a measure of ‘‘hardness’’ is obtained. In the case of elastomeric materials, the latter value is more in- dicative of the modulus than the plastic response. A critical review of this topic has recently been pre- sented by Briscoe and Sebastian [4]. Ion et al. [5] recently published a study of the behaviour of an amorphous and drawn polyethylene terephthalate (PET) upon indentation with a micro- mechanical probe using a corner cube pointed inden- ter, with an apical angle of 35 °. Despite such a sharp indenter these authors found only a minor contribu- tion to the penetration depth caused by creep. The strong temperature sensitivity of the visco-elastic be- haviour, particularly at temperatures approaching the glass transition temperature, ¹  , often leads to difficul- ties in interpreting force— displacement curves gener- ated using micromechanical probes particularly for the determination of hardness and modulus. Ni et al. [6] have applied finite element methods to assist with the interpretation of such force— displacement curves for gelatin based films widely used for photographic purposes. The use of hardness or contact pressure to deter- mine the yield stress of materials has had a long history for metals and some involvement but less independent verification for polymers, ceramics and inorganic glasses. Tabor [7], summarizing a wealth of existing data, showed that for ductile metals the hard- ness, H, and yield stress,   , could be related by the simple relationship H"C  (1) where C2.8, is a constant. Subsequent research by Marsh [8] attempted to extend this type of relation- ship to less ductile materials such as glasses, for which no other simple measure of yield stress was available. Marsh utilized a spherical cavity elastic—plastic solu- tion developed by Hill [9] to relate the hardness and yield stress for such materials. Later Hirst and Howse [10] extended the concept to that of indentation of polymers with wedges of different included angles. Both Marsh and Hirst and Howse suggested that for materials with high hardness to modulus (H/E) ratios 0022—2461  1997 Chapman & Hall 4493