STUDY OF THE WEAR BEHAVIOR OF AISI 304 AUSTENITIC STAINLESS STEEL USING RESPONSE SURFACE METHODOLOGY Mar´ ıa Cristina Mor´ e Far´ ıas Surface Phenomena Laboratory - Department of Mechanical Engineering - Polytechnic School of University of S˜ ao Paulo Av. Prof. Mello de Moraes, 2231 - S˜ ao Paulo - SP - 05508-900, Brazil e-mail: maria.farias@poli.usp.br Deniol Katsuki Tanaka Surface Phenomena Laboratory - Department of Mechanical Engineering - Polytechnic School of University of S˜ ao Paulo Av. Prof. Mello de Moraes, 2231 - S˜ ao Paulo - SP - 05508-900, Brazil e-mail: deniol.tanaka@poli.usp.br Amilton Sinatora Surface Phenomena Laboratory - Department of Mechanical Engineering - Polytechnic School of University of S˜ ao Paulo Av. Prof. Mello de Moraes, 2231 - S˜ ao Paulo - SP - 05508-900, Brazil e-mail: sinatora@usp.br Maria Elena Santos Taqueda Department of Chemical Engineering - Polytechnic School of University of S˜ao Paulo Caixa Postal 61548, S˜ ao Paulo - SP - 05424-970, Brazil e-mail: santos.taqueda@poli.usp.br Abstract: Rather than using the wear theories to determine a material wear rate as a function of contact conditions, the response surface methodology (RSM) was used as an alternative to the classical one-variable-at-a-time strategy to describe tribological behavior of materials. An attractive feature of using statistical design technique is that it requires few experimental procedures to set up dependence of the tribological parameters as a function of operating conditions. In this investigation the non-lubricated wear mechanisms of AISI 304 austenitic stainless steel were studied at room temperature. Response surface methodology, using a second-order composite design for two factors was implemented. The independent variables were: applied load and sliding velocity. The wear rate of material was determined on a pin-on-disc machine, with applied load range from 6 N to 20 N and sliding speed range from 0.07 m/s to 0.81 m/s. Scanning electron microscopy (SEM) was used to characterize the worn surfaces and the debris particles morphologies. The use of the experiment design, in addition to the reduction of the number of experiments, the statistical analysis and modeling provided useful and precise information to show the significance of the observed wear trends. A second-degree polynomial was used to represent a curved surface, which fits the experimental data, instead of two different straight lines for the wear rate. The analysis of the response surface for the wear rate and the characterization of worn specimens revealed a change of wear mechanism from oxidative wear to plastic deformation as a function of tangential speed and applied load. Keywords: sliding wear, austenitic stainless steel, response surface methodology 1. Introduction Austenitic stainless steels are used in nuclear reactor and more generally in the hot regions of chemical and power generation plants. These materials are considered to have poor wear characteristics, but its behavior depends on the wear conditions. Various works have been developed to study the dry sliding wear of these materials regarding the operational parameters, load, sliding speed and sliding distance, at different environment conditions. Hsu et al. (1980) studied the wear behavior of AISI 304 and AISI 316 stainless steel blocks sliding against AISI 440C rings, at various levels of load (63, 133 and 200 N) and concluded that the greater tendency to form strain-induced α ′ -martensite on the AISI304 steel was probably responsible for its poor wear behavior compared to the AISI 316 steel. Yang et al. (1985) results confirmed Hsu et al. (1980) observations where the sliding wear of 304 and 316 blocks against M2 tool steel was studied in the same range of applied load. Apart from the α ′ -martensite induced by plastic deformation in the 304 steel, Yang et al. (1985) concluded that the nature of the transfer process and the relative hardness of the contacting bodies also affect the sliding wear behavior of the austenitic stainless steels. Smith (1984) investigated the reciprocating wear of AISI 316 steel on itself as a function of the sliding distance in air at room temperature in the load range 8 to 50 N. A linear increase in the debris amount with the sliding distance was observed. Also, a change in the appearance of wear debris with load from powder debris particles to large thin flakes was observed. In an other work, Smith (1985) investigated the effect of the sliding