Oxidation temperatures of WC-Co cemented tungsten carbides

OBRABOTKAMETALLOV MATERIAL SCIENCE Vol. 26 No. 2 2024 Since the density of the substance changes during chemical reactions, the dimensions of the specimen also change. This makes it possible to quantify oxide layers directly by comparing the dimensions of the specimen before and after heating. However, this requires interrupting the heating process and removing the specimen from the furnace. Dilatometers designed to determine the thermal coefficient of linear expansion (TCLE) of a material can also be used to measure the thickness of oxide formations directly during heating [33, 34]. This paper is a continuation of the work of the authors [35] on dilatometry of WC-Co cemented tungsten carbides and its purpose is to determine the temperatures at which the oxidation of these cemented carbides with different cobalt content begins, as well as to establish the relationship of cobalt content with the oxidation temperatures and the dynamics of oxide formation. Achieving this purpose requires studying the dynamics of the WC-Co oxide growth with a wide variation in cobalt content by heating to high temperatures in a laboratory furnace and by dilatometry. Methods Small specimens were heated in a laboratory furnace in air atmosphere to qualitatively assess the dynamics of oxidation. For this purpose, six specimens with cobalt content Co = 3, 6, 8, 10, 15, 20 wt. % were prepared. These specimens were made from commercially produced carbide cutting tool inserts and cylindrical blanks for milling cutters. The composition of WC-Co cemented tungsten carbides corresponded to the VK (WC-Co) group of the Russian Standard [36]. The Fisher grain size of tungsten carbide powders was 4.0–9.0 μm (according to the manufacturer). The specimens’ sizes ranged from 3.5 to 5 mm. In the furnace, the specimens’ were placed on a corundum substrate and heated to 900 °C at a rate of 7 °C/min in air, after which, without holding at the maximum temperature, naturally cooled to 20 °C without removing them from the furnace. The most common and reliable push-rod dilatometer Netzsch 402 PC was used in this work to quantify the dynamics of oxidation. The dilatometer’s push rod and specimen holder are made of alumina ceramics (Al2O3). Tests were carried out in air. The specimen was heated and cooled at a rate of 7°C/min. The contact force of pressing the push rod against the specimen was 0.35 N. To measure the temperature of the specimen, an S-type thermocouple was used, located in close proximity to it. Nitrogen was supplied to its housing at a flow rate of 30 ml/min to protect the displacement sensor. Specimens for dilatometry with a 5×5 mm square cross section were made from the same blanks as the specimens for heating in the furnace. The nominal length of the specimens corresponded to the length of the reference specimen used to calibrate the dilatometer and was 25±0,001 mm. To remove the surface layer obtained during sintering of the workpieces, the ends of the specimens in contact with the holder and push rod were polished with a diamond grinding wheel. The initial length of each specimen was determined at room temperature using a micrometer with a 1 μm scale division in three dimensions. Prior to the experiment, the dilatometer was calibrated at a constant heating rate of 7 °C/min to 1,200 °C using a standard Al2O3 reference specimen. The test specimens were heated at the same rate to 850 °C. One specimen with cobalt content Co = 8 % was heated to 1,150 °C. Cooling to room temperature was carried out at the same rate as heating. Thus, the time of thermal exposure of the specimen was the same for all specimens heated to 850 °C. Simultaneous recording of temperature and specimen expansion was conducted continuously one time per second. The data obtained during dilatometrywere approximated by cubic splines. The characteristic temperatures of oxidation were determined from the inflection points of both lines (heating and cooling) of the thermal expansion curves by finding the extrema of the first and second derivatives. The resulting (after cooling) thickness of the oxide layers was determined using the thermal expansion curves that have a characteristic hysteresis to determine the average oxidation rate of each specimen. The experimental results were processed using standard tools in MS Excel by fetching subroutines specially written in MATLAB with the built-in functions for working with splines using the Spreadsheet Link plugin.

RkJQdWJsaXNoZXIy MTk0ODM1