Degree Name

Doctor of Philosophy


School of Mechanical, Materials and Mechatronics Engineering


The oxidation of high speed steel roll material has been investigated using a high temperature microscope with a CCD camera as a detector, in temperatures ranging from 500 to 700 oC, in both dry and humid atmospheres (up to 12.5% H2O). The whole oxidation process was observed in-situ and recorded. In-situ observations indicate that oxide scales from high speed steel first nucleate at the carbides/matrix interfaces, then spread rapidly outward to cover the vanadium rich MC carbides, followed by a continuous growth over the whole surface. The surface morphology of the oxidised high speed steel is not homogenous due to severe selective oxidation. The temperature at which oxidation commences has a significant influence on the oxidation behaviour of the material because as the temperature increases, the rate of oxidation increases dramatically. The sample surface was oxidised far more severely at 700 oC than at 650 oC in both a dry and humid atmosphere. These results indicate that water vapour in the atmosphere greatly enhances the oxidation of high speed steel, as well as having a significant influence on the morphology of the oxidised surface. Water vapour in the reacting atmosphere reduces the grain size and increases the porosity of the iron oxide. Different types of carbides in the material show large differences in their resistance to oxidation, for example, chromium rich M7C3 carbides strongly resist oxidation while vanadium rich MC carbides oxidise much easier. A TEM investigation of oxide scale formed on the HSS sample indicates that it consists of two sub-layers: a fine grained (Fe,Cr)-rich oxides inner layer and a larger columnar Fe oxide outer layer.

Oxidation tests under a high humidity (46.5% H2O) atmosphere were carried out in the Gleeble 3500 thermal mechanical simulator. The results show that the water content in the atmosphere not only increases the oxidation rate of high speed steel, but it also influences the microstructure and morphology of the oxide scale. The formed oxide scale is non-uniform and more porous and fractured than the compact, uniform layer formed with dry air. With an increasing content of water vapour, more porous oxide scales form and more Fe2O3 phase is formed in the oxide scale.

Nano-indentation tests were used to characterise the mechanical properties of the oxide scale formed on the HSS sample. A nano-indentation was performed on the cross section of the oxide scale with a matrix (5 by 5 and/or 7 by 7) routine. The tests successfully revealed the different mechanical properties of the inner and outer sub-layer. The results indicate that the outer layer with larger columnar Fe oxides was the hardest (13-16GPa), followed by the inner sub-layer (9-11GPa) with fine grained (Fe,Cr) rich oxides, and then the HSS matrix (approximate 9GPa). The HSS matrix had the highest Young’s modulus, around 270GPa, the outer oxide layer was 200-240GPa, and the inner oxide layer was the lowest at 130-180GPa. The mechanical properties of the oxide scale were significantly influenced by the chemical composition and microstructure of the scales. Because these oxides scales have different hardness/modules than the matrix, they contribute significantly to the frictional behaviour and wear resistance.

A high temperature pin-on-disc tribological test was carried out to simulate the contact and tribological behaviours of oxide scale in the roll bite. The pin represents the HSS work roll, and the mild carbon steel disc represents the hot rolled strip. Only the sliding part of the motion was considered, the rolling part was neglected. The results show that a typical tribological test can be sub-divided into three stages: i) stage I corresponds to the start of the friction curve when the coefficient of friction decreases, ii) stage II corresponds to an increase in the coefficient of friction after the minimum value, and iii) stage III is the stabilisation step of the friction. Stage I and stage II can be summarised as a running-in period which lasts less than 300 seconds from the start of the test. A thin, compact, and smooth “glaze” oxide scale around 800-850nm formed on the surface of the pin during the running-in period. This “glaze” oxide layer in the contact zone helps stabilise the friction and protects the pin from wearing. Adhesive wear is the predominant wear mechanism on the pin during this period. At stage III the wear mechanism on the pin becomes complicated. In addition to oxidation on the HSS pin, the oxides transfer from the disc to the pin, which thickens the oxide scale on the surface of the pin quite significantly. Large cracks and pores could be found inside the oxide scale, which indicates that a severe “banding” phenomenon could occur when the oxide scale reaches a critical value. A large amount of wear debris observed on the pin wear track confirmed that abrasion happens at this stage. It is the balance between adhesion, abrasion, and oxidation.

For the first time a mini-rolling mill was successfully incorporated into the Gleeble 3500 thermal mechanical simulator to simulate stalled hot rolling. Two types of roller surfaces, e.g. a relatively virgin surface and a pre-oxidised surface were investigated. The experimental results show that these two surfaces exhibit quite different tribological behaviours. The rolling force and friction of pre-oxidised rolls is always higher than fresh rolls for different reductions and temperatures. With the pre-oxidised roll, a high rolling temperature can cause the oxide scale to peel off from the strip and stick onto the surface of the roll.

FoR codes (2008)




Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.