The Effect of Heat Treatment Atmosphere on Hardening of Surface Region of H13 Tool Steel ()
1. Introduction
Heat treatment is an important process to the manufacturing industry, as the mechanical properties of metals can be improved in various ways during the process. This modification has significant influence on the performance of the die material [1]. There are different ways to perform heat treatment hardening, such as pack carburization, gas carburization, vacuum heat treatment, induction heat treatment, and salt bath [2]. Heat treatment is a process consisting of four main stages: preheat cycle, austenite formation stage, quenching and tempering. The target of heat treatment hardening is to harden the material by changing the structure from austenite structure, which is large, shape-edged, coarse and irregular structure to martensite structure, which is fine grain structure of hardened carbide.
In the world of aluminium extrusion industry, heat treatment of steel plays a major role on the determination of process efficiency and product quality. This is because the die, which is made of tool steel, must undergo a series of heat treatment processes to obtain the desired properties. As the extrusion die covers around 35% - 50% of the total manufacturing cost [3], it is essential to obtain thorough understanding on the effectiveness and the kinetics of the die heat treatment, so precise process design can be achieved with a good quality control.
Bjork et al. [4] stated the main issue of extending the service life time of extrusion die is by delaying the removal of surface layer. Surface coating technology is always applied on H13 steel as extrusion die material,to achieve better wear and corrosion resistance to counter the consequence of exposure to severe mechanical, chemical and thermal conditions during the extrusion process. Nowadays, one of the most common surface hardening technologies applied on extrusion die material is nitriding [5]. Nitriding of steel should be conducted at temperature around 500˚C to give the highest diffusion rate of nitrogen in steel [6]. Normally, the nitrated layer should not be thicker than 0.3 mm because it increases its brittleness and lowers the thermal fatigue resistance [7].
Many researches have been conducted related to the heat treatment of steel (see e.g. [8-11]); however, understanding of the decarburization during heat treatment is still limited, especially for H13 tool steel. Although Arain [12] investigated the difference between the open atmosphere heat treatment and the vacuum heat treatment, his focus was mainly on the toughness behaviour and the effect of the atmosphere condition on the hardness of the H13 tool steel is still not clear.
Although the cost of the heat treatment process is only a minor portion of the total production cost, it is arguably the most important and crucial stage on the determination of material quality. This paper investigates how the surrounding condition during heat treatment process influences the material hardness profile. The influence of carbon content of the quenched material on the response of the tempering and the performance of the nitriding is also studied. Samples of the H13 steel with specific sizes would be subjected to heat treatment process with different duration time and under different atmospheric conditions. Hardness profile of each sample would then be measured. It is also of interest to investigate the difference in effectiveness of the gas nitriding process on the samples heat treated without atmospheric control and the samples heat treated with atmospheric control. The heat treated samples would further be subjected to nitriding case hardening process with hardness profile being measured and compared.
2. Experimental Procedures
The four different heat treatment and atmospheric conditions investigated in this study are: 1) heat treatment without atmospheric control, 2) heat treatment with stainless steel foil wrapping, 3) pack carburizing heat treatment, and 4) vacuum heat treatment. Further treatment would also be conducted to investigate the effect of carbon content on the efficiency of the nitriding case hardening process. After quenching, the samples were subjected to two tempering processes followed by gas nitriding. Between each process, a sample was collected for analysis. Table 1 lists the summary of the experimental plan.
Specimens with size of 20 × 10 × 60 mm3 were cut from an H13 circular log which was provided by EXCO Limited, NZ under annealed condition, with the initial hardness of ~12 HRC. The circular log was divided into six equal sections and four of them were used for this research. Each section was dedicated to one heat treatment atmosphere condition as mentioned above. Metal strips with thickness of 10 mm were machined from each section and rectangular samples of size 7 × 10 × 60 mm3 were then sectioned from the metal strip. All specimens were then surface machined and the new specimen dimensions were then measured to ensure similar surface finish as industrial practice and to produce fairly flat surfaces for carbon diffusion modelling.
For the heat treatment without atmospheric control, the specimen was heated in a muffle furnace, at austenitizing temperature of 1020˚C for a specified time period. The samples were positioned at the centre region of the muffle furnace and were in direct contact with the surrounding atmosphere. For this atmospheric condition, carbon in steel could freely react with the ambient atmosphere. An electrical heated open atmosphere furnace (muffle furnace) was used for all heat treatment processes except vacuum heat treatment process. Data logger with a thermocouple was used to monitor and ensure the right treatment temperature was maintained during the process.
In the heat treatment with stainless steel foil wrapping, the specimens were fully wrapped with a piece of stainless steel foil to reduce the rate of chemical diffusion between the specimen and the furnace atmosphere. This method is commonly used in industry and the suggested wrapping procedures can be found in Bryson [13]. For this research, each sample was first wrapped with the long side (the length) double folded, then double folded inwardly from the other two ends (the widths). This experiment setting aimed to minimise the continuous carbon reaction and oxidation between the sample and the ambient atmosphere by the existence of stainless steel foil. The stainless steel foil acts as a barrier to restrict the carbon reaction between the specimen and the surroundings.
In pack carburizationheat treatment, a steel box holding a specimen was fully packed with charcoal with case hardening crystal, barium salt, chemical formula of Ba(ClO3)2 and was heated to a temperature of 1020˚C. The specimen is located at the centre of the steel box and is fully covered by barium salt, so each specimen surface is in contact with the same carburized atmosphere condition.
The vacuum treatment was conducted in an Abar vacuum furnace at approximate 25 microns and preheated at temperature of 650˚C and 850˚C. Each preheating stage took 1 hour. Then it was heated up to 1040˚C and held for either 60, 90 or 120 minutes, and finally cooled to room temperature in a rate of 30˚C per minutes.
Once the austenitizing time is reached, the specimen must be rapidly cooled from the austenite state to the room temperature to form martensite. Two different cooling methods were applied with the first three atmospheric conditions, i.e. fan cooling and water quenching. For the fan cooling, the specimens were taken out from the furnace and were cooled in front of a running fan. The specimens were kept rotating so the cooling rate would be even on all surfaces. In the water quenching, the specimens were put into a pool of water, and kept stirring in the water for 2 minutes. Due to practical difficulties, the vacuum heat treated samples were only cooled in the vacuum furnace with 2 bar of nitrogen gas and the cooling rate of 30˚C/minute. After the cooling, the specimen dimensions were measured again to look for the size changes during the process. A small sample with the size of 7 × 10 × 10 mm3 was then cut from each quenched specimens for hardness test and metallographic analysis.
The remaining part of the specimens was then subjected to two tempering processes which were held at temperature of 540˚C and 595˚C respectively in a vacuum furnace for four hours. To investigate the dynamics of the carbon content on the efficiency of case hardening by gas nitriding, the last part of the remaining treated samples were cut into three different pieces and subjected to once, twice or thrice times of nitriding case hardening process. Samples from pack carburization experiment were not subjected to case hardening process because this is not a usual practice in industry. The gas nitriding process was conducted at 530˚C under controlled atmosphere for 6.5 hours.
The microhardness test method used for this research was Vicker’s hardness test and the load applied was 300 gf. Hardness measurements were conducted from the sample edge to the centre of samples, which was approximately 5000 µm from the edge using the Vicker’s microhardness machine. The hardness measurements were measured in step of 50 µm until 1000 µm, with one extra measurement at 20 µm. For regions between 1000 µm and 5000 µm, the measurements are measured in step of 250 µm. After microhardness tests, the samples would be subjected to surface polishing again and were etched with Nital solution, 3% HNO3 in ethanol. The polished surface was washed with alcohol and dried with warm air immediately after etching, to expose the microstructure details. Metallographic pictures would be taken for the measurement of the depth of the carburization/decarburization layer or the case hardening layer, and for the microstructure examination.
3. Results and Discussions
Hardness profile of each heat treatment stage, including as quenched state, the first tempered state, the second tempered state, and all of the three nitrided states are presented in this section. It must be noted that although the microhardness tests were conducted with Vicker’s measurement, during the process of analysis, the data was converted into Rockwell (HRC) scale. It is because Rockwell scale is the common scale used in steel industries.
Figure 1 shows the hardness profile for all samples heat treated without atmospheric control. Hardness decrease can be found towards the surface region of all samples heat treated without any atmospheric control. The hardness at the region 100 μm underneath the surface increases progressively, then the hardness slowly increases toward the constant state. The figure shows that samples cooled by water generally have higher hardness (54 - 57 HRC) than samples cooled by fan air (53 - 54 HRC). The decarburized layer is found to be thicker as treatment time increases. Another notable difference is that the surface hardness (20 μm below sample surface) of the fan cooled samples is lower than those of the water quenched samples. The fan cooled samples had a surface hardness of 2 - 10 HRC compared to the surface hardness of 14 - 22 HRC for the water quenched samples.