With development of petrochemical industry, output of plastic products has increased rapidly. Most plastic products need to be formed by molds, and plastic mold steel has also developed rapidly. Currently widely used pre-hardened plastic mold steel such as Swedish 718 steel and German 2738 steel are developed on the basis of American P20 steel. This type of plastic mold steel is transported from steel mill to mold manufacturing enterprise in quenched and tempered state, and can be used directly after processing, avoiding decarburization and deformation hazards caused by subsequent heat treatment.
As a result of quenched and tempered steel, a martensitic structure is formed, a large amount of dislocations are present in the structure, carbon is in a supersaturated state, residual austenite is present in the steel, quenched steel structure is unstable, but spontaneously changes to a stable structure state. This kind of transformation must rely on diffusion of atoms in order to proceed, and different changes will occur depending on tempering conditions.
During high-temperature tempering process, martensite structure recovers, dislocation density decreases, cementite rapidly coarsens and softens significantly. All factors affecting atomic mobility and diffusion rate will affect transformation of martensite structure. Therefore, in addition to influence of tempering temperature, transformation of tempering structure is also affected by rate of tempering heating and holding time.
In order to obtain a high-strength and high-toughness steel by tempering, it is necessary to obtain a uniformly distributed cementite. Some literatures have pointed out that use of rapid heating and tempering can more effectively improve toughness of steel, and inhibit formation of coarse film-like cementite along original austenite grain boundary. It is believed that size of cementite is considered to be refined by rapid heating to tempering temperature.
At present, there are many researches on influence of tempering temperature and holding time on microstructure and properties of prehardened plastic mold steel, but influence of tempering heating rate on microstructure and properties during tempering has not been fully explained. Effect of two different tempering heating rates on microstructure and properties of 3Cr2MnNiMo steel is studied in this experiment. Purpose is to understand effect of tempering heating rate on tempering process and to provide a reference for formulation of a reasonable tempering heating process.

Test steel was smelted by a vacuum induction melting furnace and cast into steel ingots, and dimensions after forging were 75mm*75mm*460mm. Main chemical components are shown in Table 1. After forging, 25mm*15mm*95mm steel block was taken along drawing direction for quenching. Quenching temperature was 910℃ and oil was cooled to room temperature after holding for 1 hour. A cylindrical specimen with a diameter of D4mm and a length of 10mm was taken from core of quenched steel block and subjected to a tempering test on a DIL805 induction heating thermal expansion phase change instrument. Tempering heating temperature was 660℃, tempering heating rates were 1℃/s and 300 ℃/s, holding time is 1s and 1800s respectively. After tempering, samples are cooled to room temperature at a cooling rate of 3℃/s. After tempering, a metallographic sample was taken from core of cylindrical sample. Corrosive solution was a 4% nitric acid alcohol solution. Microstructure and morphology were observed using a JEOLJXA-8530F field emission electron probe (EPMA). Tempered electron backscatter diffraction (EBSD) sample is electrolytically polished to eliminate surface stress. Polishing solution is a 10% perchloric acid alcohol solution. Vickers hardness test was performed on an FM-700 micro hardness tester with a load of 200 g and a load time of 10 s. Hardness value at each location is average of 8 tests.

 

 

Fig.1 Morphology of electron probes with different tempering heating rates when tempering at 660℃ for 1s

Figure 1 shows morphology of electron probes with different tempering heating rates when tempering at 660℃ for 1s. When test steel was tempered at 660℃ for 1s, matrix structure was still dominated by lath martensite in quenched state, and microstructure at a heating rate of 300℃/s had more obvious lath-like characteristics. A large amount of carbides were precipitated at two different heating rates. At a tempering heating rate of 1℃/s, carbides precipitated along lath boundary of martensite and grain boundary of original austenite in a needle shape. At a higher heating rate (300℃/s), a large number of fine short rod-shaped carbides are precipitated on martensite matrix. These carbides not only precipitate at boundaries of martensite lath, but also exist inside lath to make carbide distribution more uniform. Comparing results of Figures 1 (c) and (d), under higher tempering heating rates, dispersion of carbides is greater, size of carbides is smaller, precipitation behavior in slab is a characteristic precipitation behavior unique to rapid heating and tempering.

 

 

Fig. 2 Morphology of electron probes with different tempering heating rates when tempering at 660 ℃ for 1 800 s

Morphology of electronic probes with different tempering heating rates at 660℃ for 1800s is shown in Figure 2. Due to higher tempering temperature, with extension of holding time, orientation of original martensite obtained by quenching has become blurred, structure characteristics of lath martensite have weakened, forming tempered martensite. A large number of fine band-shaped, short rod-shaped, and granular carbides are distributed on matrix of tempered martensite. At low temperatures, morphology of carbides is mainly strip-shaped and short-rod-shaped. As heating rate increases, carbides appear granular and ellipsoidal.
It can be seen from comparison between Fig. 2 and Fig. 1 that carbides are obviously grown at this time. Size of carbides obtained by rapid heating and tempering is still smaller than that of carbides obtained by low temperature heating and tempering. With increase of holding time, aggregates grow up, but carbides obtained by rapid heating and tempering are still uniformly dispersed.

 

 

Fig. 3. Local average misorientations (LAM) of electron backscatter diffraction for different tempering processes.

Figure 3 shows results of electron backscattered diffraction (EBSD) for different tempering processes. In FIG. 3, dark color indicates a region with a lower alignment difference, light color indicates a region with a higher alignment difference. Relative to initial quenching conditions, recovery during tempering process reduces local orientation difference. It can be seen from Fig. 3 that reducing heating rate or extending holding time will exacerbate reduction of local orientation difference. When low-temperature heating and tempering are used, effect of extending holding time on poor orientation is not obvious. When rapid heating and tempering, extending holding time, areas with lower orientation differences increase significantly. Comparing Fig. 3 (a), (b) and Fig. 3 (c), (d), it is found that area of high-orientation difference area obtained by rapid heating and tempering is larger, which indicates that under rapid heating conditions, dislocation density reduction caused by recovery of matrix structure is smaller than low heating rate at same tempering temperature. At this time, carbides are precipitated from dislocations inside martensite lath and become new nucleation sites [20,21]. With increase and change of nucleation position, precipitation of carbides becomes more diffuse, so the smaller average size of precipitated carbides, the greater amount.

 

(A) Microhardness

 

(B) Dislocation density

Figure 4 Changes in microhardness and dislocation density with heating rate and holding time during tempering at 660 ℃

Microhardness values of different tempering heating rates and holding times are shown in Figure 4 (a). As can be seen from Figure 4 (a), when tempering temperature is 660℃ and holding time is 1s, heating rate is increased from 1℃/ s to 300℃/ s, hardness of sample is increased by 53.3HV. At this time, effect of tempering heating rate on microhardness is significant; when tempering holding time is extended to 1800s, microhardness of different tempering heating rates decreases, microhardness of 1℃/ s decreases by 55.5HV, microhardness of 300℃/s decreased by 99.4 HV. With extension of tempering and holding time, martensite recovers, hardness value decreases more rapidly under condition of rapid heating, but hardness value at this time is still higher than hardness value of low speed heating and tempering. Figure 4 (b) shows dislocation density calculated from EBSD results. As can be seen from Figure 4 (b), dislocation density at 300℃/ s for 1s is greater than that for 1℃/ s for 1s. As holding time was increased to 1800s, dislocation density of sample at 1℃/ s did not change significantly, but dislocation density of sample at 300℃/ s decreased slightly. Change in dislocation density corresponded to change in microhardness, indicating that when tempering at a low heating rate, lath martensite recovery occurs during heating process, while during rapid heating and tempering process, matrix recovery is suppressed, and recovery occurs only in subsequent heat preservation stage, which causes dislocation density to decrease and hardness of matrix to decrease.