Characterization of Impact Toughness Properties of DIN39MnCrB6-2 Steel Grade

Boron added steels in quenched and tempered condition have been widely used in applications where good mechanical properties and low cost are required. In this study, impact resistance of DIN 39MnCrB6-2 steel grade was investigated due to its criticality in automotive components. Metastable phase diagrams were built to guide heat treatments. The influence heat treatment parameters and initial microstructure were evaluated. Impact toughness results showed low absorbed energy (<30J) for all tempering temperatures. CVN impact toughness was substantially increased in samples submitted to solution treatment. Microscopic observation showed coarse borocarbides in surface fracture of Q&T samples. It was concluded that the formation of borocarbides is inherent in boron steels and their coarse morphology should be avoided in order to reduce embrittlement.


Introduction
Since early studies in 1907, boron addition has been widely used to improve steel hardenability with low cost and has played a key role in development of quenched and tempered (Q&T) high strength steels. 1;2 The occurrence of borocarbide M 23 (B, C) 6 , with M = Fe or Cr, is the most remarkable microstructural feature in low alloy boron added steels. First reported in 1954, iron borocarbide Fe 23 (B,C) 6 presents FCC structure (Cr 23 C 6 prototype) and lattice parameter near 10.6 Å. 3;4;5 Previous studies have shown that M 23 (B, C) 6 borocarbides are not completely dissolved in austenitizing temperatures up to 1000ºC. 6 Its orientation relationship with austenite lattice {100} γ || {100}M 23 C 6 leads to low interface energy and delays ferrite nucleation when small borocarbide precipitates are present. 7;8 The occurrence of large borocarbides M 23 (B, C) 6 was reported to promote ferrite nucleation at the interface, explained by loss of coherence and its consequent interfacial energy increase, creating preferred sites for nucleation. 9 Figure 1 shows a schematic diagram for ferrite nucleation at M 23 (B, C) 6 /austenite interface. Hypothetical coherent and incoherent M 23 (B, C) 6 boundaries are shown in (a). Ferrite nucleation occurs preferentially at incoherent interface (b, c) because its higher interface energy. Ferrite phase may grow up to cover borocarbide particles completely (d). 10 Studies conducted in HAZ (Heat Affected Zone) of SAE 10B20 steel grade proposed the influence of cooling rate on M 23 (B, C) 6 borocarbide precipitation. Generally, higher cooling rates delay (or even suppress) borocarbide precipitation. 11;12 Massive M 23 (B, C) 6 occurrence was reported in SAE 10B22 austenitized at 1000ºC and cooled at 0.5ºC/s and 5ºC/s. In order to achieve cost saving, some automotive suspension and transmission components are made of DIN 39MnCrB6-2 steel grade. Impact tests evaluation of this material is chosen in order to better reproduce possible critical application conditions of these components.
The main objective of this study was to evaluate the influence of initial microstructure and heat treatment (solution; quenching and tempering) in the Charpy V-notch resistance of large scale hot rolled DIN 39MnCrB6-2 steel, correlating the results with the microstructural features observed in the samples.

Experimental Procedure
The samples for this study were removed from 155 mm square section bars produced by hot rolling. The chemical composition of the material is given in Table 1. B S represents soluble boron which effectively contributes to hardenability (small borocarbide particles at grain boundaries).
Chemical composition from Table 1 was used as input data for thermodynamic simulations, in ThermoCalc® software with TCFE7 database. Graphite phase was not used in calculations, allowing formation of cementite. Due to the lack of reports of M2B boride occurrence in steels with low addition of boron, M2B phase was not used in calculations as well.
Half sections of rolled bars ( Figure 2) were used to extract samples for heat treatments consisting of Solution treatment followed by Quenching and Tempering as indicated in Figure 3. Austenitizing temperature for quenching was 880ºC based on industrial practices regarding this steel grade.
Solution treatments were carried out in some samples in order to evaluate the influence of prior borocarbide size on CVN toughness. Samples submitted to solution treatment prior to quenching and tempering were identified as S + Q&T and samples submitted only to quenching and tempering were identified as Q&T. Heat treatment parameters are summarized in Table 2.
After heat treatments, blanks were taken from Surface (S), Mid-thickness (M) and Core (C), as illustrated Figure 2b. The blanks from each position were machined to 3 Charpy V-notch specimens according to ASTM E23 -Type A and tested at room temperature. Surface fracture analysis and phase chemical compositions were performed in a Hitachi® TM3000-Tabletop-Microscope with EDS system.

Thermodynamic simulations
Isopleth section with predicted phases were simulated for the alloy in Table 1 without boron, as shown in Figure 4, and with boron, as shown in Figure 5.   It is possible to observe that boron addition significantly changes phase diagram fields. Borocarbides exhibits lower free energy when compared with respective carbides in boron free compound and it results in wider range of stability in boron-containing alloys. Cementite with boron in carbon sublattice, M 3 (B, C), is more stable than cementite and it is predicted up to 1000ºC. Furthermore, iron carbide M 23 C 6 is not stable in boron-free steel and respective borocarbide M 23 (B, C) 6 is stable in boron-containing alloy. The M 23 (B, C) 6 dissolution temperature is about 900ºC for alloy with 0.36wt. %C.
Predicted volume fraction of phases for alloy with boron content is shown in Figure 6. At 880ºC, γ and M 23 (B, C) 6 are predicted in simulated system with volume fractions of 99,9704% and 0,0296 %, respectively.

As rolled samples
SEM analysis of the as-rolled bars have indicated large amount of borocarbide precipitates in the matrix, as shown in Figure 7. Precipitate higher hardness in comparison to matrix is denoted by plastic deformation around the particles. Semi-quantitative EDS analysis proved the existence of carbon and boron in the precipitates, point B in Figure 8

CVN toughness
For the specimens not submitted to solution treatment, tempering does not seem to affect significantly CVN toughness results. All the specimens reached toughness up to 25-30 J with small standard deviation, as shown in Table 3. However, it has been found a tendency for lower CVN toughness as the distance from the surface increases, as shown in Figure 9. Core specimens presented lower CVN toughness.   In order to evaluate effect of tempering temperature in other mechanical properties, hardness was measured on mid-thickness samples and the results are shown in Table 4. Average hardness decreases with increasing tempering temperature but significant difference was obtained only in higher tempering temperature. Q&T 560 560 29.5 (1.2) 19.6 (1.9) 19.6 (2.1) Figure 9. CVN toughness results for samples quenched and tempered at 520ºC (□), 540ºC (○) and 560ºC (Δ).
In agreement with CVN results, SEM analysis of surface fractures have not shown significant differences in microstructures of Q&T 520 , Q&T 540 and Q&T 560 specimens. A SEM fracture micrograph from mid-radius CVN specimen Q&T 520 is shown in Figure 10.
It is possible to observe second phase particles with average diameter of 15µm (indicated as A in Figure 10) in Q&T 520 , Q&T 540 and Q&T 560 specimens. Those particles were characterized as iron borocarbides ( Figure 11). The occurrence of cracks linked to iron borocarbides particles indicated as B in Figure 10 is a strong evidence that M 23 (B, C) 6 /γ interface is the preferred path for crack propagation. It is reasonable to assume that the embrittlement is caused mainly by the occurrence of these coarse borocarbide particles.
The morphology and distribution of borocarbides particles found in Q&T samples is quite similar to those found in as-rolled sample. This suggests that austenitizing in quenching treatment (880ºC/120min) was not sufficient to promote their complete dissolution, which agrees with data from thermodynamic simulation shown in Figure 5.
The influence of a previous solution treatment on CVN toughness results is shown in Figure 12. The main objective of solution treatment was to promote the dissolution of coarse M 23 (B, C) 6 borocarbide. Solution treatment at 1200ºC Table 4. Hardness measurement for mid-thickness samples which were not submitted to solution treatment. Average and standard deviation for 3 indentations.

Hardness (Standard Deviation) HRc
Q&T 520 520 33.5 (0.6)  decreases CVN toughness because it leads to significantly austenitic grain growth. However, solution treatment at 930ºC increased CVN toughness to approximately 50J. This temperature is high enough to promote M 23 (B, C) 6 dissolution and precipitation of small borocarbides particles during cooling. In addition, at this temperature grain growth is not significant. SEM micrographs from surface fracture of specimens S 930 + Q&T 520 have not shown coarse borocarbides particles, as shown in Figure 13, in agreement with previous discussion. Since fracture main micromechanism is similar in samples submitted to solution treatment and samples without solution treatment, borocarbides occurrence in fracture is the main difference between samples submitted to solution and samples submitted only to quenching and tempering.

Conclusions
From the results obtained in this investigation, the following conclusions can be summarized.
1) Small amount of boron significantly changes the equilibrium phase diagram of steel DIN 39MnCrB6-2, allowing formation of M 23 (B, C) 6 borocarbide; 2) Austenitizing temperature of 880ºC did not promote complete dissolution of coarse M 23 (B, C) 6 borocarbides; 3) The presence of large borocarbides leads to low CVN toughness; 4) In samples submitted to solution treatment at 930ºC, higher CVN values were obtained. Charpy impact fracture analyses have not shown large borocarbides, indicating a change in crack propagation path, which can be explained by partial dissolution of borocarbides at 930ºC.