Similarity between Effects of Laves Phase and M6C in Alloy Steels

In this paper some experimental results obtained within the last two decades are summarized and related to the role of Laves phase and M6C in processes of phosphorus segregation in alloy steels on long-term thermal exposures. In the investigations coordinated by the author of this paper, light microscopy (LM), transmission electron microscopy (TEM), energy-dispersive X-ray spectroscopy (EDX), electron diffraction, and Auger electron spectroscopy (AES) were used.


Introduction
In this paper some experimental results obtained within the last two decades are summarized and related to the role of Laves phase and M 6 C in processes of phosphorus segregation in alloy steels on long-term thermal exposures. In the investigations coordinated by the author of this paper, light microscopy (LM), transmission electron microscopy (TEM), energy-dispersive X-ray spectroscopy (EDX), electron diffraction, and Auger electron spectroscopy (AES) were used.

Laves phase and M6C
The formula Fe 2 Mo is usually used to characterize the Laves phase present in Cr-Mo alloy steels. This phase exhibiting hexagonal crystal structure [1] is isostructural with MgZn 2 (Table 1). It precipitates mostly in the form of discrete particles inside the δferrite grains [2,3] or in the form of foils along the boundaries of prior austenite grains in ferrite [4,5]. Particles of Laves phase precipitated in 12CrMoV steel (Table 2) after annealing at 773 K for times exceeding 10 000 h contain besides iron and molybdenum also chromium, silicon and phosphorus as illustrated by EDXspectrum in Fig. 1. Laves phase was not identified in 12CrMoV steel after annealing for shorter times than 10 000 h at 773 K [6].
Parameters characterizing the crystal structure of M 6 C are also given in Table 1. This carbide precipitates mostly in molybdenum and/or tungsten containing steels after long-term annealing [7]. The higher the bulk contents of molybdenum and/or tungsten the shorter the time preceding the M 6 C precipitation [8]. M 6 C forms mostly bulky foil-shape particles along the boundaries of prior austenite grains. The chemical composition of M 6 C in Cr-Mo steels is very similar to that of Laves phase. As follows from EDXspectrum in Fig. 2 [6], M 6 C in 3CrMoV steel ( Mass contents of elements in investigated steels in %. diffraction was used. M 6 C was not found in the steels after annealing at 853 K for 1 000 h or for shorter times [6]. Both Laves phase identified in 12CrMoV steel [5] and M 6 C identified in 3CrMoV steels [6] have some identical properties: • they start to precipitate after longer annealing periods (Laves phase in 12CrMoV steel after 10 000 h at 773 K and M6C in 3CrMoV steels after 5 000 h at 853 K), • molybdenum and iron dominate in chemical compositions of both phases, • both the phases are able to dissolve silicon and phosphorus.

Segregation induced grain boundary precipitation of Laves phase and M6C
Laves phase and M 6 C as Mo-rich phases in Cr-Mo alloy steels nucleate preferencially in areas with enhanced molybdenum con-centration. Such areas are also grain boundaries, because molybdenum shows tendency to segregate in Fe-base materials [9]. The grain boundary segregation of molybdenum however can be strengthened by the phosphorus-molybdenum co-segregation when phosphorus is present in Cr-Mo steels. For 3CrMoV steel, the segregation enthalpy of molybdenum and the P-Mo interaction coefficient are -21 and -23 kJ.mol Ϫ1 [10], respectively. This gives evidence about a strong attractive interaction between phosphorus and molybdenum [11] and reduces a long incubation period exceeding the Laves phase and M 6 C precipitation.

Reduction of phosphorus enrichment at grain boundaries due to the Laves phase and M6C precipitation
The appearance of the Laves phase and/or M 6 C particles at the grain boundaries evokes the redistribution of originally segregated phosphorus between the matrix and the newly formed particles. It is because of the particle ability to dissolve phosphorus and silicon. Consequently, the grain boundary concentration of phosphorus starts to decrease because phosphorus entering Laves phase and/or M 6 C particles diffuse intensively into the particle interior. Driving force of this process is effort of phosphorus to be distributed even across the particle [6]. After precipitation of Laves phase in 12CrMoV and M 6 C in 3CrMoV steels, the grain boundary concentration of phosphorus starts to decrease. This is illustrated in Figs. 3 (for 12CrMoV steel) and 4 (for 3CrMoV steel).
Processes taking part in the investigated steels during annealing contribute to the anomaly in the kinetic dependence of the phosphorus grain boundary concentration. The anomaly, even if it appeared in different kinds of steels, shows the same nature and can be attributed to the same mechanism. The decisive assumption for applicability of this mechanism resides in the similarity of Fe 2 Mo and M 6 C phases. Specific behaviour of these phases in influencing the phosphorus grain boundary segregation in alloy  Fig. 2 EDX-spectrum of M6C particle present in 3CrMoV steel after annealing for 5 000 h at 853 K [6] steels seems also to be interesting from the practical point of view. Changes in phosphorus concentration evoked by precipitation of the Fe 2 Mo and/or M 6 C phases should also show a direct impact on the steel intergranular brittleness. A study of the relationship between the phosphorus enrichment at grain boundaries and fracture toughness of alloy steels containing M 6 C or Laves phase is therefore highly topical [12].

Conclusions
It has been shown that the Laves phase and/or M 6 C particles precipitating along the boundaries of prior austenite grains can evoke the anomaly in the kinetic dependence of phosphorus grain boundary concentration. This is due to the ability of both phases to dissolve non-metallic elements such as phosphorus or silicon. In comparison to other phases usually present in Cr-Mo alloy steels, Laves phases and M 6 C precipitate after longer periods of annealing. A subsequent partitioning of originally segregated phosphorus between the matrix and the newly precipitated particles of Laves phase and/or M 6 C leads to the decrease in the phosphorus grain boundary concentration.