Rare-Earth-Metals (REM) treatment

Rare-Earth-Metals (REM) treatment

Recently, the interest in the application of Rare-Earth-Metals (REM) for modification of NMI in different steel grades has increased sharply. A large number of experimental works (including laboratory experiments and industrial trials) have been carried out by different researchers. Here, a REM element is often described to have a high affinity to harmful impurities in steel such as O, S and N. Therefore, the influence of REM on the final properties of steel products corresponds to a reduction of the soluble contents of these harmful impurities in the steel. This is due to the formation of non-metallic inclusions with required characteristics. Even though the melting temperature of REM elements is comparably low (798–1016 °C), the melting temperature of the formed non-metallic inclusions varies in the ranges from 1690 to ~2291 °C for oxides, from 1940 to 1990 °C for oxy-sulfides, and from 1795 to 2450 °C for sulfides. When REM (lanthanum, cerium, praseodymium, neodymium, etc.) are added into the melt as a mischmetal or other REM-alloys, the oxides, oxy-sulfides and sulfides of the REM elements are formed as solid particles in the liquid steel. It can therefore be expected that most of the sulfur will react with REM in the melt. The precipitated REM oxy-sulfides and sulfides are small sized (about 0.5–3.0 µm) and more homogeneously distributed in the solidified steel compared to large size MnS inclusions (Type II and Type III). The latter precipitate in the last parts of a solidified steel.

Only REM2O3 will precipitate in the melt if the ratio of [S]init/[O]init is smaller than 10 (point A). During precipitation of REM2O3 oxides, the content of oxygen is reduced without a change of the S content to the point B. The REM2O2S (REMO2·REMS) oxy-sulfides will precipitate in the melt between the points B and C. The concentrations of O and S will decrease in the melt according to the stoichiometric ratio of O and S in the precipitated oxy-sulfide. Then, the REM oxy-sulfides and sulfides will precipitate together simultaneously and the corresponding concentrations of O and S will change according to the line CE. If the [S]init/[O]init ratio is in the range from 10 to 100 (point F), the REM oxy-sulfides (FG segment) and oxy-sulfides + sulfides (GH line) will precipitate in the melt. If the [S]init/[O]init ratio is larger than 100 (point K), the REM sulfides will precipitate to a cross section with the 1/100 line without changing of the O content in the melt with a following simultaneous precipitation of oxy-sulfides and sulfides. The final contents of the dissolved O and S, [O]final and [S]final, depend on the content of added REM. As a result, the non-metallic inclusions in the steel after REM addition can have a complex (multiphase or multilayer) structure (Figure 18b): oxide or oxy-sulfide core covered by one or several layers of oxy-sulfides and sulfides. According to previous work [48], the composition of sulfide phase can change from REM2S3 to REM3S4 and further to REMS, as the sulfur content in steel decrease. Sulfides of type REM3S4 and REMS predominate in the most industrial steel grades. However, if the rest content of soluble REM in the liquid steel decreases below some critical value due, to the reoxidation or reaction with refractories and slag, some amount of S can be recovered from REM sulfides and oxy-sulfides. In this case, the MnS inclusions can precipitate during solidification of the steel melt. A thermodynamic evaluation regarding the formation of different REM inclusions in the liquid steel is limited because of the lack of reliable data. For instance, the values of REM activity calculated based on thermodynamic data given by different authors for a reaction of REM and S may vary 10–1000 times. This big difference can be explained by the various conditions of experiments and by some other reasons.
The effect of REM additions on the mechanical properties of different steel grades are reported in many publications. For instance, it was reported in reference that the impact strength in transverse samples can be increased two and more times at a ratio of added REM and S contents of about 3 (%REM:%S = 3–4 or %Ce:%S = 1.5–1.7), at which the formation of MnS inclusions was avoided. Figure 19 shows values of the impact strength in longitudinal (LS) and transverse (TS) samples as a function of the ratio of REM and S contents (in mass-%) in steel with additions of mischmetal or REM silicide.

According to results obtained by Ha et al., an addition of mischmetal up to a value of 0.067% REM (at %REM/%S = 3.7) in 25% duplex stainless steel with sulfur contents of 0.016%–0.028% leads to a significant decrease of the size, area fraction and number of oxy-sulfide inclusions per unit area in steel. As a result, a resistance to pitting corrosion increased by ~34% at a REM content of 0.067%. However, further additions of mischmetal up to 0.078% REM (at %REM/%S = 4.9) decreased the resistance to pitting corrosion. This was caused by a significantly increased number and area fraction of oxy-sulfides and a change of the inclusion shape from an angular or granular shape to a needle-like shape. Wang et al. also reported that the addition of an appropriate amount of REM alloys (0.014%–0.081% REM) in various advanced low-alloyed steels (14 MnNb, X60, 10 MnV, etc.) with a 0.008% S content for modification of inclusions, resulted in a deep purification and refinement of the grain size. This led an increased strength and toughness of these steel grades. Moreover, the corrosion resistance of weather resisting steels is also improved. For instance, the corrosion rate decreased on average by 17%–54% when adding 0.029% REM (%REM/%S ~ 2.0–3.6).

Among the many beneficial effects of adding REM alloys to steels, significant improvements in ductility, transverse impact strength, susceptibility to lamellar tearing in welding and bend formability have been reported. Kang and Gow have also reported an enhanced impact strength in REM treated rail steels. Moreover, they found that the REM treatment has made a significant improvement of the fatigue strength of the axle steels (0.02% S). In addition, an effort has been made to understand this improvement through the shape control of sulfide inclusions in steels. It was found that small REM inclusions were less active in both crack initiation and propagation of the fatigue fracture, compared to large MnS inclusions.

However, published experimental data obtained from laboratory experiments and industrial trials are very scattered and often contradictory. It can be explained by the imperfect technique of a REM addition, the variation of yields of an added REM and by insufficient control of the concentrations of REM, Al, O and S in liquid steel. Up to now, the optimal amount of added REM has been determined experimentally for various steel grades in different companies. Furthermore, for a given equipment and technology of steelmaking.

An improved machinability of the re-sulfurized free-cutting steels by modification of non-metallic inclusions due to addition of REM has been reported in previous work. In that work, 0.027% to 0.050% REM was added with and without similar levels of Ca. It can be observed in Figure 20 that the flank wear can be decreased by an average of 32%–34% for trials with Ca-addition, by 41%–43% for trials with Ca and REM additions, and by 49%–54% for trials with REM-addition in comparison to the reference steel (without Ca and REM additions). It was found that the flank wear decreased significantly with increased REM contents in the steel. From another side, the tensile strength of the experimental steels was reduced by only 1%.

Reference: Niclas Anmark, Andrey Karasev and Pär Göran Jönsson, The Effect of Different Non-Metallic Inclusions on the Machinability of Steels, Materials 2015, 8, 751-783.