Effect of Rare Earth Elements on Chill Formation and Nodule Count in Thin-section Spheroidal-Graphite Cast Iron

It was found that there was a close relationship between chill formation and nodule count in spheroidal-graphite (s-g) cast irons and, further, that there was a critical nodule count above which chill did not appear. The relationship between the critical nodule count and the cooling rate of specimens was expressed in the following quadratic equation:

N=0.58R²+19.07R+1.01

where N is the critical nodule count (mm^?2) and R is the cooling rate (°C/sec). Nodule count was increased by adding small amounts of a rare earth in combination with small amounts of calcium. Some heterogeneous phases with diameters of 1 μ or more were observed at the centres of graphite spheroids. These phases were composed of spherical rare-earth sulphides containing MgS, CaS and MgO. It is suggested that rare-earth sulphides contribute as a substrate on which graphite spheroids are formed in thin section cast irons, containing rare earth elements.     

Some Properties of a Boronised Spheroidal-graphite Cast Iron

Abstract

Boronising treatment of spheroidal graphite cast iron was investigated by means of electrolysis in a molten salt mainly consisting of B₂O₃, K₂O and Na₂O at various temperatures and over different periods of time. The microhardness and the thickness of boride layers were measured and the distributions of B, Si and C on the surface of specimens were observed by an X-ray microanalyser. Microscopic examination and the results of X-ray microanalysis showed that the boride layer consisted of two layers; an outer FeB layer and an inner Fe₂B layer. The microhardness of these boride layers were approximately HVN 1500–1800. The thickness of the boride layer increased in proportion to the square root of treatment time at constant temperature. The activation energy for diffusion of boron in the specimen was 19.5 kcal/mol, as obtained from the slope of Arrhenius plots.

A Si rich layer was formed in a region outside the boride layer in specimens after boronising. Moreover, graphite formation was observed in the Si rich layer in specimens boronised at austenitising temperatures for a prolonged time. Graphite formation in the Si rich layer can be considered to be a consequence of the precipitation of carbon, during the cooling process after boronising, which diffused from spheroidal graphite to the austenite matrix at the boronising temperature. Large amounts of graphite formed in the Si-rich layer resulted in a crack passing through these graphite particles.     

Influence of Bismuth on Graphite Nodule Count in Thin-section Spheroidal-graphite Cast Iron

Abstract

Cast irons containing 3.8% C and 2.2% Si were melted in a graphite crucible using a high-frequency induction furnace. Nodularisation treatment was carried out at 1530°C by the addition of 1.6% Fe-Si-Mg alloy and various amounts of bismuth as metallic bismuth, bismuth oxide and bismuth sulphide. When the magnesium reaction subsided, the melts were post-inoculated with Fe-Si and then poured into CO₂ Process sand moulds. The occurrence of chilled structure and graphite-nodule counts were determined using an optical microscope and an image analyser. X-ray microanalysis and emission spectrochemical analysis were carried out to detect Bi in the graphite nodules and also in nodules extracted with dilute hydrochloric acid. The nodule count was enhanced by the addition of extremely small amounts of metallic bismuth, bismuth oxide and/or bismuth sulphide. Some heterogeneous phases containing Bi, Mg, Ti, Fe and Si, with diameters of 1 μm or less were observed at the centres of graphite spheroids using back-scattered electron imaging and an energy dispersion X-ray microanalyser. Graphite crystallises on these phases during eutectic solidification. It is suggested that large numbers of fine liquid bismuth particles contribute to the formation of graphite spheroids by acting on the substrate.     

Fluidity and solidification microstructures of flake graphite cast iron in thin sections

Fluidity determinations have been undertaken using a silica tube as the fluidity channel, metal being drawn in to the test tubes by means of a regulated vacuum system. Fe-Si and Ca-Si inoculants were added to a hypoeutectic cast iron melt which was then tested for fluidity. A linear relationship between fluidity length and tube diameter was obtained. Fluidity length increased with increasing suction pressure and with increasing temperature of the molten iron. Fluidity length is proportional to the square root of effective suction pressure. Structures within the fluidity specimens have been classified, from the tip backwards, into three zones, containing ledeburite, mottled and flake graphite structures. The length of each structural zone increases with increasing suction pressure. The total length of the ledeburite zone, plus the mottled zone, is constant for increasing tube diameter. Thus, the flake graphite zone increases linearly with tube diameter. The effects of inoculation on the structure and its fading were evaluated by the length of the flake-graphite zone.     

Effect of Eight Individual Rare-earth Elements on the Graphite Nodule Count of a Thin-section of Spheroidal-graphite Cast Iron

Abstract

Eight rare-earth elements, Y, La, Ce, Pr, Nd, Sm, Dy, and Yb, were added individually to two irons, one containing 0.02% S and the other 0.08% S. After nodularisation, the metal was poured into CO₂ Process sand moulds to produce stepped test-specimens with thicknesses of 3,6 and 9 mm and into shell moulds to provide ‘break-off’ type specimens with a dia of 30 mm. The effect of each of these rare earth elements on the nodule number, and chill formation was investigated and the mechanism of the nodule number increase studied.

The increase in the nodule number differed significantly with each kind of rare-earth element added to the molten iron. The largest rise in the nodule number was with Ce followed in turn by La, Pr, Nd, Sm, Yb and Y. It was seen that the increase in nodule number was more pronounced when the elements were added to the iron with the higher sulphur content. To maximise the increase in nodule number, it was found that the optimum added quantity of these rare-earth elements was almost stoichiometric with the sulphur content of the iron.

X-ray microanalysis of the prepared specimens containing these rare-earth elements confirmed that only sulphides formed in the melts treated with Ce, La, Pr, Nd, or Yb, whilst both sulphides and phosphides formed when additions were made of Dy, Sm or Y. The specimens poured into a metal mould during the eutectic solidification stage, and subsequently quenched showed that graphite had formed on the rare-earth sulphide substrates, but not on the phosphide. It was concluded that the latter cannot act as substrates for the formation of graphite.

Measurement of the average diameters of the rare-earth sulphides (Ce, La, Pr, and Nd) showed that the largest effects upon nodule numbers were associated with elements producing sulphides of smaller diameter. This phenomena may reasonably be explained. A rare-earth sulphide exists as a liquid phase in an iron melt and aggregates until it solidifies, increasing its diameter. The degree of aggregation differs amongst the different rare-earth sulphides, resulting in a different number of substrates for each graphite formation.     

Removal of manganese from molten cast iron using potassium sulphide

Small melts of Fe-C-Mn (1.5%) alloy were treated with up to 20 wt% of potassium sulphide to assess its effect on the removal of manganese. The manganese and sulphur contents were determined on chill-cast samples, whilst microscopic observation and electroprobe X-ray microanalysis (XMA) procedures were used to analyse the reaction products. It was found that up to 80% of the original Mn content was removed by an addition of 7.5 wt% of K₂S; the rate of removal from molten iron into the slag was very high. Large quantities of polygonal reaction products and blow holes with diameters above 200 mm were found in the quenched samples. The reaction product was shown to be manganese sulphide and evidence indicates that this is assisted to the surface of the melt by bubbles of potassium vapour.