To cite this version:Urko de La Torre, Aitor Loizaga, Jacques Lacaze, Jon Sertucha. As cast high silicon ductile irons with optimised mechanical properties and remarkable fatigue properties. Materials Science and Technology, Maney Publishing, 2014, vol. 30 (n°12) The present work shows a comparative study regarding the mechanical properties of 25 as cast ferritic ductile iron alloys, nine of them with silicon contents higher than 3?00% and carbon contents lower than 3?60%. In a first step, different carbon equivalent values have been used in order to analyse the effect of this parameter on the mechanical properties. After this comparative analysis, the composition ranges C53?30-3?40 wt-% and Si53?75-3?80 wt-% have been selected as the most proper ones to optimise the tensile and impact properties among the high silicon ductile iron alloys. Finally, a second study was carried out to compare the tensile and fatigue properties of the optimised high silicon alloy with the corresponding ones obtained from an EN GJS 400-18-LT grade alloy with low silicon content. Although the room temperature impact values obtained from the high silicon ductile iron are lower than 6 J cm 22 , the measured fatigue limit of this alloy (358 MPa) is clearly higher than the one obtained from the low silicon cast iron (170 MPa). A discussion about the benefits and advantages of the high silicon alloy is included.

Amongst the most important graphite shapes, nodules and compacted particles are of particular interest as they can coexist in castings with relevant changes in properties. In this context, it has been often reported that the compacted shape is an intermediate form between nodules and graphite lamellas that may be seen as a degeneracy from the nodular one. The present work shows the microstructure evolution of an initially ductile iron alloy with a silicon content of about 2.4 wt.% when reducing progressively the magnesium content by holding a large melt batch in a nitrogen-pressurized pouring unit for 8 hours. Thermal cups with and without inoculant were cast at a regular time interval together with a sample for chemical analysis. Interestingly, the thermal records of the inoculated samples show no significant changes with time while the structure evolved from fully spheroidal to half spheroidal half compacted graphite. Conversely, the thermal curves of the non-inoculated samples showed two arrests, one at nearly the same temperature as for inoculated alloys and a second one at a temperature decreasing with holding time until being below the metastable eutectic temperature. Microstructure observations showed the presence of a limited number of compacted cells which decreases as well with holding time. These observations suggest that these cells start developing during the temperature interval between the first and second arrests, leading to a bulk eutectic transformation either above or below the metastable eutectic temperature. These results support the view that a fully compacted structure can be obtained only with a controlled inoculation which should not be too high to avoid too high nodularity.

The paper introduces a new linear displacement analysis (LDA)/thermal analysis (TA) experimental device for measuring linear displacement during the solidification of cast iron. The experimental device comprises a sand mould encased in a steel shell that prevents mould wall movements. Thus, only the linear displacement caused by the shrinkage or expansion of the metal is recorded by the transducers. Two quartz rods introduced directly at different heights into the liquid metal and connected to two transducers record the linear displacement during the liquid-solid transformation and subsequent cooling. Two thermocouples positioned at the same height with the quartz rods allow for the concomitant TA and LDA and thus for the direct correlation between expansion/contraction and the temperature change during solidification events such as graphite formation. The LDA device was used to study the differences in the solidification mechanisms of irons with different graphite morphologies (lamellar, compacted/ vermicular and spheroidal) at carbon equivalent in the range of 3?7-4?4%. The analysis included the LDA and TA curves and full metallographic characterisation of the cast irons. In general, graphite expansion increased as the graphite shape changed from lamellar, to compacted and then to spheroidal. The most important process variables are the magnesium and carbon contents. Higher Mg residual and C in the iron produced more graphite expansion. Compacted graphite (CG) iron was particularly sensitive to the Mg residual. Indeed, the high Mg CG irons exhibited similar graphite expansion to that of spheroidal graphite (SG) iron, while the low Mg CG iron expansion was closer to that of the lamellar graphite (LG) iron. Graphite expansion increased for all data with the time interval over which graphite expansion occurred. It also increased with both carbon and carbon equivalent. The time for graphite expansion increased noticeably with the carbon content of the iron. It did not depend on the graphite shape. By combining TA and LDA, it was possible to plot the evolution of graphite expansion as a function of the fraction solid and thus to understand the kinetics of graphite expansion. The amount of expansion available at the end of solidification was quantified. Such data, when correlated with process variables, will be useful in decreasing microshrinkage and in producing riserless compacted and SG irons.

Evaluation of solidification kinetics by thermal analysis is a useful tool for quality control of Al-Si melts before pouring provided it is rapid and highly reproducible. Series of thermal analysis records made with standard cups are presented that show good reproducibility. They are evaluated using a Newton's like approach to get the instantaneous heat evolution and from it solidification kinetics. An alternative way of calculating the zero line is proposed which is validated by the fact that the latent heat of solidification thus evaluated is within 5% of the value calculated from thermodynamic data. Solidification kinetics was found highly reproducible provided appropriate experimental conditions were achieved: high enough casting temperature for the cup to heat up to the metal temperature well before solidification starts; and equal and homogeneous temperatures of the metal and of the cup at any time in the temperature range used for integration.

No abstract