Mechanical Properties in Thin-Wall Ductile Iron Castings.

By evaluating the variables affecting the mechanical properties of thin-wall ductile iron castings, foundries can reduce carbide formation and improve casting quality.

Applications for ductile iron have increased steadily since its development in the 1940s due to its relatively low production cost and ability to achieve a range of microstructures with different mechanical properties. In particular, there has been an increase in demand for thin-wall ductile iron castings to provide components with high strength-to-weight ratios. For example, the automotive and truck industries are utilizing thin-wall ductile iron as a way to reduce vehicle weight.

In production, however, the high cooling rate of thin-section ductile iron results in increased amounts of carbides with a corresponding loss in mechanical properties, specifically ductility and toughness. In addition, there is only limited data for "quality" thin-wall ductile irons that meet or exceed the ASTM specifications. These are two obstacles foundrymen must overcome to increase the functionality of thin-wall ductile iron castings.

This article examines the effects of melt chemistry and molten metal processing variables on the tensile and impact properties of thin-wall ductile iron castings (3-mm sections) and compares the properties of 3- and 12-mm sections within the same casting. It will evaluate and identify the most important variables affecting the mechanical properties and carbide formation in thin-wall ductile iron castings and to suggest the most economical ways of reducing or eliminating the formation of these carbides.

EXPERIMENT AND RESULTS

Melts of 100 kg were prepared in an alumina-lined induction furnace. Variables studied included composition, inoculation type and practice, base iron pre-conditioning, and pouring temperature. The composition of the 20 melts made during this investigation ranged from 3.7-3.95% carbon (C), 2.35-3.4% silicon (Si), 0.025-0.045% magnesium (Mg), 0.005-0.009% cerium (Ce) and [less than]0.1% manganese (Mn). The samples for mechanical property evaluation were obtained from 3-and 12-mm sections of step block castings poured in [CO.sub.2] bonded silica sand molds. Tensile test specimens (2.5-mm-thick flat and 6.35-mm round) were machined from 3- and 12-mm sections, respectively. From the 3-mm sections, subsize Charpy impact test specimens measuring 2.5 x l0 x 55-mm were prepared, while 10 x l0 x 55-mm standard Charpy test specimens were machined from the 12-mm sections.

Tensile Properties

Overall Effect of Section Size on Tensile Properties--Quality baseline diagrams are devised by plotting the minimum values of yield strength vs. elongation and/or the minimum values of tensile strength vs. elongation from the five grades of ASTM A536 specifications. Irons with mechanical properties that plot on or above the baseline are acceptable. The higher the mechanical properties plot above the baseline, the higher the quality. High quality iron is obtained when the graphite is fully spheroidal and when the matrix structure is more uniform in composition (highernodule count, reduced intercellular structure/carbides, etc.). Ductile irons with properties that plot below the quality baseline are considered unacceptable.

Evaluation of acceptability of the 3-and 12-mm-section ductile irons using the tensile strength-elongation quality baseline is shown in Fig. 1. The 3-mm sections show a significant increase in strength compared with 12-mm sections, even though a substantial difference in elongation is realized due to the section size effect (i.e. variations in nodularity, nodule count and matrix structure).

It is clear that while many 3-mm samples showed low elongation values (likely caused by a high pearlite content or the presence of carbides), others showed strengths well above those required in ASTM A536 grades. At low elongations, some of the thin-wall samples showed superior strengths that normally can be obtained in ductile iron only through heat treatment. At moderate-to-high elongations, the thin-wall samples were significantly stronger than those from identical irons of 12-mm section. In general, irons with properties at lower elongation reflect lower Si (higher pearlite)/presence of carbides/lower nodule count or nodularity, while irons at higher elongation represent higher Si (higherferrite)/improved nodularity and higher nodule count.

Effect of Chemical Composition on Tensile Properties--Both the yield and tensile strength of 3-mm sections decrease significantly with an increase in carbon equivalent (CE) as a result of increase in ferrite content. On the other hand, 12-mm sections show a slight increase in yield strength with an increase in CE, but the tensile strength does not show any effect. The increase in CE shows a small improvement in the elongation of 12-mm sections as a result of an increase in ferrite, but the 3-mm sections are not affected much by changes in CE. The elongation of 3-mm sections reduces with an increase in carbon content, but the 12-mm sections do not show any effect. The increase in carbon content results in a slight increase in the strength of 3-mm sections, whereas the strength of 12-mm sections does not significantly vary despite changes in the carbon content.

Si shows the most significant effect on the tensile and yield strength of 3-mm sections (Fig. 2). The decrease in strength at higher Si content results from an increase in ferrite content/nodule count. The 12-mm sections show a moderate increase in yield strength with an increase in Si content due to the solid-solution strengthening of the ferrite, whereas the tensile strength is not affected much by change in the Si level. A strong effect of Si in increasing the elongation of both sections is shown in Fig. 3.

An increase in Mg content is shown to decrease the elongation in both sections as a result of the carbide stabilizing effect of Mg in cast iron. On the other hand, the increase in Ce content shows a small increase in the elongation of both sections as a result of an increase in nodule count and thus ferrite content. Mg and Ce do not show any effect on the strength of both sections. Small changes in aluminum (Al), sulfur (S) and phosphorus (P) content have no effect on the elongation and strength of either section. However, a small variation in chromium (Cr) content shows a significant effect on the elongation but no effect on the strength of both sections.

Effect of Processing Variables on Tensile Properties--An increase in the pouring temperature shows some improvement in the elongation of 12-mm sections, but the elongation in 3-mm sections and strength in both sections does not show any effect.

Pre-conditioning of the base iron was done by the addition of 0.2% ferrosilicon (FeSi) (75%), 0.2% silicon carbide (SiC), or 0.1% FeSi + 0.1% SiC, immediately before tapping. Pre-conditioning increases the nucleation potential and, thus, minimizes the potential for primary carbide formation in the final iron while improving the elongation.

In stream inoculation experiments, each melt was used to test three kinds of inoculation treatments with 1% of 75% FeSi in the first tap, 1% of 75% FeSi with rare earths in the second tap and 1% of 75% FeSi with rare earths and bismuth (Bi) in the third tap. In the late-inoculation procedure, an additional 0.1% inoculant was placed on a filter at the base of the pouring cup. The late-inoculation was much more effective in increasing nodule count and reducing carbides and thereby improving elongation of both sections.

In the late-inoculation experiments, two different methods for the addition of the 75% FeSi inoculant were used. In one, 1% of the FeSi was added to the metal stream during transfer from the tundish to the pouring ladle. In the other, the same amount of inoculant was added but at four different stages--0.2% onto the metal stream during tapping from the furnace into the tundish ladle, 0.3% as a cover for the Mg treatment alloy in the tundish ladle, 0.5% stirred into the tundish ladle after Mg treatment and right before pouring the castings (no reladling was done to avoid loss of temperature) and finally the 0.1% late-inoculation in-spree addition. The addition of inoculants at different stages of molten metal processing resulted in the highest nodule count and, therefore, led to improved elongation.

In late-inoculation experiments, five types of inoculant were investigated in terms of their effectiveness in increasing the nodule count and chill reduction. The best results were obtained from 75% FeSi containing Bi and rare earths. There is an increase in elongation and decrease in both yield and tensile strength in the 3-mm section as a result of the increase in nodule count and ferrite content.

Pre-conditioning and late-inoculation with 75% FeSi containing Bi and rare earths significantly decrease tensile and yield strengths in 3-mm sections as a result of higher nodule count (with a corresponding higher ferrite content) whereas in the 12-mm section, the strength is not affected much.

Impact Properties

Overall Effect of Section Size on Impact Properties--The effect of section size on Charpy Vnotched impact energy at room temperature on both sections is shown in Fig. 4. For comparison purposes, the impact energies of 3-mm sections is shown four times higher to account for the total cross-sectional areas of the test specimens (i.e. 25 sq mm for 2.5 mm and 100 sq mm for 10-mm-thick Charpy bars). It appears that the smaller bars are less tough, but there may be an intrinsic size effect, making direct comparison invalid. It may, however, be noted that for identical conditions (i.e. two sections of the same casting), the 3-mm section will contain more pearlite than the 12-mm section, and so the 3-mm sections are expected to be less tough.

Effect of Chemical Composition on Impact Properties--The impact energies of 12-mm sections decrease with an increase in CE and C, whereas the 3-mm sections do not show much effect. On the other hand, Si shows a strong effect on the impact energies of 12-mm sections, while the 3-mm sections show a much smaller effect (Fig. 5). P as low as 0.01-0.016% shows a significant embrittling effect in the 12-mm sections. The impact properties in 3-mm sections also show a slight reduction with increase in P content. An increase in the Ce content produces an improvement in the impact properties of both sections. This might be a result of an increase in the nodule count with the addition of Ce, which promotes ferrite. The increase in Mg content results in a slight decrease in the impact values of 12-mm sections, but the 3-mm sections are not affected much by changes in Mg. The small variations in Mn and S content have no effect on the impact properties. On the other hand, the presence of certain residual elements such as Grand Al, even in small ranges, shows a significant effect on the impact properties of 12-mm sections but no effect in 3-mm sections.

Effect of Processing Variables on Impact Properties--Pre-conditioning of the base iron and late-inoculation shows a slight improvement in the impact values of 12-mm sections but does not show any effect in 3-mm sections. Inoculant type and method of addition of inoculant have a significant effect on the impact properties of 12-mm sections but no effect in 3-mm sections. The impact energies of both 3-and 12-mm sections are not affected by changes in pouring temperature.

Conclusions

In view of the extremely high nodule count in thin-wall ductile iron, unusual mechanical properties can be expected. Considering the quality index plots (Fig. 1), it is clear that while many samples showed low elongation values (likely caused by a high pearlite content or the presence of carbides), many others showed strengths well above those required in ASTM A536 grades. Indeed, at low elongations, some of the thin-wall samples showed superior strengths that normally can be obtained in ductile iron only through heat treatment. At moderate-to-high elongations, the thin-wall samples were significantly stronger than samples from identical irons of 12-mm section.

A direct comparison between Charpy impact values could not be made due to intrinsic effects associated with specimen size, but it is clear that toughness in the two section sizes was roughly equivalent when account was made for the total cross-sectional area. The main difference between the impact properties in the two section sizes lay in the relative insensitivity of the thin-section specimens to either melt chemistry or molten metal processing variables.

Of the elements contained in the iron, Si had the greatest effect on the tensile properties of the thin-wall sections. A 2.5-3.4% increase led to large decreases in tensile and yield strength, and a large increase in elongation as a result of higher ferrite content in the thin-wall sections. The same increase in Si content of the thin-wall sections had little effect on impact toughness. As expected, any processing variable that led to an increase in nodule count (with a corresponding increase in ferrite content) led to greater ductility, lower strength and improved toughness. Of the variables studied, the greatest effect was found to be from late inoculation, base iron pre-conditioning, and the use of an inoculant containing Bland rare earths.

This article was adapted from a paper (00089) presented at the 2000 AFS Casting Congress and is available from the AFS Library at 800/537-4237.