See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/309453204
Kinetics of Nucleation and Growth of Graphite at Different Stages of Solidification for Spheroidal Graphite Iron
Article in International Journal of Metalcasting · October 2016
DOI: 10.1007/s40962-016-0094-7
CITATIONS
24
READS
1,062 6 authors, including:
G. Alonso IK4-AZTERLAN
36PUBLICATIONS 394CITATIONS SEE PROFILE
Pello Larranaga IK4-AZTERLAN
54PUBLICATIONS 688CITATIONS SEE PROFILE
Doru M. Stefanescu The Ohio State University 423PUBLICATIONS 7,052CITATIONS
SEE PROFILE
Amaia Natxiondo IK4-AZTERLAN
5PUBLICATIONS 40CITATIONS SEE PROFILE
KINETICS OF NUCLEATION AND GROWTH OF GRAPHITE AT DIFFERENT STAGES OF SOLIDIFICATION FOR SPHEROIDAL GRAPHITE IRON
G. Alonso , P. Larran˜aga, and E. De la Fuente
R&D of Metallurgical Processes, IK4-AZTERLAN, Durango, Bizkaia, Spain D. M. Stefanescu
University of Alabama and Ohio State University, Columbus, OH, USA A. Natxiondo
Veigalan Estudio 2010 S.L.U., Durango, Bizkaia, Spain R. Suarez
R&D of Metallurgical Processes, IK4-AZTERLAN, Durango, Bizkaia, Spain Veigalan Estudio 2010 S.L.U., Durango, Bizkaia, Spain
CopyrightÓ2016 American Foundry Society DOI 10.1007/s40962-016-0094-7
Abstract
The importance of the nucleation and growth phenomena that controls the solidification of castings on the mechanical properties and soundness of cast iron cannot be over- emphasized. The graphite nucleation mechanism is directly related to the carbon content of the iron and the inoculation treatment. To further understand these phenomena, inter- rupted solidification experiments were conducted on spheroidal graphite irons at three different levels of carbon equivalent (4.0, 4.2, 4.4), with and without the addition of a commercial inoculant. A detailed scanning electron microscopy investigation was carried out to analyze and
quantify the possible nucleation sites at different solid fractions, as well as the influence of the inoculant in their formation. Thermodynamic software was used to evaluate the probability of formation of the compounds. A detailed discussion on the differences in nucleation of graphite between the beginning and end of solidification is provided.
Keywords: spheroidal graphite iron, nucleation, solidification, inclusion, FEG-SEM, field emission gun-scanning electron microscopy
Introduction
Over the years, a multitude of theories have been developed attempting to explain the nucleation process of spheroidal graphite (SG) iron.1,2 Nevertheless, significant questions remain unanswered. Because of the highly non-equilibrium conditions, the wide temperature intervals over which the nuclei are formed, and the complex reactions involved, their chemical composition is highly variable. This makes it dif- ficult to obtain a clear and simple answer of the nucleation process. In general terms, it is possible to explain all of these mechanisms for graphite nucleation in the liquid according to two theories: homogeneous nucleation on carbon-rich cluster or undissolved graphite (a one-stage nucleation) and heterogeneous nucleation (one- or multi-stage nucleation). It
should be mentioned that while these first types of nuclei are not homogeneous nuclei in the classical sense of the term since they are postulated to preexist in the melt, they are of the same nature with the graphite phase growing on them.
These theories predict that residual graphite should be an ideal nucleant for the formation of graphite during solidi- fication, but also that some micro-inclusions formed in molten iron are also possible sites for the heterogeneous nucleation of graphite, depending on the purity of the base metal, holding times and temperature, as well as liquid treatment processes.
The inoculation process is a way of controlling micro- structures, providing a suitable phase for the graphite
nodule nucleation upon cooling. Commercial inoculants are based on ferrosilicon alloys containing small quantities of elements, such as Ca, Ba, Sr, Al, Zr and rare earths.
These elements must have a strong affinity toward oxygen and sulfur in the melt, producing complex oxides and sulfides that act as nucleation sites for graphite.
The heterogeneous nucleation theory developed in the last 30 years is focused on the non-metallic inclusions present in all commercial cast irons. These particles must satisfy some specific conditions to act as possible nucleation sites, includ- ing: good crystallographic compatibility, low lattice disregistry or mismatch (not more than 3 %), fine dispersion in the melt (1–3lm) and high stability at elevated temperatures.3,4 The Gas Bubble Theory
This hypothesis was first advanced by Nieuwland,5 and then further developed by Karsay.6It asserts that the tiny gas bubbles in the liquid metal are ideal sites for nuclei on which graphite nodules can grow. In this theory, the gra- phite grows radially from the outside into the bubble as is indicated in Figure1, where (A) is gas bubble, (B) is graphite, (C) is melt and (D) is austenite.
Karsay’s gas bubble theory is in principle based on the presence of carbon monoxide bubbles. However, in industrial ductile iron heats, strong deoxidizers are added to neutralize any oxygen in the form of dissolved O or as CO gas. Furthermore, it is highly unlikely that a complete graphite nodule will extend into the entire volume of a gas bubble, since this eventually would have to involve diffu- sion of C through the graphite shell.
The Graphite or Carbon-Rich Cluster Theory Boyles8presented this theory in 1947. According to him, graphite grows from small-size crystalline graphite already present in the melt. Previously, Eash9presented the idea of Si-rich regions around the dissolving graphite particles, when the melt was treated with Si-based inoculants, which could promote the precipitation of graphite. Later, Feest et al.10argued that this supposition is not correct, since the dissolution time of ferrosilicon in liquid iron is only a few seconds and graphite forms at the interface between dissol- ving particles and liquid iron. They modified the theory and proposed that these seed crystals will be preserved in the melt, provided that Ba or Sr is present in sufficient amounts to prevent the graphite from dissolving back into the melt.
Using X-ray and neutron wide-angle diffraction on molten Fe–C alloys, Steeb and Maier11 brought supporting evi- dence of short-range order regions in melts with higher than 3.5 % C. While the exact structure of these regions was not established, the authors argue that they are carbon clusters containing approximately 15 atoms. At carbon contents consistent with the short-range order, the melt
exhibited increased viscosity.12 These and other13,14 experiments indicate that eitherCnor (Fe3C)nclusters exist in dynamic equilibrium in molten Fe–C alloys. These carbon-rich clusters (or molecules) may serve as nuclei for graphite. Recently, some of the authors of this work15 demonstrated that, in low-carbon gray irons, graphite nucleates at the austenite–liquid interface without the presence of any foreign inclusions. This supports the nucleation on carbon-rich clusters theory.
The Salt-Like Carbide Theory
According to Lux,16 certain elements like Ca, Sr and Ba forms salt-like carbides in the liquid, which offers a close match of lattice parameters to graphite and reduces the interfacial energy between the nucleus and the substrate to allow for extensive graphite nucleation during solidification.
Figure 1. Karsay’s gas bubble theory.7
However, it is highly unlikely that the inoculating active elements (Ca, Sr or Ba) prefer to combine with C to form salt-like carbides instead of forming oxides or sulfides which are significantly more stable.
The Silicon Carbide Theory
Wang and Fredriksson17,18 observed that, following the dissolution of ferrosilicon in liquid cast iron, SiC crystals and graphite particles were formed in the melt close to the dissolving ferrosilicon particles. A theory was developed, and calculations were performed to explain the nucleation of graphite and the fading mechanism of these particles.
The theory was based on the assumption of the existence of local supersaturation of C and Si in the melt subsequent to the SiC dissolution, which provides the necessary driving force for homogeneous nucleation of graphite. The fading effect was explained by the homogenization of C and Si in the melt through convection and diffusion.
However, the critical role of elements such as Ca, Sr and Ba in the FeSi inoculant cannot be explained by this hypothesis.
The Sulfide/Oxide Theory
Several investigators19–23have suggested that the sulfides, oxides or nitrides, which are formed after the addition of the inoculant, can act as nucleation sites during the
solidification of graphite. Lalich and Hitchings24 and then Kusakawa25 confirmed this theory, by demonstrating the importance of non-metallic inclusions. They concluded that the majority of graphite nodules in SG iron are associated with non-metallic inclusions, mainly, MgCa sulfides.
Other scientists such as Warrick26suggested that nuclei for lamellar and spheroidal graphite are composed of complex oxides and sulfides. Also Skaland27reported that oxy-sul- fide particles will have at least one lattice spacing that could match the graphite lattice spacing and create the possibility of a favored substrate for graphite growth.
Igarashi et al.28 found MgO/MgS enveloped by (Mg,Si,Al)N, nucleating a nodule. Also, he observed CaS, MgO and Al2O3in the core, while (Al,Mg,Si)N in the shell surrounded by graphite nodule.
The Silicate Theory
Skaland et al.3showed that the majority of the inclusions in ductile cast iron are primary or secondary products of the Mg treatment and reported that the hexagonal nuclei are composed of double-layer compounds, namely a MgS and CaS core surrounded by MgO SiO2and 2 MgO SiO2with an epitaxial growth mechanism of graphite on the oxide.
Table 1. Chemical Composition (mass %) of Experimental Cast Irons
Quenching C Si CE Mn Mg S Ti P
I 3.78 1.93 4.38 0.22 0.038 0.009 0.030 0.016
II 3.56 2.03 4.20 0.19 0.039 0.011 0.021 0.017
III 3.30 2.00 3.93 0.19 0.041 0.010 0.020 0.018
Figure 2. Example of cooling curves obtained during the quenching experiments showing the time and tempera- ture of quenching (QI inoculated).
Figure 3. Example of the calculation of the amount of fraction solid through the grid method;fS=0.42.
Each of these hypotheses was supported by a variety of techniques. Jacobs et al.29 attempted to identify nodular graphite nuclei using transmission and scanning electron microscopy coupled with energy-dispersive X-ray analysis (EDX). Lalich and Hitchings24 used the electron micro- scope and the scanning electron microscope to investigate nodular iron graphite nuclei in metallurgical cross sections.
Igarashi and Nakae30recently described the heterogeneous nuclei of SG using a field emission-type Auger electron spectroscope and a transmission electron microscope.
The present investigation was designed to identify the nature of nucleation sites for nodular graphite by evaluat- ing the chemistry of inclusions found in the graphite spheroids and in the melt. Two types of particles were mainly considered, namely sulfides and titanium com- pounds. In most cases, these particles are accompanied by small nitrides and silicates. The nature of these inclusions was studied through a field emission gun-scanning electron microscopy (FEG-SEM) using different techniques, such as spectrums, mappings and line scan, and using different detectors which allows locating inclusions in the graphite under the polished surface of the metallographic sample.
The main objective of the present paper is to analyze a large number of inclusions that may act as possible
nucleation sites for graphite and to study their evolution (number and nature) as a function of the solid fraction. The investigation was carried out on inoculated and non- inoculated samples with the same chemical composition.
Experimental Strategy
An in-depth FEG-SEM study was carried out on eutectic Fe–C–Si alloys produced from melts of commercial com- position with different carbon equivalent. The thermo- dynamic software FactSage was used to evaluate the Figure 4. FEG-SEM images of the same graphite in heat
QIIIS1 (3.3 % C): (a) graphite without etching; (b) graphite with deep etching.
Figure 5. EDX spectrums of s1, s2 and s3 from Figure4 before and after MgS removal.
probability of formation of metallic and non-metallic inclusions.
Melting and Casting
Three heats were produced in a 100-kg medium-frequency induction furnace (250 Hz, 100 kw). The charge for each heat consisted of 32 kg of ductile iron returns and 18 kg of high-purity iron. Predetermined amounts of a commercial graphite (98.9 wt% C, 0.03 wt% S) and of FeSi alloy (74.6 wt% Si, 0.3 wt% Ca, 0.7 wt% Al) were also added to the metallic charges. After melting, the composition was checked and adjusted according to the required target.
After superheating to 1500°C, the iron was transferred into the pouring ladle for Mg treatment with 0.55 kg (1.1 wt%
of the batch weight) of a FeSiMg alloy (47.2 wt% Si, 6.02 wt% Mg, 1.15 wt% Ca, 0.24 wt% Al, 0.30 wt% Mn, and 0.88 wt% RE) by the sandwich method. The FeSiMg was positioned at the bottom of the ladle and then covered with steel scrap, before tapping the melt from the furnace.
From each melt, six standard thermal analysis cups were poured (three inoculated and three not inoculated) and the cooling curves were recorded. Inoculation was made
directly in the cups through the addition of 0.2 % of a commercial inoculant (62.6 % Si, 0.22 % Mg, 1.01 % Al, 1.79 % Ca, 5.96 % Mn, 0.13 % Ti, 6.77 % Zr, 0.65 % Ba and less than 0.07 % lanthanides).
Figure 6. FEG-SEM images of the same graphite of iron QIS1 (3.78 % C) under different detectors: (a) in-lens detector; (b) in-lens and AsB detector (sulfide and Ti compound inclusions are detected).
Figure 7. FEG-SEM observations of graphite with differ- ent inclusions in the core: (a) oxide; (b) Ti compounds and sulfides; (c) oxides, carbides, carbo-nitrides, sul- fides and nitrides.
The chemical analysis of all experimental irons is reported in Table 1. In addition to the elements listed in the table, the alloys contained 0.04 % Cr, 0.01 % Mo, 0.07 % Cu and less than 0.01/ % Al and 0.005 % Sn.
The solidification of the iron in the cup was interrupted by quenching in brine at increasing times, to obtain informa- tion on the microstructure and on the nucleation sites, at various stages during solidification (1: immediately after pouring, 2: after 10 s and 3: close to the end of solidifi- cation), as shown as an example in Figure2. After cooling to room temperature, the cups were sectioned and prepared for metallographic examination.
Characterization
After marking the metallographic samples without etching, they were examined in the FEG-SEM to look for possible nucleation sites. At each mark, a picture was taken and most of the graphite particles were analyzed with EDX.
Next, the samples were etched with Nital in order to evaluate the solid fraction.
The fraction of the area occupied by the solid after dif- ferent solidification times was measured through quantita- tive metallography techniques on color and Nital-etched samples, as described in detail by Alonso et al.31The liquid fraction includes cementite and ledeburite, while the solid fraction is formed of graphite, martensite, retained auste- nite and sometimes pearlite. An example of the method is provided in Figure3.
A grid was superimposed on the micrographs. The inter- section of the grid with the liquid fraction was marked by dots. The ratio between the number of dots and the total number of the intersections on the grid represents the area fraction of the liquid,fL. The solid fraction is then calcu- lated simply asfS=1-fL.
Extensive FEG-SEM examination of the nucleation of graphite was carried out on non-etched and deep-etched samples (Figure4). It is important to note that deep etch- ing, besides removing the matrix, also dissolves the sulfide inclusions in the graphite (Figures4b,5).
To identify possible nucleation sites, an Ultra PLUS Carl Zeiss SMT (0.8-nm resolution at 30 kV) in the STEM mode was used in combination with an X-Max 20 Oxford Instruments EDX detector with a resolution of 127 eV/
mm2. Three different detectors have been used for the generation of images: (a) an in-Lens detector (annular SE detector) for the surface structure, (b) an Everhart–Thorn- ley type detector (SE2) for topography and c) an angular- selective backscattered electron detector (AsB) for the compositional contrast, which allows detection of inclu- sions hidden inside the graphite (Figure6). This study is Figure 8. Different types of inclusions of RE. (a) oxides
in graphite; (b) sulfides in graphite; (c) oxides not associated graphite.
complemented with spectrums, mapping and line scans to analyze the main elements present in the inclusion and to estimate which are the possible compounds formed.
Experimental Results
Non-metallic inclusions of varying composition have been observed in the matrix and at the centers of graphite nodules of the different irons (Figure7). They include silicates, oxides and carbo-nitrides.
In most of the cases, these oxides are of Mg and rare earth (RE) (Figure8a). It is expected that the formation of stable rare earth oxides and sulfides (Figure8b) plays an important role in the heterogeneous nucleation of graphite in SG iron throughout the entire solidification range.32However, it is easier to find RE inclusions at the grain boundaries without any associated graphite (Figure8c) than in the graphite core.
Over 70 % of the inclusions which were assumed to play a role in graphite nucleation were identified as (Mg,Ca)S neighboring Ti compounds, which have been assumed
many times to be carbides due to their polygonal shape and because there is not any other element associated with the peak of Ti (e.g., Figure7b). This tendency was confirmed in non-inoculated and inoculated irons, inde- pendent of iron melt chemistry and of solid fraction.
Calcium, which is considered crucial for eutectic graphite nucleation,33 appears in a great number of inclusions forming sulfides, but its influence seems to be much lower than magnesium. It should be noted too, that Ba (a strong sulfide and oxide former) was never found in any nucleation site, maybe because of the low Ba content (0.65 % Ba) in the inoculant.
The chemical composition in different regions of the gra- phite was found through the use of EDX. X-ray composi- tion maps (Figure9) showed two different inclusions in the core of graphite: a complex sulfide (quasi-homogeneous distribution of S and Mg and to a lesser extent of Ca) and a nitride (similar distribution of N, Ti and Zr). The Zr source was the inoculant. Previous investigations have shown that Ti has a very high affinity to O and S, but in the samples analyzed in this work, it seems to be present much more as a nitrides or any other compound than a sulfide or an oxide.
Figure 9. X ray composition maps in a graphite of the QIII2I iron.
Normally, the Ti compound inclusions have a polygonal shape (square) and look whiter than sulfides, which exhibit more rounded shapes. According to Alonso,34 it is rea- sonable to assume that cube-shaped Ti compounds act as nuclei for lamellar graphite (Figure10a), but until now, there was little evidence of nucleation of SG by Ti com- pounds. Figure10b, c provides a clear example of a Ti compound nucleus in SG. In most cases, the sulfides appear to act as nucleation sites for the Ti compound and at the same time seem to nucleate the graphite. These inclusions (Ti compound?MgS) may be accompanied by other inclusions much smaller as silicates, oxides or nitrides, as illustrated in Figure7c.
The chemical composition along a line through the core of a graphite particle in Figure11 illustrates the presence of fairly well-defined zones within the inclusion. Analyzing the distribution curves, it is possible to determine the presence of a large magnesium sulfide as central inclusion
(similar distribution of Mg and S) accompanied by two Ti compounds on the corners.
A more detailed analysis using the graphics recorded for each element (Figure12) revealed the existence of small but significant peaks at the same positions, which allow to infer the presence of other complex oxides (Si–Al–Mg–O), simple nitrides (Ti, Al and Mg) and complex nitrides (Mg–
Al–Si–N) described previously by Solberg,35oxides (TiO, MgO) or carbo-nitrides Ti(CN).
Discussion
The nucleation of graphite in SG iron occurs hetero- geneously on a large variety of micro-inclusions in the melt. Minor elements, such as Ca, Al, Zr, Ti and Ce present in the inoculants, play an important role in the nucleation process.
Figure 10. TiC act as nucleus in different types of graphite. (a) Lamellar graphite.34 (b) Spheroidal graphite.
(c) detail of center in (b).
Figure 11. Change in chemical composition along a line through the core of a graphite from the QIIS1.
The thermodynamics software FactSage was used to assess the probability of formation of various compounds. Tita- nium nitrides and carbo-nitrides or Ti-Zr carbo-nitrides were not found in the database of the software. Selected free energies of formation of the significant compounds are presented in Table2.
According to Askeland36 and Francis,37 the efficiency of graphite nucleation is greater on oxides than on sulfides and nitrides. Thermodynamic data in Table2 seem to confirm this statement, particularly if complex oxides are included.
In this research, a total of 2456 graphite spheroids were investigated. About 38 % of the nodules showed any a nucleation site, even when the nucleus was under the surface of the metallographic field and thus not visible. Based on the frequency of elements and the thermodynamics free energy of formation, it appears that the majority of the nuclei are MgS which often are accompanied by TiC (Figure13). A large amount of Ti compounds were also identified as nucleation sites by themselves or also accompanied by MgO or any other oxide. As TiC has a relatively high energy of formation, the Ti compounds are probably more complex carbo-nitrides containing Zr, Al, Si and Mg.
Also, some complex oxides that mainly include Si, Al, Mg and sometimes other elements such as RE, Ti, Fe, Ca or Zr were identified.
Effect of the Inoculation Process
The micro-inclusions formed during inoculation are com- plex and of a rather intricate chemical nature. According to Skaland et al.,3after inoculation, hexagonal silicate phases appear at the surface of previous inclusions (from the Mg treatment). They will enhance the nucleating potency of these inclusions with respect to the graphite.
It appears that inoculation of ductile iron does not provide formation of new nuclei particles in the iron, but activates the inclusions already present in the melt. Table3 shows the average nodule count obtained in different areas from the central vertical cut38for each quenching experiment (X means ‘‘no available data’’).
No clear trend in the nodule count due to the addition of inoculant was detected. There are too many unknown and uncontrollable factors affecting inoculation, even more if quenching is performed. Determining their influence in the nucleation of graphite is not an easy task.
Figure 12. Concentration of different elements of a graphite from the QIIS1.
Some differences regarding the type of inclusions that may constitute the core of graphite were established (Fig- ure13). Independently of the inoculation, Ti compounds and sulfides seem to be the main compounds which appear in the nucleus of graphite. Sulfides are more common in the
case of not inoculated irons. Oxides as a group are important in both cases. Inclusions as nitrides (or carbo- nitrides) and silicates nucleate in lesser extent, being favored by the inoculation.
Effect of Solid Fraction
Correlation of the solid fraction with the percentage of nucleation sites detected at different solidification times is shown in Figure14. Samples quenched at the end of the solidification have not been considered because in all of them, the solid fraction is about 1. The correlation is poor when all data are considered and no relevant conclusions can be drawn.
The number of detectable nuclei is lower at higher solid fractions due to the size of graphite and the difficulty to find the nucleus with the EDX, and also because the number of nucleation sites increases significantly in the quenching performed at the early stages of solidification.
However, when only the Ti compounds are analyzed, this tendency is much clearer (Figure 15). The difference in the number of nucleation sites for different cooling rates is more apparent. The number of Ti compounds that act as nuclei increases with the cooling rate.
Conclusions
The goal of this paper was to identify the main inclusions acting as nuclei for spheroidal graphite and to study the Figure 13. Comparison of nucleation sites nature
according to the inoculation process.
Table 2. Free Energy of Formation of Possible Compounds in SG Nuclei
DG(J/mol) DG(J/mol) DG(J/mol)
Complex oxides Oxides Nitrides
2MgO2CaO14Al2O3 -2.62E?07 Al2O3 -1.21E?06 AlN -5.38E?05
5CaO4TiO2 -5.47E?06 Ti2O3 -1.12E?06 Ca3N2 -1.39E?05
3CaOAl2O33SiO2 -4.85E?06 Fe3O4 -6.49E?05 Mg3N2 -1.18E?05
2MgOCaO2SiO2 -3.98E?06 SiO2 -6.49E?05
2CaO FeO SiO2 -3.70E?06 MgO -5.85E?05
Double oxides CaO -4.78E?05
3CaO2SiO2 -2.96E?06 Sulfides Carbides
Al2O3SiO2 -1.86E?06 Fe9S10 -9.06E?05 TiC -1.63E?05
2MgOSiO2 -1.57E?06 Ti2S3 -6.38E?05 Al4C3 -1.23E?05
2FeOSiO2 -1.01E?06 ZrS3 -5.11E?05
5CaO4TiO2 -5.47E?06 CaS -4.25E?05 Carbo-nitrides
MgO2TiO2 -1.82E?06 MgS -2.94E?05 CaCN2 -1.39E?05
FeO2TiO2 -1.55E?06 FeS -1.07E?05
Al2O3SiO2 -1.86E?06 2TiO2ZrO2 -2.19E?06
influence of the inoculation process in the nucleation of graphite in ductile iron. A detailed analysis of the graphite through FEG-SEM was performed, with particular emphasis on the various inclusions which may serve as nuclei.
Theories inferring that graphite spheroids nucleate on non- metallic inclusions that contain an MgS core surrounded by an oxide shell, or with an outer shell of complex magnesium silicates, do not, in most cases, explain some of the findings in this work, as in many instances, the nucleus was made of two or three different compounds, and all of them were in contact with the graphite. MgS and Ti compounds (as a
carbide or nitride as their simplest form) seem to be the major sites for spheroidal graphite nucleation in this work.
A variety of different inclusions of complex composition (sulfides, oxides, nitrides and silicates) was found in the liquid.
The inoculation process did not seem to have a significant influence on the type of inclusions found in the core of the graphite spheroids.
The results of this research demonstrate once again the enormous complexity of the study of nucleation of Table 3. Average Nodule Count (Nodules/mm2) and fs (Solid Fraction) for the Different Quenching
Quenching Sample Non-inoculated Inoculated
Nod/mm2 fs Nod/mm2 fs
QI 1 444 0.44 354 0.35
2 426 0.57 414 0.41
3 172 0.97 276 1
QII 1 320 0.55 X X
2 471 0.70 486 0.65
3 219 1 215 0.99
QIII 1 X X 400 0.54
2 292 0.64 297 0.58
3 126 1 182 1
Figure 14. Effect of the solid fraction on the number of nucleation sites (inoculated samples).
spheroidal graphite. As such, these results must be con- sidered only the beginning of our effort in this direction.
Acknowledgments
The authors would like to acknowledge Diputacio´n Foral de Bizkaia for supporting this project.
REFERENCES
1. R.J. Warrick, AFS Cast Met. Res. J.2, 97–108 (1966) 2. H. Nakae, Y. Igarashi, Y. Ono, J. Jap. Foundry Eng.
Doc73, 111–117 (2001)
3. T. Skaland, T. Grong, Met. Trans.24A(10), 2321–2345 (1993)
4. M. Chisamera, I. Riposan, S. Stan, T. Skaland.
Proceedings of the 64th World Foundry Congress, Paris, France (Paper No. 62) (2000)
5. P.E.A. Van Nieuwland, Metalen, Jaargang VIII, nr.18, 333–335 (2004)
6. S.I. Karsay, Uitgave QIT, 1976 and 1980
7. S.I. Karsay,Ductile Iron I: Production, Quebec Iron
&Titanium Corp. (1976)
8. A. Boyles, Ohio: Metals Park (1947) 9. J.T. Eash, AFS Trans.49, 887–906 (1941)
10. G.A. Feest, G. McHugh, D.O. Morton, I.S. Welch, I.A.
Cook,Proceedings of Solidification Technology in the Foundry and Casthouse(1983)
11. S. Steeb, U. Maier, inGeorgi Publishing, ed. by B.
Lux, I. Minkoff, F. Mollard (St. Saphorin, Switzerland, 1974), pp. 1–11
12. W. Krieger, H. Trenkler, Arch. Eisenhuttenwesen42, 175 (1971)
13. A.A. Vertman, A.M. Samarin, Dokl. Akad. Nauk SSSR134, 629 (1960)
14. A.M. Samarin, V.A. Izmailov, Soviet Phys. Dokladi 14, 392 (1969)
15. D.M. Stefanescu, G. Alonso, P. Larran˜aga, R. Suarez, Acta mater.103, 103–114 (2016)
16. B. Lux, Modern Cast.45, 222–232 (1964) 17. C.H. Wang, H. Fredriksson, 48th International
Foundry Congress, Varna, Bulgaria (1981) 18. H. Fredriksson, Mat. Sci. Eng.65, 137–144 (1984) 19. M.A. Gadd, G.H.J. Bennett, 3rd International
Symposium on the Physical Metallurgy of cast Iron, Stockholm (1984)
20. M.H. Jacobs, T.J. Law, D.A. Melford, M.J. Stowell, Met. Tech.1, 490–500 (1974)
Figure 15. Correlation between solid fraction and the percentage of Ti compounds acting as nucleus (samples without inoculation).
21. K.M. Muzumdar, J.F. Wallace, AFS Trans.81, 412–423 (1973)
22. R.L. Naro, J.F. Wallace, AFS Trans.78, 229–238 (1970)
23. G.X. Sun, C.R. Loper, AFS Trans.91, 639–646 (1983) 24. M.J. Lalich, J.R. Hitchings, AFS Trans.84, 653–664
(1976)
25. T. Kusakawa, S. Okimoto, K. Kobayashi, K. Ide, H.
Okita,The Casting Research Laboratory(Waseda University, Tokyo, 1988)
26. R.J. Warrick, AFS Trans. 74, 722–733 (1966) 27. T. Skaland, AFS Trans.105, 77–88 (2001)
28. Y. Igarashi, S. Okada, Int. J. Cast. Met. Res.11, 83–88 (1998)
29. M.H. Jacobs, T.J. Law, D.A. Melford, M.J. Stowell, Met. Tech.3, 98–108 (1976)
30. Y. Igarashi, H. Nakae, J. Jap, Foundry Eng. Soc.74, 30–35 (2002)
31. G. Alonso, D.M. Stefanescu, P. Larran˜aga, J. Sertucha, R. Suarez, AFS Trans.120, 329–335 (2012)
32. L.J. Kozlov, A.P. Vorobyev, Cast Met.4(1), 7–10 (1991)
33. P.J. Bilek, J.M. Dong, T.K. McCluhan, AFS Trans.80, 183–188 (1972)
34. G. Alonso, D.M. Stefanescu, P. Larran˜aga, E. De la Fuente, R. Suarez,AFS Proceedings, Paper 16-020 (2016)
35. J.K. Soldberg, M.I. Onsoien, Mater. Sci. Technol. 17, 1238–1252 (2001)
36. D.R. Askeland, P.K. Trojan, R. Flinn, AFS Trans.80, 349–358 (1972)
37. B. Francis, Metall.l Trans. A, 10A, 1979, 21-31 38. A. Natxiondo, R. Sua´rez, J. Sertucha, P. Larran˜aga,
Metals5, 239–255 (2015)