Introduction.
The most characteristic structural component found in cast iron, is without any doubt the graphite phase. Its shape, distribution and dimensions contribute largely to the mechanical and physical properties that can be obtained with this unique material. No publication on cast iron is complete without mentioning the possible formation of carbides, which is generally regarded as an undesirable event.
On these pages, an overview will be given about the way present day's ideas about carbide formation have developed, how various names were given to carbide compounds and the unexpected variety of existing theories on carbide formation that still exists today. An alternative view on carbide formation has already been published on this website.
The first ideas.
At the start of the 19th century, it was already well-known
that in iron-carbon alloys, carbon could occur in quite different forms.
Carbon could be present in elementary form, as graphite, or in combined form, as ironcarbide.
The German metallurgist Karsten succeeded in separating this last component and to establish the chemical composition as an iron-carbon compound containing 6.7% carbon. It lasted however decades before it was generally agreed that its composition was indeed Fe3C(1).
Although the majority of analysis had been performed on high carbon steels, it appeared that the iron-carbide component found in white and mottled cast iron showed the same composition, indicating that its formation mechanism probably was the same.
Various mechanisms.
Various views, however, existed on this formation mechanism.
-At that time, it was generally assumed that iron-carbon alloys solidified as a homogeneous mass. Any separation of specific compounds, such as iron-carbide, therefore, had to take place after solidification, i.e. in the solid state (1,2).
The temperature region in which iron-carbide was supposed to form was estimated at 600-7000C (3,4).
-In 1885, Osmond and Werth published their "Cell-Theory", in which not only the existence of allotropic forms of iron was proposed (now known as austenite and ferrite), but in which also a new look at carbide formation was given.
Their research on high-carbon steels, showed that the matrix consisted of grains or cells of iron, encapsulated by a thin layer of iron carbide. During solidification, iron globules, or cells, are formed first and continue to grow. The remaining melt solidifies as ironcarbide. In this way, the carbide-phase actually glues or binds the previous formed cells together. This view makes it understandable why Osmond called the iron-carbide thus formed, "Ciment" (French for binder or cement). Figure 1. Three-D view of ironcarbide in hypereutectoid steel after dissolution of austenite.
Up till the present day, the name cementite actually reminds us of this mechanism as it was proposed in 1885.
Introduction of the Iron Carbon diagram.
In 1897, Roberts-Austen published his first preliminary Iron-carbon diagram, incorporating the new view that the behavior of alloys( either liquid or solid) could be compared with the behavior of all other mixtures.
This new view made it possible to explain structure formation in alloys in a completely different way. Jüptner (5) for instance deduced that when carbides are found in the structure, it merely means that these carbides must have existed in the liquid, before solidification started.
If on the other hand, graphite was formed, it meant that carbon in elementary form was dissolved in the liquid. This fundamental question of the form in which carbon is present in liquid iron would remain important in the decades to come.
In 1899, Roberts-Austen published his final iron-carbon diagram. In this diagram the hypothetical curve for cementite formation in cast iron was drawn as the dotted line x-y-z. Figure 2. First Iron Carbon diagram, published by Roberts-Austen in 1899.
Eutectic or not?
According to Osmond, the characteristic cementite structure in cast iron should be regarded as a eutectic of iron and ironcarbide. The carbide region forms one phase of the eutectic, the embedded small islands (which are transformed at a later stage into pearlite) form the other. Figure 3. First sketch of iron-ironcarbide eutectic structure as it appeared in literature(ref.6, 1900).
This new "eutectic" view of cementite was, however not immediately accepted.
The findings that lead to the construction of the first Iron-carbon diagram were critically reviewed by Heyn (6). He was of the opinion that the cementite (carbide) region is in fact composed of many small carbide-crystals, grown together in an irregular way and enclosing certain regions of the original liquid.
When solidification is completed, these regions transform into pearlite. Such a structure merely gives the impression of being a eutectic, but actually isn't as shown in figure 4. >
Figure 4. Alternative view of the cementite eutectic.
The famous Dutch scientist Bakhuis-Roozeboom (7) made a major contribution to the new Iron-Carbon diagram, by adding a theoretical line A-a, indicating an increasing carbon content of iron crystals (austenite) that separate at decreasing temperatures.
At the same time, however, he also doubted the actual existence of an ironcarbide eutectic and proposed a completely new and different mechanism for carbide formation.
In this view, the formation of ironcarbide is the result of a chemical reaction between austenite crystals and graphite. This transformation was supposed to take place in the temperature region between 1000 and 1100 C. Figure 5. The Iron Carbon diagram, adjusted by Bakhuis-Roozeboom in 1900. Notice line A-a and the temperature range between 1000 and 11000 C, where carbide formation was suppose to take place as the result of a chemical reaction between graphite and austenite (at that time called martensite!).
Among scientists this view would be the start of a struggle for the right theory in the years to come.
The proposed mechanism could, however, not explain what was actually found in practice. Slow cooling castings should show, according to this theory, an increase in carbide content, as the available time for this reaction was long. Thin castings should show only a little amount of carbides and a high graphite content, as the available reaction time is short. In both examples, the contrary, however, is the case.
Reactions, provoked by this controversial view, would lead to more discussions about carbide formation and new idea's were proposed (9), such as:
-A lack of nuclei for graphite formation leads to carbide formation,
-When graphite forms, the internal pressure will increase which automatically suppresses further graphite formation and increases the tendency for carbide formation.
The stable and metastable system, a convenient solution.
To overcome the difficulties in explaining the occurrence of so many possible phases, Le Chatelier (10) suggests that it might be better to use two separate diagrams to explaining the formation of all phases that can occur in iron carbon alloys.
The first to put that in practice was Heyn (11), who published his idea's mainly based on microscopic research, and presented his idea's in 1904 at the same conference where Bakhuis-Roozeboom withdrew his controversial theory(8).
Heyn emphasizes the fact that under-cooling, i.e. unstable equilibrium conditions, should be brought in to explain metallurgical processes. When this is applied to cast iron alloys, the solidification process can be explained as follows:
-Only in the case of hypereutectic compositions, enough nuclei (graphite particles) are available to force graphite to precipitate directly from the melt. -In all other cases, these nuclei are absent, causing under-cooling which forces the melt to solidify according to the metastable system.
The carbides, so formed however, are metastable, and can easily desintegrate into austenite and graphite.
Independently thereof, Charpy (12) had also proposed two separate iron carbon diagrams, divided into a stable and a metastable system. All diagrams were slightly modified and combined (13) by Heyn and are in principle still in use today as the Heyn-Charpy double diagram.
From now on, the existence of a metastable eutectic consisting of austenite and ironcarbide appears to be generally accepted. As this eutectic was still nameless,Wüst (14) proposes to call the cementite eutectic "Ledeburite", to honor the famous German scientist A.Ledebur.
End of Part I.
References:
(1) F.C.G.Müller,
Grundzüge einer Theorie des Stahls, Stahl und Eisen, Vol.8 (1888), pp.291-97.
(2) F.C.G.Müller,
Die kritischen Punkte der Eisenlegierungen nach den Untersuchungen Osmonds,
Stahl und Eisen, Vol.11(1891), pp.634-42.
(3) A.Ledebur,
Ueber die Benennung der verschiedenen Kohlenstoffformen in Eisen,
Stahl und Eisen, Vol.8 (1888), pp.742-47.
(4) A.Ledebur,
Neuere Untersuchungen uber den Kohlenstoffgehalt des Eisens,
Stahl und Eisen, Vol.11 (1891), pp.294-99.
(5) H,v.Jüptner,
Beitrage zur Losungstheorie von Eisen und Stahl,
Stahl und Eisen, Vol.18 (1898),pp506-510,552-557,616-620.
(6)
E.Heyn,
Die Theorie der Eisen-Kohlenstofflegirungen nach Osmond und Roberts-Austen**
**Proc. Inst.Mech.Eng." 1899, Febr. Fifth report to the Alloys research Committee, adapted by E.Heyn.
Stahl und Eisen, Vol.20 (1900), pp.625-36.
(7)
H.E.Bakhuis-Roozeboom,
Eisen und Stahl vom Standpunkte der Phasenlehre.
Zeitschrift für physikalische Chemie, Vol.34 (1900), pp.437-87.
(8)
H.E.Bakhuis-Roozeboom,
Uber die Anwendung der Phasenlehre auf die Gemische von Eisen und Kohlenstoff,
Zeitschrift für Elektrochemie, Vol.10 (1904),pp 489-91.
(9)
H.v.Jüptner,
Eisen und Stahl vom Standpunkte der Phasenlehre,
Stahl und Eisen, Vol.20(1900), pp.1205-12+1269-73.
(10)
H.v.Jüptner,
Eisen und Stahl vom Standpunkte der Phasenlehre,
Stahl und Eisen, Vol.21(1901), pp.795-801.
(11)
E.Heyn,
Labile und metastabile Gleichgewichte in Eisen-Kohlenstoff-Legierungen,
Zeitschrift für Elektrochemie, Vol.10 (1904), pp.491-503.
(12)
Anon.
Das gleichsgewichtsdiagramm der Eisenkohlenstoff-Legierungen.
In: Referate und kleiner Mitteilungen.
Stahl und Eisen, Vol.26 (1906), pp.426-27.
(13)
E.Heyn, O.Bauer,
Zur Metallographie des Roheisens,
Stahl und Eisen, Vol.27 (1907), pp.1565-71+1621-25.
(14)
F.Wüst,
Uber die Entwicklung des Zustandsdiagrammes der Eisen-Kohlenstofflegierungen,
Metallurgie, Vol.VI(1909), pp.512-30.
>
