Dear Chuck,
Dear all,
here is a first humble attempt to find an approach on what we have discussed about.
I have tried again to find any hints in regard to explain the physical mechanisms of intermetallic phases or compounds(?) or both(?) and if their origin may be based on stoichiometrical or other reasons. And... thanks Prof. "GOTTSTEIN" (Physikalische Grundlagen der Materialkunde - a SPRINGER textbook) and his explanations of physical material sciences I have experienced a kind of slight "enlightenment". And I hope very much that I will find the right words once again for explaining what GOTTSTEIN has described in a very - diplomatically expressed - scientific view on the matter of subject, being the object of our interesting theoretical exchange of thoughts.
Considering what I mean to have understood by reading the "high purity" theoretical explanations, I would like to start very simple in the field of alloying metals. Most of metallic materials are forming systems being fully miscible in melting condition. But there are also systems being totally immiscible in solid and liquid state. Well known, and thus being used here, is the system "Iron-Lead" showing a miscibility gap between liquid and solid state and is finally immiscible in both conditions. And as I understood, the miscibility gap at all, is the main factor for all further complex explanations in regard to the forming of intermetallic phases in different metallic alloys. What can be stated is, that in the predominant number of alloy systems the alloying elements are completely miscible whereas while solidification occurs this ratio is varying to a lower level of miscibility. This is, in solid state of these alloys governs limited solubility.
What I could find out is that the topic we are talking about is - as far as I have once mentioned already - totally tricky. This is being induced through the immensely high number of existing intermetallic phases and compounds, which lies in the order of ~ 5000(!). When I have read that, I was sure that we - my dear appreciated colleagues - have selected one of the "easier" things in metal physics ;-) .
Well, you remember, what I have asked was about any coherence in comparing non stoichiometric structures like Carbides with other structures like Sigma (FeCr) and if these compounds being now non stoichiometric or - what is being suggested by the term "compound" - rather stoichiometric in sense of chemical aspects, which means that the rules of constant proportions and thus constant valences should be able to be determined. But, as we have discussed already Hume-Rothery has clarified that in case of many intermetallic phases the composition - expressed by the empirical formula (e.g. FeCr) - does not fully predict the valence of what should be expressed by this formula. This was the main reason of my question. Why, when we have an empirical formula - normally expressing stoichiometry as the driving force for forming the compound, why do we then not also have a constant valence of the compound?
GOTTSTEIN - and at the latest then I saw, I am not the only one getting grew hair when talking about Sigma et al - described the discrepancies in dealing with intermetallic phases are being based on the typically "non chemical character" of these structures although - and this is important - chemical empirical formulism is being used to express their composition. Therefore he proposed firstly to find a common and clear terminology having the benefit of separation what is meant by using the term "compound" or "phase". He explained that the reasons of the origin of intermetallic phases are numerous and thus the definition of why intermetallic phases are being formed is complicated, additionally, since different interacting factors may be the reason for the origin of intermetallic phases. The very main reason is - as in most sequences in material science too - the enthalpy, which I do not want to deal with here further on (I hope you may agree).
Crucial point is, to recognize that the term "intermetallic compound" is a bit "misleading". In German he used "irreführend". Subsequently I would like to state the recommendations for the separation of technical terms in regard to "Intermetallic Systems"
1. The term "intermetallic compound" should be used when the composition of the substance is s t r i n g e n t stoichiometrical based. I.e. MgZn2 or NaTl. In German we say "Strichphase" which might be translated approx. with "dash-phase". These phases have a stringent composition, often being accompanied from solidus-line-maxima and being represented as a dash within the phase-diagrams. Therefore, following from that, in most cases, this term is not recommended to be used in most of intermetallic systems (You see, this replies my question already partially).
2. The term "Intermetallic phase" is being understood as a "neutral" determination under special consideration of the rules of thermodynamics. What was impressing me - and at the latest here I understood the main difference - was that in case of "phases" also the rules of "phase-variations" are being allowed. This is, it explains why I have read so often different percentage compositions of Sigma (FeCr). It may vary, since it has an allowed width of phase-composition(!). In case of FeCr-Sigma (our famous weld-metal phase) the variation-width of Chromium content within the area of existence of the phase is between 43... 49% Cr and thus the variation of Iron content within the phase is between 57... 51%. Another example for varying phase-width is the so called "epsilon-AgCd3" which has a variation width of 70... 82% Cd. The latter may only be an example, since I do not guess that we may be confronted with this phase anytime in welding.
So far so good, but the different (most important) mechanisms of intermetallic-phase generation are so complex that it would certainly impossible to deal with them within this humble post. Nonetheless they can - in a very strong simplification - be stated as following. I have allowed myself to prepare some small Portable Document Files showing the most important phases and compounds in a general treatment. I hope very much that these sketches can clarify the sequences and mechanisms in an understandable way.
Since as mentioned above, there are a number of interacting and overlapping sequences for intermetallic-phases formation:
· Phases determined by Valence
These kinds of phases (also called Zintl-phases) being originated due to different locations of atoms within the groups of the periodic table of elements. This is, strong electropositive elements are forming phases with strong electronegative elements. See also attachment Zintl.pdf, showing the binary phase-diagram of Ca (electropositive) and Silicon (electronegative). As can be seen, these phases have stringent stoichiometric induced structures ("dash-phases").
· Phases with high package density (Laves-Phases)
High density packed phases are being counted to the "Laves-phase-group". Depending on which elements react both can occur, either phases with stringent stoichiometric ratios or phases with phase-width variations. Identical Valence-Electron-Concentrations (VEC) and comparable metal-radii lead to homogeneous solubility. Differing VEC and atom-radii lead to defined package-density orientated compounds, see also the attachment Laves.pdf. These kinds of intermetallic systems having most an optimal ratio between the system forming element-radii (A/B). In case of the system Mg/Cu this Mg/Cu radii-ratio is 1.228 and thus a system with a high package density (~72%) can be formed. By the way, also Sigma-phase in welding is a kind of high-density package phase forming the transition to the Hume-Rothery-phases and having a phase-width, and thus a concentration-variation. See also attachment Sigma.pdf.
· Hume-Rothery-Phases
Basing on extremely intricate coherences - I beg your understanding for not treating here further - in regard to the so called Valence-Electron-Concentration (VEC) these phases are being originated. The elements of which the phases are formed of are located under "A2" (green accentuated) and "B1" (violet accentuated) within the period table of elements, see also attachment Hume-Rothery.pdf.
To be honest, when having a look onto the Iron-Chromium phase-diagram, where can be seen that Sigma has a variation in composition, it should normally have been clear for me that Sigma may not have a stringent stoichiometrical induced structure and should count to transition phases to Hume-Rothery. But due to I have been confused about the different meanings of chemical compounds (stoichiometric) and phases (stoichiometric or non stoichiometric) and covalent phases and Valence-Electron-Concentration (VEC) I was - honestly - unable to see what Sigma really is inside.
Dear Sirs, this was a very humble first attempt of creating clearness in a field of structure-chemistry which is not so easy to see through for a simple non academic - but "proud to be a" - welder like me!
I must admit - I love physics, but structure chemistry won't be my favourite topic in the future ;-) .
I do really hope now that you supernal stainless-steel-experts may correct me if I have explained or understood something wrong. To be corrected by you would be a honour and certainly never a disgrace!
I remain with greatest respect, best regards and in hope to reading soon from you,
Stephan
P.S. It's ~ 1.30 a.m. Sunday morning now where I live... Now as I have written this above I will find a bit of "relaxation" to have a nice rest of the weekend ;-) .