Austenite-Martensite volume change 1

[AndrewFinAustralia]

Hi All,

Former materials Engineer (Hons) now teaching at a high school - yes, we teach ferrous metallurgy to 17 year olds in NSW if they want to study engineering.

I was always taught that increasing the amount of carbon in martensite increased hardness through increased lattice distortion and volume increase.

I went looking for details of the austenite-martensite volume expansion for course notes - conflicting information that I could not make sense of without a trip to the local (60km away) university.

See attached images.

http://files.engineering.com/getfil...d-4bc1-a4b6-25f9c681c4bc&file=For_webpost.png

ASM handbooks (thanks google books) gives the volume increase ferrite-martensite is maximum at zero percent carbon (4.53%) as well as minimum at zero

Is anybody able to explain? - I'll back intuition and suggest increased % carbon = increased volume expansion due to deformation of lattice but am (now) completely confused on this point. Doesn't make sense to me as I know the formulae in the handbook pass strict peer review

 

[tbuelna]

I never went to college, but if I'd attended high school in NSW I could probably answer your question!

 

[racookpe1978]

What's the percentage increase in carbon in steel between the two crystal forms?

See, you're imagining the movement of the carbon into the crystal boundaries increases volume noticeably: But the amount of carbon in the steel is not increased, is it? So "Where was the carbon before the transition?" so that the transition requires a positive volumes change?

 

[AndrewFinAustralia]

Hi racookepe1978,

Thanks for the answer.

The austenite-martensite transformation is diffusionless. It goes from fcc austenite (atomic packing factor maximum - densely packed) to a looser-packed (more volume of interstices) martensite, at approx 250C

Carbon exists preferentially in the octahedral interstices in austenite. On transformation, this carbon distorts the lattice. More carbon = more distortion = harder (and I'd think, more volume increase, intuitively.)

I've got the two completely different relationships between prior %C and volume expansion from the authority literature that are poles apart. Just wondering if anyone could understand what I've missed. (which is probably me missing something ... obvious) I've spent too long on this already before I posted.

Cheers,

AF

 

[racookpe1978]

But that's what I wanted you to see: The crystal shape (of the iron!) changed, so the volume required by the total volume of iron increased slightly. The individual carbon now stuck in the crystal grains does not change the total steel volume very much -> More precisely, the 0.30 to 0.5 percent carbon moving around in the steel does not change the total volume much, compared to the much larger Fe->Fe change in crystal shape of every crystal around the carbon.

 

[AndrewFinAustralia]

Thanks again racookepe1978

Whilst more academic than practical, it is the cause of quench cracking.

The ASM handbooks give the following two contradictory relationships from the linked image above

(1) One is that the volume change is delta V/Vsub0 = 4.64-0.53% times the carbon content (meaning that the displacive transformation is greatest at zero carbon)
(2) The other is a linear relationship strain = 0.01 x Carbon content approx (meaning that at zero carbon, the martensite formed is zero) Assuming that steel is isotropic (not strictly true due to rolling texture,) that makes delta V/Vsub0 = (1+%C/100)^3

I'd instinctively choose the second based upon my understanding of theory, however, I also know and understand that heat treatment gives the first credence, to a point where it's referenced multiple times in ASM handbooks. BEng Hons (Materials) 1992, so theory was long ago. I'm at a loss on this one.

My question is: "can anybody see what I'm missing in the first explanation - why does the relative volume change to martensite decrease with increased carbon content in the austenite?"

I'd rather not guess or suppose for a set of state teaching notes.

Thanks again - I do understand where you're coming from, does the revised question help?

 

[CoryPad]

You need the equation before Vam, which is Vsa = -4.64 + 2.21 (2.21 % C). Read page 347 that you attached, especially the paragraph before Tempering. It shows higher volume change for higher carbon steel.

 

[AndrewFinAustralia]

THanks Cory.

I'm completely on the page and understand what you're saying. I looked at that to make sure I wasn't overthinking things.

When fully austenitised, the steel has zero stress - so the expansion when forming martensite is the key to quench cracking.

Just looking for a pointer to help me understand how less carbon in a steel equals higher volume expansion when it forms martensite. Seems counterintuitive and I may be thinking too deeply here.

One equation says a 0.5% carbon steel expands in volume on quench by 4.64-0.53*0.5 = 4.38%
The other expansion relationship (inferred) states that 0.5% carbon steel expands by (approx) 1.005^3-1 = 1.5% approx

Intuitively, more carbon should equal more expansion on quenching from austenite to martensite so I'm at a loss to understand the mechanism behind how the first equation should be valid. But it is widely reported and used in peer-reviewed literature, therefore has a sound theoretical basis.

I know a 0.8% carbon steel is WAY more susceptible than a 0.6% steel or a 0.12% carbon steel to quench cracking, (assume iced brine quench, I have done this successfully with 0.12%C steel.)

Massive difference in the two formulae.

Which one of the two above is more correct (or less incorrect) is my question - I haven't worked directly in heat treating, just annealing of 1000's of tonnes of steel sheet per day, so this wasn't something we needed to cover or understand at work.

 

[metengr]

AndrewFinAustralia;
Volume changes associated with phase transformation are predictable based on equations in ASM Handbook, Volume 4, Heat Treating. Linear changes are approximately 1/3rd of the volume change predicted by

Delta Volume of Austenite-Martensite = 4.64-0.53 × (%C)

Looking at atomic volume changes from ASM Hanbook, Volume 4A

Austenite 11.401 + 0.329 C(%) apparent atomic volume in Angstroms
Martensite 11.789 + 0.370 C(%)


Also, for higher carbon contents you have increased possibility of retained austenite that will affect volume change estimates and also possibly lead to cracking after quenching.

 

[metengr]

You may also find this paper interesting

http://www.researchgate.net/profile...bon_Steels/links/00b7d53b4e0e091cea000000.pdf

 

[AndrewFinAustralia]

THanks for that Meteng,

I'll look up the paper as well.

Is the rule of thumb still 0.7% C at which austenite is retained at a room temp quench?

AF

 

[metengr]

I believe the paper states a value of 0.6%C.

 

[Lyrl]

Higher carbon causes a more distorted lattice and higher stresses. It's been a long time since college, but my understanding is a more distorted bcc lattice is "closer" to the close packed fcc structure - I think of it as the stresses prevented the lattice from completing its movement to fcc - and it wouldn't surprise me if its volume change was less (compared to undistorted bcc).

Quench cracking isn't about volume change being different (compared to alloys that don't crack) but about the material being unable to handle the stress. Higher carbon martensite is more brittle than lower carbon martensite and cracks at lower stress levels.

Also, having worked in commercial heat treatment for a decade, it's very difficult to get retained austenite in a plain carbon steel (steel where a little manganese is the only alloying element). You might be able to see some with X-ray diffraction in the .6-.7% C range, but to see it visually I'd guess closer to 1.2% C. Then there are tool steels like H13 that have nominal .35% C and retain austenite heavily.

 

[AndrewFinAustralia]

THanks all

That paper that Metengr put a link to above is a free download. I've just read it and it completely changes my understanding of the martensitic transformation - this would be worthwhile reading for all involved in hardening. Sherby is a world expert in this field.

Brilliant summary - It clarified things perfectly and changed my understanding from the simplified explanations given at uni. (which work for basic understanding but didn't make sense when challenged)

Lyrl -> thanks also. I didn't mention the brittleness of martensite or the high carbon/low carbon martensite, I was aware of the two forms -> in light of the linked paper above, I now think of it differently. Thanks for your input here.