Dendrite vs Grain

[MagBen]

Is dendrite bigger than grain, or grain bigger than dendrite, or they are the same?
I have a binary alloy. It is easy to see the dendrite structure, but I never see grain structure no matter what etching method is used.

 

A dendrite is actually a single grain. It is the tree and branch structure resulting from solidification kinetics for a particular alloy system at a particular cooling rate. Change the supercooling of the melt and the solidification structure may change to something like columnar. Such structures are subsequently destroyed by forging, rolling, recrystallization, heat treatment, etc.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[EdStainless]

If you still see the dendrite structure then that is from the solidification and the alloy has not been annealed and it is still highly segregated. From the material in the dendrites to the material between the arms you will have full solidification range of compositions.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[MagBen]

@ironic metallurgist: making sense but not totally agree. Please take a look at a typical picture attached (sorry cannot tell the alloy name), I found it hard to say a dendrite is a grain. What is for sure (confirmed by EDS) is that the dendrite (white) is one phase, the grey is the other phase.

 

[MagBen]

having issue with uploading image

 

[MagBen]

here is another sample from an atomized particle

 

[EdStainless]

OK, time to look at the phase diagram.
Are there compounds that you expect to form? Or is this just a wide freezing range and resulting segregation?
Can you calculate the solidification rate from the dendrite arm spacing in the powder?

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[MagBen]

Ed, good questions!
yes I expect two phases/two compounds, which are supposed be two LINE intermetallic compounds, so there is no micro-segregation. It is either this phase or the other, nothing in between.
and yes, the composition has a wide liquidus-solidus range. The solidify rate was calculated to be 10^4 - 10^5 C/s for powder (second attachment), 10-100C/s for metal mold casting (the first attachment).

 

[EdStainless]

Well then is the phase balance what you expected?
The ingot looks like a well developed partitioning, the powder may not be as well developed.
Are you sure that there is no boundary zone around the second phase?

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[MagBen]

yes for the cast ingot there is no boundary zone around. this is highly pure binary, impurity level is about 200ppm. I am guessing the grain boundary maybe super light if there exists a grain boundary. however, for the powder (more dirt), i did see a black boundary that separated the two phases (brown and white). That is why i had the original question: is a dendrite a grain? my thought is grain >= dendrite. i.e. one grain can include one or couple of dendrites, while one dendrite cannot include multiple grains.

 

Not dendritic in the first picture, but in classical solidification the appearance varies widely depending on the sectioning plane.
Both pictures show two clearly defined phases.
As EdS asked, what is the alloy system and what does the phase diagram predict? Keep in mind the phase diagram describes equilibrium states, not often observed in real world alloys.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[MagBen]

@IM, i attached another cast image with dendrite, a slightly different composition though

 

[MagBen]

This is a relatively newly developed phase diagram. I doubt it is completed, especially I questioned the line compound may actually have a composition range. My calculation and analysis are based on the phase diagram.

 

[EdStainless]

It is not uncommon to have an as solidified structure that is dendritic with NO grain boundaries.
After a thermal treatment this will change. And hence why the slower cooled material is more developed.
Perhaps heat treating some powder would allow the smaller island to dissolve and merge with larger ones and 'fix' some of the chemistry variations.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[MagBen]

Ed, see attached again in which the powder was hot pressed at a temperature lower than solidus point. The dendrites were isolated and spacing increased from 2 microns to 10 microns! The previous particle boundaries are visible, but still no grain boundaries? I believe the particle boundaries were due to surface contamination.
so gain back to original question: where is the grain, where is the grain boundary? Or, it never existed?!

 

[EdStainless]

So not vac melt and vac atomized I gather.
In this alloy the only grain boundaries are between the phases.
Look at a duplex SS, there are two phases with (almost) no grain boundaries within each phase, just between them.
The grains are regions of identical composition but different crystallographic orientation.
I could see if this compact was forged and then reheat treated you may form some grains within each single phase region.
What structure are you trying to get?

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed