Hi Juan, I recommend you take a look in the book, Recrystallization and annealing phenomena, by F. J. Humpreys, this is an excellent reference for microstructural phenomena.

Mr.Juan,

''Under high temperature deformation with high fluid activity'' which process are we talking about here? Probably, there could be some idea behind it....

best

RS

Hi everyone and thank you for those replies!

I'm thinking about geomaterials which are at the best similar to impure ceramics. The problem is that there's no direct correlation between the "well known" behaviour described in Metals and so and between those natural mixtures. High T means above 0.7Tm with fluid films at the grain boundaries, deforming at very slow strain rate (10E-13 s-1). In those cases we could see evidences of very high grain boundary mobility (no pinning effect by secondary phases) and the resulting microstructure is one with extreme grain growth. Of course when deformation stop Temperature could be rather high and some kind of annealing could be active. The resulting polycrystal is frequently very similar to one obtained by static recrystallization. So, under those conditions should we look for some specific feature to distinguish between the static and dynamic origin?

I guess texture could be an important source of information, but also grain boundary orientarions, and misorientation relationships across the aggregate...

Between static and dynamic RX microstructure (assuming single-phase aggregate), I expect some difference on grain size distribution (normal and log-normal distributions) and on LPO and texture strength (assuming deformation by dislocation creep, weak vs high). For example, a highly recrystallised microstructure (straight boundaries, low energy dihedral angles, etc), following a normal distribution but with a strong LPO (or CPO) is the expected case for a strong annealed rock previously deformed.

During annealing of a deformed rock , grain boundary migration is a transient process between GBM driven by (dominant-) strain stored energy (at initial stages) to GBM driven by boundary energy (at mature stages). This is the expected for a general case, but there are several possibilities attending the strain story of the rock, microstructures, number of phase, diffusion, etc. Then, the static RX of a deformed rock is not going to follow “normal growth” but probably will follow a transient stage between “dynamic-to-normal” growth.

Abnormal grain growth happens when there are some heterogeneous distribution of driven energies. It's possible to be observed at low and high temperature RX. At low T, only a small fraction of grains shows abnormal grain growth and large difference on grain size between populations are observed. As grain boundary mobility is exponentially temperature dependent, at high temperature it's more suitable that a larger population of grains show abnormal, and differences between large and small grains are expected to be minimized. But no trivial if only based on microstructure …

Regarding abnormal grain growth, many claim that anything natural is also normal. A rule of the thumb says that the symmetry of the effects reproduces the symmetry of the causes, so preferred orientation of the grains (dimensional, but also lattice - LPO) indicates growth/recrystallization in a stress/strain field. Under static recrystallization lattice and boundary energy are the dominant driving forces, producing "normal" aspect ratios and low-energy boundaries. Interesting and relevant diagram presented by Volker Klemm, though it's mixing an intensive parameter like temperature with an additive one like the amount of deformation, and an instantaneous property as grain size. That's why secondary recristallization plots at high temperatures, high amount of deformation and allegedly produces huge grain sizes, which is only a crude approximation. As regards the question, the answer lies, I think, in preferred orientation of the grains.

Thanks for reply guys!

I agree with Albert, this is what we could expect, but particularly at high T that is not always the case, It is Ok when dealing with synthetic 100% pure aggregates, but it is kind of weird thing to find the normal distribution in e.g. quartzites. Even the term log-normal should be used with care because of the sometimes loose use of statistic proofs in the literature. Besides, the assumption of "pure" dislocation creep as the mechanism that generate LPO at High T is questionable. The point is to what extent could we define a RX texture as a diagnostic feature ...

Abnormal growth or secondary recrystallization happens after metadynamic or dynamic recrystallization are finished and temperature is not lowered. Even at high temperatures you should distinguish between both from grain sizes in your microstructure. If some grains are much larger than others and are equiaxial you can be pretty much sure the occurred due to secondary recrystallization.

Hi David,

Yes you are right, so if grains are non equiaxial and all grains are really big one, you would say primary recrystallization is involved?

thanks for your reply!

Juan