- Joined
- Sep 9, 2003
- Messages
- 2,361
An excellent question since it get right tot hear tof the matter. Let's say we have a piece of 1084 steel that is fully lammelar annealed and is comprised entirely grains of pearlite...
In this image you will see lines or ridges that look like finger prints. those lines and ridges are alternating iron bands of iron and iron carbide that have separated out to form pearlite. On heating, once you pass the lower critical temperature the carbon in the pearlite will start to move and go into solution in the iron portion of the pearlite...
Here you can see the dark pearlite in the lower left and the remnants of it above that is partially in solution. This image is from a sample that I took just to non-magnetic where the transformation was incomplete in order to get a picture of the process. Notice how the little bands have turned to little strings of beads as they dissolve and the large field of lighter colored background that is a new structure forming because of it.
This process will begin at singular points of high energy, just like ice will form first around the edges of a pond or around a stick breaking the surface (or as Dr. Verhoeven would say, like bubbles in a soft drink coming from the same surface defect inside the glass). Thus the corners of the grain boundaries make a great starting point, as well as undissolved carbides and other bits. It is at these choice points that fresh baby austenite grains will begin to form.
Once formed the steel will be of a duplex structure, partially new austenite and partially old pearlite, but with more heat the process will continue and the austenite will replace the pearlite as it grows. The new austenite grains will grow until they meet each other and use up all the pearlite. Imagine a circle that only has three little grains growing until they meet each other, the final grains will be much larger due to the amount of space each will have to themselves. Now if you increase the number of baby grains to 10 when all the pearlite is gone those ten grains will have to be much smaller to occupy the same area- this is how grain refinement works!
However if you keep heating until every little carbide is dissolved the grain boundaries of the new austenite will become unstable and start to collapse and two grains will become one larger grain this is grain growth! But if you stop short of this and then cool those little austenite grains yet another process will begin once again at those points of high energy when the carbon comes back out of solution to form all new pearlite grains.
So, on any heating and cooling cycle no less than three separate phases with their grains, are involved- the initial structure, the new austenite grains, and the final crystalline version of the phase obtained on cooling. When you consider that just heating and cooling steel in any fashion results in two new sets of crystals every time, you can see how simply controlling the rates of heating that cooling can give you a whole lot of power in determining the internal makeup of the steel.
The baby grains forming in those points of high energy is called nucleation, and it is the rate of nucleation that determines the amount of grain refinement. Martensite has higher strain energy thus it can drop the grain size in one heat what may take two with pearlite. Heating and cooling rates can also effect how things come in and out of solution. Serious carbides like those made with vanadium or chromium will act as sticks in the pond and greatly increase the rates of nucleation.
Hope this helps.
.
