10/31/2016 Intro to MSE deformation mechanisms
Oct 15, 2023
Callister Intro to MSE, materials science, slip in single crystals, resolved shear stress, twinning vs dislocation motion, vector cross products, deformation mechanism maps, ductility vs strength, work hardening, recovery, recrystallization, grain growth, ostwald ripening
Okay, welcome to class, happy Halloween.
Okay, to start with class today, let's start with this brief video, because it's pretty rad.
Let me grab them, make sure I turn it on.
Okay, you'll have to listen for this, that's the key part of this video, is what you can hear.
In this demonstration, we're going to illustrate deformation twinning using these two bars of Crescent Cast Tin.
So tin has a tetragonal crystal structure, which is a much lower symmetry than raised metals,
which makes dislocation motion more difficult, and therefore favours deformation twinning.
So I'm going to take one of these bars, and I'm going to heat it up using Crescent Boiled Water.
So I'm going to leave that there for just a little while, and in the meantime,
I'm going to take this bar that's just at room temperature, and I'm going to bend it.
And I'm going to do that quite close to the microphone, and you should be able to hear why in a second.
So now I'm going to take the one that's been heating up, and do exactly the same thing to it.
Okay, so same material, only thing that's different is two different temperatures.
One of them makes a really just awful noise as you bend it, and the other one's silent.
That just makes me crazy.
Okay, so we're going to talk about what's the difference there.
Why does one make that awful creaky noise and one does not, right?
So to pick up where we left off last time, so we'll get to that in about five minutes,
but last time I said we'd pick up with calculating the number of slip systems in an FCC crystal.
So I'm going to give you a—let's see how much time it's set for.
We'll give you actually like a few minutes here, but I'd like you to calculate for an FCC crystal
how many slip systems does it have.
That means I want you to find out how many planes there are that are highest density,
and then how many directions there are that are highest density directions.
Now if there's a direction going forwards and going backwards, that's still just one slip system, right?
We don't count that twice. That's just one slip system.
Okay, so with a partner, with whoever you want, calculate whether it's A, B, C, or D here.
Okay, I'm going to go ahead and start the timer.
So 60 seconds, get your answers in.
Okay, 25 seconds to get this in.
If you're still struggling, I think one way to think of this is to picture where these three sides of the triangle are.
We know those are slip systems. Where would those be at on the cube?
And then add them up.
Okay, let's hope you have your answers in, A, B, C, or D.
People think—okay, overwhelming majority gets that it's B.
So for those that didn't quite understand why it was B, how do we understand it?
Somebody want to explain it to us that feels confident in their answer?
Anybody? What do you think?
Visualizing it is the key here. Let's try and draw it, right?
So let's draw our cube, right?
So a cube, just like a dice, has six sides.
And these things, these edges of the triangle, those cut across the faces of the cube, right?
So we could say that there are, like, on this bottom side, there's those two.
On the top side, there's those two.
And there's six sides, right? So six times two gives us our 12, right?
So that's where the 12 comes from here.
Yeah, Matt?
We'll do that in just a moment.
That one's harder to visualize.
So maybe we'll do it for a homework question next week or something.
I don't think I can get through it and still get to the material I want to cover for today.
Nobody stone Matt, please, for doing that.
Okay, so let's talk about the learning objectives for today.
We're going to talk about slip and single crystals.
We'll actually calculate it. That's where we left off last time.
We'll describe how you can deform via both dislocation motion and twinning.
And that goes back to the video we started class with.
We'll describe what is Hall-Petch.
It's one of these, like, early fundamental material science equations.
And how we use it to calculate yield strength as a function of grain size.
We'll talk about ductility and hardness.
We'll actually calculate it.
We'll introduce these important concepts of annealing, recovery, recrystallization, and grain growth.
We'll talk about deformation in ceramics or the lack of deformation in ceramics.
But we will talk about how glasses, elastomers, and metals deform.
If we have time, we'll get to viscosity.
And we'll even tie it into molecular weight and crystallinity of polymers.
So we left off last time.
We said that if you have a single crystal.
Imagine this big cylinder is a single crystal, right?
It's a giant, like, crystal of silicon.
They actually make them huge, right?
They're 12 inches.
They even make them 15 inches wide now and, like, 12 feet tall.
All perfectly oriented in one direction all the way through, right?
So if you take that and you pull on it.
What I've said in class is that you're only going to get motion along your slip planes, right?
So on your slip planes and in the slip direction.
So it has to happen on your slip system.
But what you don't know is whether or not the way that you grew your crystal with its orientation.
If it's going to, you know, cooperate with you trying to deform it.
So, for example, in this picture, if the slip direction has to happen along this red direction here.
But you're pulling it not, you know, parallel to that red direction.
How do you get it to slip?
Well, we said, all right, you have to take the force that you're applying to it, which is going to turn into a stress.
And you have to break that stress up into components, right?
And so there will be a resolved shear stress basically along this direction.
So Matt at the end of class last time said, well, does that mean that when you pull it, you're going to get a bunch of series of small slips?
That's exactly what you're going to get, right?
So this is an actual sample of a single crystal that they pulled.
And you can see that the slip planes would be sort of like this here.
And the direction was that way, right?
So as they pulled it, it did slip a little bit like that.
So the overall deformation is a vertical one, right, in this one relative to this one.
But it doesn't achieve it by stretching the atoms out.
They actually have to slide past each other along whatever its slip system was, right?
So if you grew this crystal in a different way and the planes were oriented differently, then your little keats here would be slightly different spacing, right?
But the principle still stands that you have to have slip along there unless you have twinning.
We'll get to twinning in a second.
So how do you calculate it?
All right.
Our resolved shear stress, that's tau r, is going to be equal to the stress that we apply, sigma, multiplied by the cosine of phi.
And phi is the angle between the normal of the slip plane and the loading direction.
And then cosine of lambda.
And lambda is going to be the difference between the loading direction and the slip direction, right?
So we need to calculate those two angles so that we can take the cosine of them to figure out this equation, right?
Therefore, since the critical resolved shear stress is a material's property, it's going to be a material's property, right?
This one's known.
The tau CRS, that's the critical resolved shear stress.
We would like to know as a function of whatever this tau CRSS is, for a given angle, what stress will cause this thing to start slipping, right?
So when they started pulling on this thing here, probably nothing happened when they were pulling on this one until they hit a certain stress,
the yield stress when this thing did start to slowly start sliding past each other.
So how do we calculate how big that is?
We use this equation right here, right?
Now, one last catch here is that these angles can't just be arbitrary.
They can't be anything.
In fact, we know that phi and lambda have to add up to be at least 90 degrees.
Why?
Why does phi and lambda have to add up to be at least 90 degrees?
This is oftentimes hard for students to grasp.
So let's imagine in this plane down here, this is the slip plane, right?
So it's maybe sitting on this table.
So the table is where the slip plane is going to happen, right?
And then the slip direction might be wherever it's pointing, right?
So right now, the slip direction would be towards Ryan, but it's along this plane, right?
Fair enough?
Now, the normal to the slip plane is this thing.
It's pointing straight up from it, right?
It has to be the normal, right?
So going back to our picture, we needed this blue line.
That's the slip plane normal, right?
That's this one.
And then the red line, that's where this thing's pointing.
So these we know obviously have to be 90 degrees, right?
To each other.
So the question is why does, when you apply a force, why do these have to be at least 90 degrees?
Well, because at best, if now we introduce another vector here,
now this is the way we're pulling it.
Let's say that I'm pulling it in plane of these two things, right?
Then it still has to add up to be at least 90 degrees, right?
45 from, hang on to this.
Checking on.
45 here maybe, and 45 here.
So at least it has to be 90 degrees, but it can be more.
Because think, this could be like that, but I could be pulling on it like that.
And so then you could have greater than 90 degrees when you were to add those two angles up, right?
Does that make sense?
Are there any questions on that?
Okay, that's hard to grasp sometimes.
So because of that, they have to be at least 90 degrees.
At what point will they be the largest?
So we can plot it, right?
And it turns out if you plot right here.
So we're plotting in 3D the cosine of phi times the cosine of lambda in degrees.
We're going to plot it from 45 to 90.
And we find that the largest value there corresponds when both of them are equal to 45 degrees.
Okay?
Which means this is our slip system, right?
If you want slip to happen, the best case scenario would be to have this lined up in it.
If it's all coplanar, then you'll have the highest likelihood of slip.
This will be less likely because your resolved force on it is going to be a little bit less.
That's all that means.
Okay, so what we really need to figure out then is how do you calculate these angles?
That's what it really comes down to, right?
Because ultimately I'm going to tell you that it's loaded along some direction.
The slip system is along these planes.
How are you actually going to turn that into angles, right?
So we know how to calculate these things.
And we do so with something that you did back in geometry or trig.
At some point in your math career you would have seen this.
That we can figure out the angle between two vectors by using the dot product, right?
So the dot product of two vectors, vector A and vector B,
is going to equal the length of vector A times the length of vector B times the cosine of the angle between them, right?
So if these are the two angles that we're trying to figure out on a question,
and we need to, or sorry, those are the two vectors, so we need to figure out that angle,
we can just apply the dot product rule, okay?
Now, how do we figure out the direction normal to a plane?
Let's say that the slip system is the 1, 1, 1 plane,
and we said it's along the 1, 1, 0 family of directions.
How do we figure out the actual planes associated with that?
First we have to figure out what direction is normal to this plane, right?
So normal to that plane, that's the perpendicular sign,
it's actually this direction, 1, 1, 1, in the cubic system.
That makes our life really easy.
It means if this plane of this table is the 1, 1, 0 plane,
then the normal is just the 1, 1, 0 direction.
That only works in the cubic system, right?
I won't give you anything harder than the cubic system in this class.
In your other courses you may have to do that.
Hexagonal or tetragonal, it doesn't work like that.
But in the cubic system, in this class, it makes it easy.
If I give you the slip plane, what's the normal to it?
Same thing as a plane, but now instead of writing it as a plane, you write it as a direction, okay?
Any questions on that?
So let me ask this.
If it wasn't cubic, if it was tetragonal or some other system,
how would you figure out the normal, right?
How might you figure out the normal to a plane?
First of all, do we need to convince ourselves that it's actually normal?
Imagine you've got your cubic crystal, right?
So we said that the 1, 1, 1 plane is this one.
So is the 1, 1, 1 direction normal to that?
Sure, because 1, 1, 1 direction, that means we go 1 this way,
we go 1 that way, we go 1 that way.
So our overall direction goes from this corner up to that one.
So sure enough, that thing will be normal to that plane, right?
How would you figure out if it wasn't cubic,
and we don't have this handy rule that makes our life easy?
Any ideas?
Yeah, how would it work?
How do you know that that would give us something normal?
So basically, cross product says you have vector 1 and vector 2.
If you take the cross product of those, you'll always get something normal to it.
So all you have to do is say, all right, in this plane,
I know maybe what this direction is.
I know what that direction is.
I could take the cross product of those two, and I would get something normal.
And that works for cubic as well, but we don't have to use it because cubic is easy.
Whatever the plane is, same numbers are the direction.
So if you don't remember how to do length of vectors and dot product,
I've written it out here,
the angle would just be equal to the inverse cosine of all of that jazz,
where u1, v1, w1, u2, v2, w2 are your two directions
you're trying to figure out the angle of.
So with that in mind, let's figure out this next clicker question.
So it says the following.
It says, determine the tensile stress that has to be applied
along the 1, 1, 0 axis of a silver crystal
in order to cause slip on the 1, 1, 1, 0, 1, 1 slip system
if the critical resolve shear stress is 6 megapascals.
So let's remind ourselves of the formula here before we start.
We know that tau CRSS is going to be equal to the applied stress
times cosine of phi times cosine of lambda.
And that is our equation for the angles.
So I'd like you to give your answer within one decimal place
and give the answer in megapascals, I'm sure.
Anybody need help getting started?
It's just rearranging the one that I did give you.
It's just that one.
It's just rearranging it.
So from this formula, you can solve it.
Do you need help getting started?
Real quick announcement.
So theta, what is theta?
Theta, this is just the angle, a generic angle between any two vectors.
If you have vector 1 right here and vector 2 right here,
theta is just the angle between any two generic vectors, right?
So you plug in vector 1 and vector 2.
So for example, this might be your loading direction.
And this one would then be maybe your normal to slip plane.
And if you do that, then you get one of the angles that you're looking for.
Then you get phi.
You'll solve for phi, right?
Then you do it again, but this time, instead of slip plane,
you're going to do it to slip direction,
between the loading direction and slip direction,
and you'll get lambda.
So this theta is just a generic angle between two vectors.
Other questions?
OK.
OK, I'm not seeing a lot of right answers,
so we're going to give a little more time on this.
Your answers should be ballpark 14-ish.
So keep working.
We won't stop just yet.
Keep working.
We'll wrap up.
Okay.
We're going to wrap it up.
Get your answers in.
Can we end?
I'm going to end.
Okay.
Answers in.
I'm going to end.
Okay.
Good.
Most people got the correct answer which is 14.7.
So let's actually work through this together.
So I told you in the question that the resolved shear stress which again that's a material's
property.
This thing, whenever the resolved shear stress hits six mega pascals, this thing is going
to start to slip.
So the point at which it hits it, it's going to be the stress that we're solving for times
cosine of phi times cosine of lambda, right? So let's do cosine of phi first. Cosine of
phi is going to be equal to A dot B over the length of A over the length of B, right? So
again, the two angles here that we're talking about for phi, and it doesn't matter which
one's which, but phi technically is between the loading direction and the slip normal.
So that's between, that's the dot between, excuse me, 1, 1, 0 and 1, 1, 1. And then
here this is the length of the 1, 1, 0 and the 1, 1, 1. When I do that, I find that that
is equal to 2 over square root of 6. Now what's great is you don't actually have to solve
for the angle itself is, because in this equation up here, all it wants is the cosine of that.
You just solve for the cosine of that. So you don't actually have to plug in the angle
if you don't want. Save yourself the mathematical step there. So let's do the next one. So cosine
of lambda, again, same equation, but this time it's going to be between 1, 1, 0 and
0, 1, 1. So we're taking the length of the 1, 1, 0 vector times the length of the 0,
1, 1 vector. When I do that, I get 1 over square root of 4. So plugging in, I solve
for stress should be equal to 14.7 megapascals. Any questions on how we did this? Any confusion
at all there? Yeah, Martin? So I gave it to you in this question. Obviously, a question
like this will be on the next midterm for sure. I always use one, because there's lots
of ways I can ask questions around it, right? I can say, I wouldn't do this on a midterm
because you wouldn't have time, but on a homework question, I could say, this resolves shear
stress and this loading stress caused slip to occur. What slip system might have been
involved? And you'd have to guess at what the slip system might be. So I'm not going
to do that on a test, because that would take too long. That's a reasonable thing on a homework
though, because if I said it's an SEC crystal, then you know that the slip system has to
be the 1, 1, 1 plane. So then all you have to do is figure out which direction it happened
on. So you'd set up a little Excel spreadsheet. It wouldn't actually be too bad. So there's
lots and lots of ways to ask these questions, because all it is is, it's one formula, but
it has to do with some vectors, right? So this tests whether or not you understand vectors.
It understands whether you understand like Miller indices and notation, you know, the
curly versus square versus whatever brackets, right? So lots here to cover. Any other questions
on it that you was unclear about? Anything else? So again, the two vectors, or the two
angles we care about are always the loading direction and the normal, the loading direction
and the slip direction, right? Those are the two angles we have to figure out. And they're
actually not so bad to do. Okay. So plastic deformation is different in polycrystalline
materials. This equation is pretty handy, right? But it only works for single crystals,
right? Because in polycrystalline materials, we have the problem that every single one
of these little domains here is a different grain boundary. It has a different orientation
and our equation depended on understanding the orientation of the crystal, right? So
in fact, the strength of polycrystalline materials is going to be stronger than single crystalline
materials, because let's say you have a dislocation present right here, and you know, you're pulling
on this thing in such a way that this thing wants to move that way. Well, when it gets
to a grain boundary, because the orientation of the adjacent grain is going to be different,
it's going to have to redirect its direction a little bit. Maybe it's going to have to
move to like that to move through the next grain, right? So every one of these grain
boundaries serves as a scattering site or a resistance of that dislocation to try and
move through the material. Therefore, the more grain boundaries you have, the stronger
your material, right? And this is actually a pretty common tool. This actually gets to
Hall-Petch in a minute. If you can just introduce more and more grain boundaries and use smaller
and smaller grain materials, it'll get stronger and stronger, okay? Okay, before we talk more
about that with Hall-Petch, let's talk about deformation by twinning. So if a dislocation
can't move through a crystal, maybe because it doesn't have very many crystal slip systems,
right, or because grain boundaries are causing it problems, you can have twinning occur,
right? So twinning, I think we mentioned this, that a shear force, right, so they're applying
a force this way and that way, so they're shearing it. It causes a collective motion
of atoms, right, in the shear or the, sorry, the twin region or the shear region, okay?
This picture's not strictly correct because a twin should always have mirror images on
both sides, and that's not quite mirror. It's off by a little bit, but it's pretty close.
So twinning can also occur. So both of these mechanisms can both produce deformation, right?
On the left, we've got slipping occurring, right? And in here, remember in the slip case,
the orientation is preserved throughout our single crystal, but in this case, the orientation
is not preserved, but we can achieve deformation by both routes, right? So twinning will be
more favorable or more probable if the slip system is unfavorable. So if you can't have
slip, you'll take twinning, and that's exactly what you see with the tin cry demo at the
beginning of the class, right? In the first case, at room temperature, she took this piece
of metal, and as she bent it, you could hear it crackling. What do you think that crackling
is? Is it slip or is it twinning? What would make noise? What do you think should make
noise? Slip or twinning? So twinning is cooperative motion of many atoms. That's a sound wave,
right? When a bunch of atoms shift, that creates a sound wave. It's a pressure wave that would
then go through the material. So twinning creates that crying, that crickling noise.
That's literally the shift of atoms. Slip doesn't have that because it doesn't have
as many atoms all shifting at once, right? Yeah, so crying occurs during the twinning
deformation, but then when we heat it up and we give it some thermal energy, all of a sudden,
the more favorable mechanism is slip, and so it's able to deform via slip, and you
don't get that crying motion of the collective motion of atoms. Okay? Any questions on that?
Yeah?
So it seems like when it was heating up after it was bent, it seemed like it still retained
most of its strength, but when it's slipping, it seemed like it would lose some of its,
I guess some of its strength would be lost.
That's a great segue into the next section. So hold that thought. We're coming to that.
As you deform these things, what does it do to the strength is really the heart of
the question there. So let's come back to that in just one second. Okay? So first off,
when we talk about all these mechanisms of deforming or strengthening of metals, there's
always a trade-off. You can never have it both ways. If you want to increase your strength,
then you lose toughness and you lose ductility, right? So this is like your general trade-off.
Increasing strength comes at the expense of ductility. If you want more ductility, you
have to lose some strength, right? And the reason why that happens is because this next
point here, that anytime you restrict or hinder dislocation motion, you render the material
harder and stronger. If you make dislocation motion easier, you make it weaker and softer.
So the blacksmith takes his piece of metal and he puts it in the forge and gets it glowing
hot, and now all of a sudden it's easy to deform. The reason it's easy to deform is
because you gave it a ton of thermal energy so these slips can move past each other a
little bit easier, right? When it cools down, it gets harder and stronger because those
dislocations can't move past each other or another. So this is a really important take-home
message for material scientists, mechanical engineers alike. If you do anything to make
it harder for them to move, you made a harder, stronger material, but you sacrificed ductility,
right? So we mentioned that reducing grain size will strengthen a material, and it does
that exactly because grain boundaries serve as scattering sites for dislocations. So you
see that there's sort of a linear relationship if you were to plot the hardness or strength
hardness against the inverse square root of grain size, right? So here's the actual
expression. Your yield strength is going to be equal to some minimum yield strength plus
kd to the negative one-half, where d to the negative one-half, that's your grain size
to the negative one-half. So plotting here, that's what they're plotting here, right?
They've got d to the negative one-half here, so basically small grains, yeah, small grains
and large grains. You see your small grain material is much harder, like a factor of
50% or 75% higher hardness, okay? And this is simply because dislocations can't move.
So because this is nice and linear over a nice range, we can actually fit it, and that
slope will give us k, which is going to be, well, k and your sigma naught, those are your
Hall-Petch constants, right? Which is great, because if you have 316 steel and they say
how strong it is, well, it really depends on how big the grains are, right? So it's
not useful for them to say, like, hey, this 316 steel is such and such a strength. They
should really say what the average grain size is, right? Because then you're going to know
why it has the strength it has, and it's going to be due to this Hall-Petch equation. It
also means that if you do something to your material that has small grains to cause them
to grow, which we'll get to in a moment, you're going to weaken your material, right?
So if you're planning on using something at high temperatures or some temperature where
you're going to see grain growth, don't be surprised if your strength goes down from
what the spec sheet says, basically, right? Also, as you go to smaller and smaller grains,
therefore more and more grain boundaries, you increase toughness because you impede
crack motion as well. It's not just dislocations that can't move, but whole cracks get blocked
by grains, right? So you increase the fracture toughness.
Okay, solid solution strength, and we've said this a couple times in class now, but like
sterling silver is a great example of this. Start with pure silver, add a little bit of
copper. Since copper is not the right size, it has a strain field around it. Therefore,
dislocation is going to want to hang out with that copper atom, and it's not going to want
to move. If you made a dislocation harder to move, you strengthened your material, right?
Okay, strain hardening or cold working. They call it cold working because as you plastically
deform something, it's typically lower than its melting point that you're doing this.
If you do it much below its melting point, then you're not constantly fixing all the
damage you put into it. So if I take, this is my piece of metal. Every time I pound on
it, I introduce dislocations into it. If this is at a low temperature, those dislocations
can't reverse and heal themselves and annihilate. Therefore, they're still there. Therefore,
every time I hit this with a hammer, I'm just adding more and more dislocations, and they
start to resist one another, right? So if you don't ever fix that, it's just going to
get harder and harder and more and more likely to fracture, which is why a blacksmith can't
just keep on hitting his piece when it's out of the oven because eventually it's just going
to fracture, okay? We'll skip. What is cold working? Great question. Why would you want
to use cold working versus hot working? By the way, hot working, do they actually use
it? They absolutely do. Here's a picture of how it's made. My kid loves these. So take
a look at this. This is how they make aluminum foil. Believe it or not, it starts out with
thousands of pounds, giant ingots. All traces of the liquid used to cool the decrusting
knives have to be eliminated. So they cast the ingot first, then what do they do? The
many steps in the thinning of the ingot begins. The aluminum block is crushed by the hot mill
rollers. Temperatures in the rollers are between 455 and 540 degrees centigrade. Pressure on
the ingot is continually verified by a technician. If too great, it's lowered, and the technician
then adjusts the pressure. The heat is so high that the ingot risks sticking to the
mill's roller. To prevent this, everything is cooled with a liquid that's 95% water
and 5% oil. They have to have that oil so it doesn't stick to the rollers themselves,
the ingot. And they just do this over and over and over. Starting from a thickness of
45 centimeters, the ingot becomes increasingly thinner with each pass through. Depending
on requirements, the ingot will go through the machine between 12 and 16 times. The ingot
now measures 7 centimeters in thickness.
So it's clear why they do it at high temperatures, right? It's just easier, and it's constantly
healing out the deformation so it's less likely to fracture. So why would you ever do cold
working, meaning you're deforming it at low temperatures? Any ideas? Yeah, Eric?
Yeah, you gonna say the same thing, Steve?
Yeah.
If you want your final foil to be more stiff, meaning it's more likely to break, but a higher
strength, especially in the rolling direction, yeah, you could just do a cold working, right?
So for a lot of applications, they do do that, and it produces a stronger... So maybe like
for your final pass, it's too weak because you've been doing it hot, but your final pass,
you might do that one cold to give it some higher strength. Matt?
You also have a hot surface, right?
Sure. Anywhere where you do cold working, you're gonna make a more brittle area. And
if you wanted to then use it as like a design feature where now it's going to break specifically
along wherever you did the cold working, yeah, absolutely. Okay, so all this talking about
introducing dislocations, getting rid of them, having it become harder, having it go back
to more ductile can be described in terms of words that we use in material science called
recovery, recrystallization, and grain growth, right? So we already said that plastic deformation
causes more dislocations. It changes the grain shape and the orientation, right? So this
is here, right? If you start out with like normal equiaxed grains, equiaxed means that
they don't prefer one orientation, right? But as you deform them, they're going to
orient themselves, right? These things become stretched out and long perpendicular to how
they're being rolled. So we've changed the grain shape, right? And their orientation
even, right? So we increase dislocations, we change the shape of grains, and we certainly
are gonna add a lot of strain hardening. Every time you hit it, dislocations have the strain
field around them, the strain fields interact with each other, and it can't move around.
So it becomes harder and harder, right? So all that happens, making it more and more
likely to break, and your final product, you don't want it to break, so they have to recover
those things, right? So they anneal it, which is just a heat treating step meant to remove
the effects caused by plastic deformation. They call it recovery as the first step because
internal strain starts to be relieved by dislocation motion, meaning let's say you had an initial
dislocation like that, another one like this, you add a little bit of thermal energy, this
thing can line up with that one, they feel an attractive force, they combine and they
annihilate, right? So it got better, right? There was recovery. So that's the first thing
that occurs when you give it a little bit of heat during the annealing, right? If your
system is so messed up and has so many dislocations and you provide enough thermal energy, your
crystal might just say, you know what, it's easier to start from scratch. Like when I
go in a room sometime and somebody's had like the chairs like all over the place, sometimes
rather than trying to take the existing chairs which are all over the place and try and make
a nice grid, it's easier just to push all the chairs to one side and start lining them
up again in some sort of nice arrangement. That's the difference between recovery and
recrystallization. Recovery says try and move the individual chairs to make it work within
the existing lattice. Recrystallization says it's too broken, let's just crystallize new
bond or new grains entirely, right? So you form entirely new strain-free grains. These
will be equiaxed, meaning that they don't have a preferred orientation, they're going
to be random and sort of spherical shaped as opposed to long and drawn out maybe. These
grow out of the parent grain and then by doing so, you get mechanical properties that are
restored to the pre-cold work set, right? So where do we see this happening? So here's
our initial material, right? So this is increasing ductility and increasing strength on different
axes, right? So that's either more ductility or strength. So hardness, as you cold work
it, hardness goes up but ductility goes down, right? As you're cold working it. You put
it in the oven and you start to get recovery and you're going to see moderate, very small
changes to hardness and very small changes to ductility, a little bit though. By the
time you get into recrystallization, your hardness is going to plummet, right? And your
ductility is going to really take off to kind of where they were when you started, right?
Then the last step is grain growth, right? So you started out growing these tiny new
little strain-free grains, right? So here's our new grains starting to crystallize, right?
Recrystallization, right? So those are new strain-free grains. They're not stretched
out. They're nice and equiaxed. They grow and then you've got a whole bunch of them
once you've gotten rid of all the old grains. Well, if you leave this at high temperatures
for a while, these small grains will grow into large grains. And the reason that happens
is due to something called Ostwald ripening. The book calls it Ostwald ripening. All that
means is if you have small grains, those ones get swallowed up and big grains are the ones
that swallow them. So small grains disappear, big ones get bigger. So if this happens and
you let it go over time, they eventually swallow up all the small ones. And you can see that
happening, right? Lots of people study this and you can see it going on here. So here's
a computer simulation of a bunch of small grains. And if they leave it at a high temperature,
right, in the model, you can see that these small grains are getting eaten up by the big
ones. So over time, they're getting swallowed up. Now, what's the driving force for this?
Surface energy, right? Every surface has an energy. We don't like it, so we'd like to
get rid of them. So it happens. Now, the mechanism we won't get into in this class.
You'll probably learn it in other ones. My ceramics, we talk about it. But basically,
big ones get bigger, small ones get smaller. It's a sad commentary on living life, right?
And this is also what ruins your ice cream, right? Actually, when you buy ice cream from
the manufacturer, it's quite small grain, which is delicious, right? But when you let
this thing, you take it out of the fridge for a minute and you serve it up and then
you put it back in, you expose it to a higher temperature. So you gave it thermal energy
for those grains to grow and you also gave it energy for the cream and the water to separate.
So when the cream and the water separates, now your ice crystal is there and it's not
tiny and it's going to grow. And every time you do this, that ice crystal is going to
get bigger and bigger and bigger. By the time you've like forgot about it for a month and
you come back to it, it's just gross. Big-grained ice cream is super gross, right?
And on that note, we'll wrap up here today. Happy Halloween. Have a good time.

