11/2/2016 Intro to MSE deformation in different materials
1 Video View·Oct 15, 2023
Callister Intro to MSE, materials science, viscosity, WLF equation, wilhelm-landau-ferry viscosity, polymer lamellae, stress relaxation, carburization, diffusion, jump distance, concentration gradient, thermally activated motion
I hope the homework was okay and doable.
Only 92 people watched the video on Weibull, and there's definitely more than 92 people
in the class still, so I don't know how you did that.
I hope if you're one of those people that didn't and you struggle on the Weibull question
that you will get help because that's not going away.
There will definitely be at least one question on that on the midterm three, and I usually
put it on the final too, so that's something that's not going away.
You definitely need to learn it, because like I said, I think it's one of the key things
that you're going to get out of this class that's most useful.
Yeah, Chris?
Is it Chris?
When is midterm three?
It's a great question.
Midterm three is currently scheduled for next Friday, so not next class period, but the
Friday thereafter.
That conflicts with a bunch of people who have multiple exams that day, so I'll never
write two exams, because it's just too hard.
It's too hard to make sure that it's fair, that it's fair to everyone who takes it, but
one thing I am considering is potentially we could take it the Monday thereafter.
I need to talk to my TAs, because that means we'd have to find a time to grade it, and
it's much easier to grade it on the weekend than to grade it on a Monday night, so if
the TAs aren't terribly opposed, now's the time to write them pleading letters, then
potentially we could do it on a Monday.
It's not working.
It's not great.
So, by next Friday, I will have an answer on when the exam will be, but plan on it being
the Friday thereafter, right?
Is it November 11th, I think?
Plan on it being November 11th, unless we can have a change of heart.
Oh, so if Monday's no better, then yeah, we're probably going to, okay.
So that said, let's get started on today.
So today we're actually a little bit behind, which is not a problem.
So today we're going to wrap up Chapter 8, which I'd hoped to finish before, and there's
not a lot left on Chapter 8, because the hard part is deformation in metals, because we
do the whole single crystal thing, and that was useful, and it's good that we spent the
time yesterday in class working that out, because I want it to stick, but today we do
need to talk about deformation in ceramics, and then we need to talk about deformation
in polymers, and there's a couple key points to cover there, and then I hope we have time
to actually start Chapter 6, which is diffusion.
So our learning objectives are to talk about mechanisms for deformation in ceramics, polymers,
glasses, basically everything else other than the single crystal we've talked about thus
far.
We'll talk about viscosity a little bit more, and why it's useful for talking about deformation,
and then finally we'll relate things like molecular weight, crystallinity, to deformation
in polymers, and the heat treatment of polymers.
So ceramics.
Ceramics, we know that they don't deform very much.
This piece of chalk, this is a ceramic, right?
I can try and bend this, but it's not going to deform, right?
I'm bending, I'm bending, and it just breaks, right?
Because it has no what?
This is a good question.
What does it lack that other materials have why it does not bend?
So a common response based on what we talked about already, it's like, well, it has no
twinning possible, or it has no slip dislocation, like slip systems, right?
So that's kind of the answer, but it's more complicated than that.
It really comes down to the fact that you have anions and cations.
So take a look at our system here, right?
Even slip can be, it's uncommon, not because you don't have highly packed planes.
You can have highly packed planes and highly packed directions, but the problem is, think
what happens when you try and slide this thing past, right?
So pick a plane, right, like in the drawing, a plane like that.
If that happens to be our slip system, and let's say it's slipping this way, what will
happen as you slide this half of a bond length over, now all of a sudden this negative is
going to be next to another negative, and this positive will be next to another positive,
right?
And that leads to really strong coulombic, basically repulsion, right?
So it has to overcome a very high energy to slide past one another, and then when it slides
to its full position over, like by the time that this negative anion has gotten to that
point, then it's back to normal, but it's getting there that you have to overcome a
really high, a prohibitively high activation energy, right?
It's just not likely.
You have to have a ton of thermal energy for that thing to slide past it without it just
saying, you know what, it's cheaper in terms of energy for me to just break.
And so oftentimes, if you're at low temperatures like that chalk just was, I'm bending it,
and instead of twinning or slip dislocations starting to move, it just says, ah, I'm just
going to break.
It's just not worthwhile for me to try and get that close to ions at the same charge,
right?
So it doesn't happen.
It has to do with electrostatic repulsion as they slide past one another.
Since metals are electrically neutral, you don't have that problem.
Covalent ceramics, slip is also uncommon, but here it's not because of electrostatic
repulsion.
It's because covalent bonds are so strong, right?
To break that covalent bond, imagine diamond, right?
Diamond is a bunch of carbon bonded to carbon in a three-dimensional lattice.
If you want to make that thing slip, you have to make carbon break a carbon bond, right?
That releases a lot of energy.
At that point, it's more common for it to just fracture, right?
And they often do have complex structures.
Because covalent bonding is directional, it leads to complex structures, which oftentimes
means you don't have these nice, really high-density planes, really nice high-density directions
to slip on.
Okay, what about glasses, non-crystalline ceramics?
Question?
No?
Okay.
So deformation is still not likely in glasses, which are non-crystalline ceramics, but now
it's because you don't have a regular atomic structure.
You don't have a really nice plane.
Like in this row of students here, right, we said that I could have slip, I could take
this couple rows here and slide them that way and it's not a problem.
But in the glass, you don't have nice regular rows.
It's just a hodgepodge jumble of atoms that are sort of frozen in place, right?
And so you don't have slip systems essentially, okay?
Instead, how do we get deformation?
You get it via viscous flow.
Glass, if you heat glass up, all of a sudden it becomes something that can flow really
easily, right?
So my students that did a glass lab that recently, they actually made their own glass.
They got it good and hot.
They got all the bubbles out and then they cast it onto these hot plates and they literally
poured it like you pour honey.
It doesn't flow like water unless you get it super hot, but you know, it flows, you
can actually flow it.
So this rate of deformation is going to be proportional for one thing to the applied
stress.
If you're pouring it, that's just gravity, right?
Gravity is pulling on it.
But you could also put this in like an injection molder.
You could apply a stress, right, and they do that.
And it's going to be proportional to that as well as the viscosity.
What's viscosity?
We've heard this term before if you've taken transport.
Viscosity is a measure of how sort of thick a liquid is, right?
And at some point, a liquid goes from something really thin like water or water with soap
in it, right, that's really thin, to really thick like honey or, you know, even thicker
to like honey that's cold, right?
So that transformation from something that flows to something that doesn't flow as a
fluid, we use viscosity to describe it.
Literally how it's measured is you take a moving plate, you drag it at some velocity.
Let's say this is like the surface of my liquid I'm testing.
I drag it, I know the velocity of the plate that I'm dragging, but that velocity gradient
occurs between the top plate and the bottom plate, right?
So I have two plates, I'm moving the top one, the liquid that's touching the bottom plate's
not moving, the one at the top is moving.
So this gradient in velocity of the liquid that's moving, that basically is proportional
to your viscosity, right?
Your viscosity is equal to the shear stress you apply divided by this gradient in velocity.
Any questions on that?
Sometimes that's hard to, I think, grasp for students.
We're not going to say more about this or how it's measured in this class.
When you take transport, mechanicals and materials, we'll have to take that, you'll learn way
more about it.
So I'm not going to say more.
Last chapter we mentioned that the glassy transition temperature is a temperature defined
at the temperature at which your viscosity is equal to 10 to the 12th pascal seconds.
That was in last chapter, I double checked the notes, we're just confirming that again
here, right?
That is a totally arbitrary definition, right?
Totally arbitrary.
They just picked it.
They said at what point does a glass kind of start to move around a little bit?
When the viscosity is around 10 to the 12th pascal seconds, right?
Totally arbitrary, but the whole community has accepted that, right?
That said, viscosity can be calculated at any temperature using a series of different
expressions.
In this class, the only one we're going to use is the WLF equation, Wilhelm Landau-Ferry,
which says the following.
The viscosity at any arbitrary temperature is equal to the glassy transition viscosity
10 to the 12th pascal seconds times exponential of this, negative 17.4 times the temperature
that you're at minus the glassy transition temperature divided by the quantity of 51.6
plus that temperature minus the glassy transition, right?
So these WLF constants like this one and this one, those technically are constants for a
given material, right?
You have to look them up for a given material, right?
These are general values that work for a lot of things, but if you want to be very specific,
you have to know what those constants are for a system.
So if you buy motor oil, that might be something that's provided on the spec sheet or something.
What are those constants?
So you know, all right, how's the oil in my car going to change when I start it and the
oil's thick because it's cold versus when it's running and it's running at whatever
temperature, some high temperature, right?
And now it's thinner.
Okay, let's talk about polymer.
So that's really it.
You really can't deform ceramics and glasses in other ways than that.
Ceramics, if you want them to like have slip dislocation, you've got to heat it way up
and I'm talking like really, really high temperatures before you get even nominal deformation.
It's not going to be a lot, but it'll be some.
And then in glasses, again, you've got to heat it up so it can flow, okay?
Very different than the slip and the twinning we talked about previous.
Good question.
So I've added the negative from the constant in there, that's a good question, right?
So you could use this as is, that equation, as it's written, that's a great question.
Okay, what about polymers?
How do they deform?
Different, very differently, right?
So we're going to talk about crystalline versus semi-crystalline.
First let's talk about semi-crystalline polymers.
We talked about this a couple chapters ago when we talked about the structure of polymers.
There's regions of the polymer that are kind of like this middle section of the row where
you've got like folds of the polymer folding back on itself and there's bonding between
these layers.
It's not super strong bonding, it's van der Waals, right, but it's bonding, right?
Along the chain, like along the chain, that's where it's strong bonding.
But between these things, right, which we called, again, the same word as we'll know
when we talk about phase diagrams, but a kind of different meaning.
So between these sections, you've got a region of something else and then you've got another
section, right?
So what happens in here?
This is where the change is random, right, so you've got lots of organization separated
by sort of randomness, right, in that top left picture, kind of where you can see a
diagram of that.
So if you pull on this thing, particularly if you pull on a different direction than
how the seams are oriented, right, so this middle row is oriented like that, so if I
grab this polymer and I start pulling it this way at some odd direction, right, like that,
what's it going to do?
It's going to do the easiest thing first and the easiest thing is to stretch out the amorphous
region, right?
That amorphous region was already not very well bonded, so stretching and breaking those
bonds, there's not many of them to break, so it's going to stretch first, right?
So overall, you do get some elongation as you pull it up and down, right, by doing that.
If you keep on pulling it, the next thing it's going to do, it doesn't want to break
these bonds yet, but it can slip them, right?
It's going to rotate, right?
So if I was pulling this whole row of students, the rows would sort of orient toward the direction
you're pulling it, right?
They're not breaking yet, they're just, some energy went into orienting them in a different
way, right?
Keep on pulling it and finally this row is going to start to break up, sections will
come forward, sections will go that way, right, it's going to break up a little bit, and you
keep on pulling it, you're finally going to get really a dramatic example of going from
four to five.
It's just more of the same, but more dramatic, right?
That's deformation in polymers, right, very, very different than what we talked about in
all the other systems so far.
Matt?
Parts of the solid regions actually do get unraveled a little bit, right?
These lamella, the ratio of lamella to amorphous would probably shift towards amorphous a little
bit.
I couldn't say how much, but that's the concept, right, at least conceptually what's going
on here.
So these are the steps, right, we've talked about them.
Spherolites, which is the overall organization, spherolites were like the blocks of organization
separated by amorphous, another block, right?
It had like this big region.
Those get destroyed in this process, right?
We went from something that was more ordered to something that is now stretched out and
changed the order in it, okay?
So interestingly, when you heat these back up, you restore the spherolites and the sample
is going to shrink back, tending towards the shape it started at, right?
Very different than metals, right?
You took a metal and we cold or hot worked it and we achieved deformation, but when you
heat that back up, all you do is you either get rid of dislocations via recovery, or
at most you do recrystallization, you grow new grains, but the deformation, like this
object is permanent, it stays put.
Polymers, it's different, right?
When it's in this state right here, if I heat this up, I'm basically saying here's some
energy, lower your energy.
Use that thermal energy to lower your system's energy and it says, great, I can do that by
bringing these together, by closing it up, basically by trying to get back to that state.
And so when you heat a polymer up after it's been drawn or something, it's going to shrink
back to its original shape.
It's going to tend to, right?
Because you're giving it thermal energy for it to go from a higher energy, strained state
back to something where it wants to be, okay?
So that's completely opposite of metals, okay?
Because polymers can do this, we already talked about this, we talked about stress relaxation,
so I'm just going to remind you again, here's the formula.
We'll see it at some point.
We saw a variation on the homework this week, we'll see it again.
But again, if you load it up to an initial stress, stress not, so again, that's our initial,
that's our initial stress, right?
If you know what the modulus of your polymer is, how long it's going to be held there,
and the viscosity of your polymer, you can calculate what the stress will relax to after
a given time at some temperature, right?
So that's our quicker question.
Let's start with that today, which says the following.
At the glassy transition temperature, you observe that a stress of 100 megapascals produces
10% strain, okay?
What does this stress decay to after two minutes?
Give your answer in megapascals, round it to one decimal place.
And I'll remind you that at the glassy transition temperature, by definition, this equals 10
to the 12th pascal seconds.
And I'll also remind you that technically things inside of an exponential have to be
unitless, so make sure your units cancel or you won't get the right answer on this.
Your units have to cancel.
Does anybody need help getting started?
Yeah.
Okay, one more minute, get your answers in.
Okay, final answers in, closing the poll.
Are we good?
Just a second.
Yeah.
Why does this pen always break?
Okay, I'm going to close the poll.
We're good?
We're good.
Okay, what do people think?
Okay, great.
Majority of you guys, three quarters of you get the right answer, 88.7, so how do we get
that?
Where'd it come from?
If my pen's going to work, we can work it out.
Guys, it's not working.
All right, guys, bear with me.
My pen stopped working.
This is the worst thing ever.
All right, I'm going to type it out.
It'll be even easier just to type it.
Just call it.
Yeah.
Okay, so in this case, E equals the stress over strain, right?
That's way too small to read.
Let me fix that.
All right, so that's equal to 100 megapascals over 0.1, which means our modulus is equal
to 1,000 MPa.
Right?
Okay, so now we've got our modulus, so now we can just plug it directly in this equation.
We can say that sigma, our stress, equals whatever our applied stress was, which was
100 MPa times exponential of what?
It's going to be, so we've got negative 1,000 MPa.
Let's convert megapascals into pascals, so the units cancel, right?
So now it's pascals, and then we need this to be in seconds, so the seconds cancel.
Two minutes is 120 seconds, right?
So this will be over 10 to the 12th pascal seconds, right?
When you plug all this in, you get 88.7 megapascals, megapascals because this has no units and
the megapascals carried over, right?
Any questions on this?
Any questions?
Okay, let's keep going.
Okay, a few more things to say about this.
Molecular weight, the modulus is unaffected by molecular weight generally, but the tensile
strength increases with molecular weight, because molecular weight means it's longer
and longer chains, so the likelihood of this chain tangling up with another chain over
here and pulling on it and getting stronger, pulling on the covalent bonds, increases,
so tensile strength goes up because of basically more entanglement.
The degree of crystallinity, if you go from something less crystalline to more crystalline,
you're going to change the bonding, right?
Secondary bonding is prevalent in the crystalline regions, that's where there's good overlap,
so basically if you increase crystallinity, you increase modulus, and you also increase
tensile strength, but it also becomes more brittle, right?
Because the amorphous regions are the ones that are able to flex more easily.
Pre-deformation by drawing, a common way that we work with polymers, the way that you get
like your grocery bags, things that are polymers that are really, really thin, they start out
with a melt, and they take that, and they, like grocery bags, they make by actually
inflating them, it's a pretty interesting process.
They inflate it, and as you inflate a balloon, it's going to get thinner and thinner and
thinner as you fill it up, right?
So that's how they get really nice thin things for a lot of polymers, or they'll draw it,
like if you want fibers, and you want strong thin fibers, you draw it down, so climbing
ropes and that, that you want really high tensile strength in one direction, they'll
take it, and they'll draw these fibers, right?
As you draw it, you go from, you know, in the molten state, there's going to be kind
of like this, and you're going to turn it into this.
So what happens to the strength?
Well, you lined up a bunch of these chains, now when you pull on something, you're pulling
on the polymers in the chain, rather than just stretching the chains out, right?
So it's going to be a lot stronger, and a lot more likely to, it's going to be less
ductile as well, right?
So it's just like strain hardening, right?
It strengthens them, it produces oriented polymers, these are obviously anisotropic,
if you draw a polymer this way, you're going to have a lot of strength in that same direction,
and transverse to that, you're going to have a lot less strength, right?
And it can be pretty significant.
If you, for example, take something that's oriented, and then you pull 45 degrees off
of that, you could see strength reduction by factors of maybe 2 to 3, right?
Pretty huge drop in strength.
If you draw it at high temperatures, then you have to quench it to retain those high
temperature properties, right?
Because thermal energy says this polymer is going to want to shrink back down and go to
a nice spherulite.
So if you want to maintain these high temperature properties, you've got to quench it, right?
You have to cool it down, get rid of that thermal energy quickly, so it keeps its high
temperature structure.
And then, like I said before, if you do perform heat treating, you're going to go back to
the spherulites, right?
It's going to go back to how it was when it started, and in doing so, you become more
ductile and you lose strength, right?
Opposite effect of what happens in metals.
Deformation of elastomers.
Elastomers are a lot like polymers, but now we add something, we say that they're cross-linked.
So instead of just van der Waals secondary bonds, which are holding these things together,
now you add a few really strong covalent bonds periodically by adding the vulcanization
agent, right?
Sulfur or something.
So you get very similar properties, but it has more to do with chains catching and entangling
due to those cross-linking, right?
And that is all I'm going to say about that.
So let's move on to chapter six.
Chapter six is on diffusion.
So here's our learning objectives.
I think my pen has decided to work again.
It's back.
Remarkably.
All right.
So the learning objectives for this chapter are, let's talk about the needs for diffusion
in real engineering problems.
Is this just something that like, all right, scientifically exists or whatever?
Or is there actually a need to understand this?
Yeah, you bet you there is.
We'll talk about how atomic motion and the equations that describe them can be used to
govern diffusion.
And if we have time, we'll talk about activation energy and the potential gradient, chemical
potential gradient.
I don't know if we'll get much further than that, so let's get started.
So where do we use this?
We use it all over the place, right?
A couple of chapters ago, we talked about case hardening, right?
We said that if you want to strengthen material, the surface is the best place to start.
And they did something called shot peening.
What was shot peening again?
What was accomplished by shot peening?
Yeah.
We know that things usually break easier under tension rather than compression, and things
start at the surface because the flaw size is half of what it would need to be in the
inside due to Griffith's Fracture Equation we talked about.
So shot peening basically deforms the surface like crazy and puts it under compression.
Therefore, to break under tension, it has to overcome the compression first, right?
So there's another way to do it.
Shot peening is one way.
Case hardening is another, sometimes called carburizing, and we do this in my own research
group, actually.
Let me pull this up.
So if we go to Meeks, Shetty, Sparks, Tantalum, Carbide, we just wrote a paper on this.
I think I'll have that be part of the homework due next week.
But basically, we started out with a material that wasn't as hard, but it was really tough
So we wanted to make this thing both hard and tough.
We want the surface to be hard so it can be used as a cutting tool, but a cutting tool
does you no good if you go to cut something and it just fractures, right?
So we wanted it to be nice and tough, which the bulk already was, and we needed to introduce
carbon from the outside into the material by carburization to make now a functionally
graded carbide, basically.
So you'll learn more about that on the homework this week, I think.
But they do this all the time, and they've known about it for a long time.
If you have old antique guns, I think they call this patina, it's like the collector's
term for when a gun or whatever artifact or antique has that sort of tarnish on it.
That tarnish is, you know, the function of diffusion of stuff going in there.
It can also be oxidation, right?
But in this case, they've intentionally taken a steel and they put more carbon into it.
What do we know about steel as a function of carbon content?
Which one's harder?
More carbon.
And why?
Why did we decide that steel's got harder but less ductile with more and more carbon?
Yeah, let's imagine two scenarios.
One under equilibrium where the phase diagram reaches whatever the equilibrium is, and then
the other where you're just forcing something in and then you freeze it.
You don't give it time to reach equilibrium.
Under equilibrium conditions, the further you are towards the right where there's more
carbon, you're closer and closer to the cementite phase.
That is a ceramic.
No surprise that it's harder and more ductile, or less ductile, right?
On the left, you've got pure iron, right?
So that's the first reason, under equilibrium.
But this is not necessarily happening under equilibrium.
They start out, they put the gun or the artifact, whatever it is, in a atmosphere that has lots
of carbon, right?
They can either coat it with carbon or they use like carbon monoxide or carbon dioxide
in the air.
And they're driving carbon in.
And you don't let it completely equilibrate all the way through the material.
Instead, it's just in the surface.
So your carbon concentration, right, as a function of depth, x, it might have had like
some value before, right?
That's like the steel, how it was made.
But after you do this, it's going to be higher.
It might be something like that, which causes this region to have a higher strength.
It's going to be more, less ductile, but it'll be higher strength because you stuffed carbon
in all the additional sites.
And what that does, why would that strengthen something?
Actually, this is a good question.
So you had interstitials, right?
So this is our steel, right?
I take a bunch of extra carbon atoms, I'm stuffing them in here.
Why did I make the material stronger and more scratch resistant?
Yeah, because we said for deformation, that's scratching, right?
So for me to slide this row of atoms over and achieve deformation, I stuffed extra atoms
in there that messed up our slip system, right?
It made it harder for these things to slide past each other.
It started behaving a little bit more like a ceramic where there's more interactions
that it didn't like, okay?
So that's exactly what happens.
You made this thing harder and way less likely to be scratched by stuffing in extra atoms
near the surface.
Eric?
So in addition to...
Yeah, this does...
Yeah, absolutely.
Another thing it does, not only blocking slip systems, but by adding extra atoms, you're
adding compressive strain, right?
I'm stuffing too many things in here, it has to do something that, so it creates compressive
strain.
So yeah, both of those actually are effective, right?
So we've known about this for a long time, but a long time, they didn't know the science
behind it.
Now we can describe how this happens and we can control it, right?
So if you want your...
Like take these, what are these, six different materials, right?
And if you want just a very thin surface carbonization, you can achieve that.
See the two different colors here?
So you can have thin all the way up to really thick as you control two things.
Two things dominate diffusion, time and temperature, right?
Those are two independent variables you have, right?
Another one that we could say technically we have is species, right?
Some species will diffuse faster than other, but if there's a certain species you want
to use, carbon and steel, for example, you've got time and you've got temperature, is basically
what you've got, right?
So the question then that we would like to be able to answer as engineers is, all right,
how long do I have to hold something?
What temperature do I have to take it?
And then how thick am I going to get?
And when I say thick, like, is it all equal, right?
What's my final thing going to look like?
Is it going to look like that, right?
Or is it going to be a slope?
What's it going to look like?
How do we predict that?
And how do we change that with different processing steps?
So before we say more, let's do a demo.
I need somebody who feels brave, who's willing to risk life and limb for science, right?
Come on up, Steve.
You may regret this a little bit.
So we have here Axe Body Spray.
So you're going to smell a little bit like an eighth grader, but for science, you're
doing it for science.
What's going to happen when you smell the eighth grader in the room, I want you to raise
your hand, right?
You ready for this?
Keep smelling.
This is my favorite part of the drill.
Who smells it?
All right.
Rachel.
Yep.
The eighth grader's spreading.
Is this pretty?
That's so strong.
So hands up if you smell it.
What's happening?
So you're smelling it first here, and this is sort of like ... I walked right through
the wave.
Right?
It's spreading.
Now, the way that it's spreading is not technically diffusion.
It's something a lot like diffusion, right?
This is actually convection, but this is a great analogy to how diffusion works.
Diffusion, if you started with a bunch of carbon ... Y'all, let's actually open that.
That's super strong.
Okay.
Diffusion, you added a really high concentration somewhere, say carbon atoms, and those carbon
atoms have to hop right through the lattice, right?
And the more there are, the more it's going to spread that way, right?
They're going to go down the concentration gradient, right?
So what we really need to come up with is an expression that describes how atoms hop,
right?
How do atoms hop?
What are the key parameters that matter as atoms hop?
So let's talk about it.
Skip that.
We can skip that.
We'll come back to that next time.
Again, one more time, if you took copper and nickel, like my 2010 students are doing this
week in lab, they take actual copper and actual nickel, right?
They put them together, and they're heating it up, and after a while, we know that initially
the composition gradient looks like this, like copper goes 100% down to zero, right?
But over a while, it's going to flatten out and look something like that, right?
Okay.
So this is actually a good graphic of what goes on.
Think of atoms that they're always moving, right?
So there they start out with your initial state.
They're always moving, and the amount that they're moving is proportional to the thermal
energy that you give them.
So they're going to sort of pinball around where their lowest energy sites are, and occasionally,
they get enough energy to hop, right?
But they're going to generally spread out, right?
And why?
Because think, they're already here.
So this guy, he can't hop that way because it was already occupied, but he could hop
that way because it was unoccupied, right?
So it has to do with atoms hopping and available sites, right?
In the beginning, he could not hop up or to the left because those were both occupied,
but he could hop to the right or down, and presumably in and out of the page as well,
right?
So we can introduce some concepts.
Grab my notes so I do this right.
Okay, as this atom hops from this site to, say, that site, right, we can draw an energy
landscape here, okay?
Now the initial energy state is low, and the final one's low, but when it moves between
those two states, it has to break bonds, it has to crowd nearby atoms, so it's actually
going to produce a higher energy state temporarily, right?
So this right here, the distance between these two states, that is going to be our change
in free energy during the jump, right?
That's our activation energy.
For diffusion to happen, it has to overcome the activation energy.
So if you have way less thermal, it's so strong smelling still, right?
You have to have at least that much thermal energy to get over that, okay?
And then the hopping distance, the jump distance, would obviously just be from here to here,
okay?
So if you have no driving force for jumping, right, we haven't added any driving force,
then our plot might look something like this.
We've got site number one, and we've got site number two, and they are both initially at
the same energy, right?
You have to go over some activation energy.
And so what you're going to have is you're going to have a flux of atoms leaving site
one to go to site two as a function of time, dn1 dt.
But you're also going to have a flux of atoms leaving site two, dn2 dt, right?
And if there's no driving force, if we haven't done anything to drive this thing and make
it happen, then the flux of those two atoms going one way or the other is equal, right?
Just as many atoms hop from site one to site two as hop from site two back to site one,
right?
Okay?
And again, this is still our activation energy, okay?
So what does the x-axis represent?
Distance.
Yeah, so this is energy, and this is distance, like somewhere in the crystal.
So if we said Rachel's seat to Ryan's seat, right, that's literally like a spatial separation
of those.
And when somebody hops from one site to another, there's an activation energy, right?
Now, we said that it's got to be unoccupied, so for Jimmy to diffuse, he can't diffuse
this way by sitting on Ryan's seat, because that seat's occupied.
So if he wants to move that way, he has to step on Ryan's seat over here, right?
That's uncomfortable, right?
So it's possible, but it has an energy barrier, right?
But if he goes this way, there's an SMPC.
So diffusion that way would be desirable.
We'll talk about where that comes in here in just a minute.
Yeah, Eric?
So if you're thinking of site one and two as specific points, how can they both be moving
at the same time?
Because this one wouldn't move that way, because there's already one there, and so...
I see what you're saying.
We're going to imagine two scenarios.
In one scenario, there's equal chemical potential both ways, right?
In one case, this is empty and that's occupied.
Your flux would be whatever it would be, related to your activation energy.
And now imagine another scenario where this one's empty and this one's occupied.
It'd be the same flux.
Basically, all things equal.
An atom doesn't care if it goes left or right.
That's what we're saying.
Unless we bias it in some way and change the energy, which we're going to do in just a
second, right?
Everyone follow this?
Any questions?
We good?
Okay.
Let's keep going.
So we can say that the flux of atoms leaving site one or leaving site two, they're going
to be equal, and they're going to equal VF times the exponential of an activation energy
divided by a thermal energy, right?
So what is V and F?
V is our attempt frequency.
Literally, an atom vibrates, right?
So the frequency of vibration.
Every time that it goes whatever which way it's going, it's potentially making a jump.
So let's say Jimmy's shifting at maybe one shift per second, or whatever it is, and that's
his frequency.
It doesn't mean that it's a big shift, but he's shifting, right?
And then F is the fraction of atoms in the correct configuration.
For Jimmy, for him to move from seat to seat, the fraction of atoms looks pretty good.
He's got nothing up front, so we'll ignore that.
He could go backwards with his emphasy, he could go to my left with emphasy, but he cannot
go right.
So it'd be two out of three would be his fraction, right?
So other people like, say, what is it?
Matthew, I'm sorry.
Matthew can't shift at all, right?
His fraction is zero, unless he goes into an interstitial site, but we're just talking
about on-sites work right now, okay?
This makes sense?
So it has to do with not only how often you're shifting, but what's the fraction of sites
around you, and then, obviously, how much thermal energy you have.
So because the shift, the energy required to actually make a move over, he has to shift
from here, clear that much, right?
So that's going to be proportional to how much thermal energy he has.
If you only have this much thermal energy, you don't move, you don't get diffusion, right?
So now let's add a chemical potential gradient to add a driving force for diffusion.
Do you have a question, Matt?
Yes.
And like I said, you won't be tested on these derivations we're doing right here.
When you go and take your thermal class, you're going to learn a lot more about it in detail.
I'm just laying the groundwork for the equation is all, so you don't need to follow these
details super close.
Ben?
All right.
Maybe let's do it.
Let's see what the K sub B is.
Oh, yeah.
Great question.
K sub B is Boltzmann constant, and T is temperature.
So the Boltzmann constant times temperature will give you units of energy.
So when we write RT or KBT, we're in both cases talking about thermal energy, but one
of them, KBT, is thermal energy per atom, and the other one, RT, is thermal energy per
mole of substance.
Since in diffusion, we're talking about individual atoms moving, it makes sense to use the thermal
energy for an individual atom, KBT.
Great question, though.
Okay.
So now if we add a chemical potential gradient driving force, we can bias our flux, right?
So we're going to change our picture here.
Now let's say our starting energy is up here, but our ending energy is down here, right?
We're still going to have an activation energy.
Let's say it's still roughly the same.
Gosh.
All right.
So you've still got something like that.
But now imagine that your flux, let me pick the same colors as before, so gray, oh gosh,
right?
Our flux this way, this is DN1, DT, and then our flux the other way, DN2, that's brown,
right?
So DN1 is going to be far greater than DN2.
And think of it this way.
It still has a relatively small activation energy, at least comparable to what it was
before, right?
But now if you're going this way, your activation energy like doubled, right?
And that's a double inside of an exponential function, right?
We said it's right here.
This is where your activation energy is, right?
So if you double the activation energy, you dramatically reduce the flux going from two
back to one, right?
Any questions on this?
Any questions?
Nothing, Ben?
So like with the first one, you said- Yeah, let me draw it out.
So let's draw it.
We'll say that this distance right here in terms of energy, that's still going to be
our delta GM, right?
But meanwhile, on the other side, we have, let's draw this one.
This difference in energy, we're going to write that as the change in the chemical potential.
Chemical potential, I'll remind you what that was in just a moment.
That will be equal to this, our gradient in chemical potential times the separation between
the two sides, right?
What was chemical potential again?
Chemical potential by definition, if we went back to chapter 10, we'd see it is by definition
the change in energy of a system as you add or remove a component of a certain type, right?
So you guys are all different types of atoms.
If I remove one atom of Jimmy from this body of students here, what happens to the energy?
Does it go up or down, right?
If I add another Jimmy here, right, what happens to the energy of that body, right?
That's the chemical potential, right?
Now imagine the scenario that we started with.
We sprayed good Steve up here with the ax.
So the concentration of atoms here, super, super high.
So that has a different chemical potential than students in the back of the room that
don't have any of the ax body spray, right?
So there is in fact a gradient in chemical potential going from here to the back of the
room, right?
More or less linear, kind of radial, I guess it would go, because it came from a point,
I guess it would be radial, right?
But there's a gradient in our chemical potential, right?
Because of that, you get like this, so it may be linear, it may not be linear, depending
on how your concentration looks, right?
But you get a difference in chemical potential, that's what matters.
Therefore you get higher on one side than the other, okay?
Any questions?
Yeah, T?
Can you explain dn1, dc1?
Yeah, so great question.
N is the number of atoms at site 1, so n1 is number of atoms at site 1.
T is time, right?
So dn1, that's the derivative of the number of atoms at site 1 with respect to time.
So it's the change.
It's the change in atoms as a function of time, like in one second, how many atoms leave
site 1?
1, 10, 100, right?
Depends on the rate at which they're leaving, right?
Obviously the rate at which they're coming into it again, too, right?
Kylie?
So let's say dn1...
That's one, yeah, that's one way to think of it, actually, right?
Obviously if you keep on moving atoms out of it, they have to keep on coming from somewhere,
right?
But yeah, okay?
So you use, like, dn1 or dn2?
Okay, so all we've done is by adding a chemical potential, right?
We've added a chemical potential, we've added a chemical potential, we've added a chemical
potential gradient, we make one energy state lower than the other.
And maybe the way to think of it is, like, imagine I had, like, a bunch of extra people,
and I'd be like, hey, we've got to put them in this row somewhere, and I'd say, all of
you extra 20 people sit in these front couple rows, right?
They're going to, like, cram in here, right?
Which makes this front couple rows uncomfortable, energetically unfavorable, right?
So those extra rows of people are going to, like, slowly, awkwardly, like, try and, like,
shift further back, right?
Because it's lower energy.
There's less stress.
There's empty seats further back, right?
That's exactly what's happening.
You have higher energy here because I crammed extra atoms here.
The concentration was higher.
Concentration, right?
And there's lower concentration in the back of the room, so they're going to work their
way backwards because it's lowering their energy.
Atoms are doing the exact same thing.
If we go back to our...
that Wikipedia image, right?
They start out...
We had a high concentration, right?
But there's empty seats that way, so they're going to slowly diffuse down the concentration
gradient because that's moving down the chemical potential, and it has to do with flux, right?
Some of these atoms will hop back.
I hope in the graph that they include at least one hop back.
I don't know.

