Video Lecture on Engineering Fracture Mechanics, Prof. K.

Ramesh, IIT Madras 1

Module No. # 02, Lecture No. # 09: Fracture Strength by Griffith

In the last class, we had developed the equations for strain energy stored in slender
members subjected to axial load, torsion, then bending. The expressions you had already
learnt in a course in strength of materials. Nevertheless, we looked at what is the nature of
those expressions, and in fact towards the end of the chapter on energy release rate, we
would use them to find out the energy release rate.

(Refer Slide Time: 01:21)

We would take up simple problems,

where the crack is long enough, so
the portions of the object can be
considered as combination of
slender members. The crack would
demarcate the components as
slender members. And we would
use the expression for energy stored
because of the external loading and
evaluate the energy release rate
from fracture mechanics point of
view.

And we have also looked at and

asked a question what are the changes in the component when crack advances. One
obvious change which all of us could recognize easily is, stiffness of the component
decreases. Then we moved on to look at strain energy in the component decreases or
increases. This also we had a look, at the problem of double cantilever specimen. And we
were able to see for two extremum conditions of constant load and constant displacement.
We would look at them again. And the other issue what we will have to keep in mind is,
when you are having a general
situation, the points of the
component at which external loads
are applied may or may not move.

So, in one case we had looked at

constant load, where the points
were moving. We have also looked
at fixed grip, the points will not
move. And it is obvious, work is
being done on the component if
the points move. And another
important aspect what you will
have to keep in mind is, energy is
being consumed to create two new
surfaces.

(Refer Slide Time: 03:32)

Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 2

(Refer Slide Time: 04:01)

(Refer Slide Time: 04:05)

This is the key statement in the

Griffiths approach to fracture
mechanics. He recognized the
formation of two new surfaces
requires energy; there is inherent
resistance persist in the material
and formation of crack requires
two new surfaces to be formed.
They consume energy and this
energy has to come from some
source. And for the purpose of
analysis, we have looked at two
extremum conditions: one is the
constant load. We will look at the
animation, we had seen that in
detail in the last class, just observe
it. As the crack is long the stiffness
is reduced, and you find when the
load has reached the value of P1
crack has advanced. And we
calculate what is the change in
strain energy. And you find the
change in strain energy is nothing
but this triangle. And here, the
strain energy increases when the
crack advances. And you should
also note that, the load has moved a finite distance, dv. What is the external work done?
That is the rectangular area what you find here. And what is the energy available for crack
growth that is the triangular area. And it is simple arithmetic that would tell you that, the work
done is twice the strain energy available for the formation of two new crack surfaces. And in
this case, the strain energy
increases, because of external work
done.

(Refer Slide Time: 04:52)

And we have also looked at the

other case of constant
displacement. And here again you
find, when the crack is longer the
stiffness of the component
decreases. So you have the line
corresponding to a plus delta a is
below the case for the crack length
a. Watch the animation. As the
crack advances under constant
displacement, and you find the work done by the external load is 0 because the points do
not move. And what is the strain energy change? This is the final strain energy, this is the
Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 3

initial strain energy, so the change is what you find here. And in this case the strain energy
decreases. (Refer Slide Time: 05:40)

(Refer Slide Time: 05:45)

So, you have to keep this in mind. I

had also mentioned the use of
constant load or constant
displacement is convenient from
the point of developing the
mathematics and also convenient
to interpret the results from an
experiment. I can also calculate
the strain energy release rate,
once we learn it in this chapter
from a graphical approach. And we
have also made a statement in the
last class, that in a generic loading
you have a combination of
constant load and constant displacement. So, it becomes convenient for you to have the
results as a ready reckoner. What happens in constant load, and what happens in constant
displacement.

And what is the challenging question, that we have now is, suppose I have a component
with a crack how to find out the strain energy in the presence of a crack? And what did
Griffith do? Griffith took the problem of a central crack following the example of a plate with
the circular hole, a plate with the elliptical hole; he considered the problem of an infinite plate
with the central crack subjected to uniaxial tension.

(Refer Slide Time: 07:17)

Now, the question is how to find

out the strain energy in the
presence of a crack? One simple
approach could be use
dimensional analysis. The second
possibility is go for a relaxation
analogy. And the third possibility is
perform actual calculations based
on crack face displacements. That
would be the most accurate way to
find out the strain energy in the
presence of a crack. But right now,
we do not have the necessary
quantities to do that kind of a
calculation. If you have to do that, we need to know the knowledge of stress and
displacement fields in the vicinity of the crack-tip.

Once we take up crack-tip stress and displacement fields, we would develop displacement
as well as stress field then come back and do these calculations again, and satisfy ourself
that, we could find out the strain energy in the presence of a crack, from a mathematical
stand point accurately. For the current discussion, we will look at for a central crack, and
keep this information important. Crack with two tips this is also very subtle point, but equally
Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 4

important. The strain energy is given as U suffix a equal to pi sigma squared a squared
divided by E.

This is for a central crack. Once you say central crack it is crack with two tips. And whatever
the result that you have seen here is for a infinite panel of unit thickness. So it is multiplied
by unity, and if you look at from dimensional point of view, this satisfies the expression for
energy. And the suffix a denotes strain energy in the presence of a crack.

Now, what we will do is, we will look at a relaxation analogy to satisfy ourself how this
expression could be viewed at. But ultimately, you will be able to convince yourself well, only
when you do the actual calculation, this we will postpone it for the moment. Nevertheless we
will have a look at it.

(Refer Slide Time: 10:16)

Now, let me look at the relaxation

analogy. And what I do here, I take
a plate of thickness B. Because I
want to introduce in fracture
mechanics literature rightly or
wrongly, they have used the
symbol capital B to denote the
thickness. For the purpose of your
understanding you could consider
that as a rubber plate. You know
you could take a cycle tube and
then cut it and then make a sheet
out of it. And what do you do is,
you take it and stretch it, and then
clamp it rigidly.

Then, what I do is introduce a crack at the center. Remind you whatever we discuss it is
meant for a brittle material, but for visualization purpose, it is convenient for you to take a
rubber sheet and then look at what will happen. Suppose, I introduce the crack what do you
anticipate? Up to some extent, the crack will remain stable. Beyond a length, the crack will
propagate and the strip will separate itself. This is you can visualize physically.

For the formation of crack, you need to have surfaces to be formed and energy is required to
form the surfaces; this energy would have come from the strain energy of the system. And
what is the kind of situation we are analyzing? We are looking at fixed grips, so there is no
external work done.

We will just look at what way the crack is put and for the formation of these two surfaces,
some energy would have come out of the strain energy of the stretched plate, and we will
have to calculate this quantum of energy. We will make a simple calculation and to make our
life simple, we consider a triangular area in the neighborhood of the crack is what is
releasing the energy to form the two new surfaces. It could be any shape. This is just to
illustrate a simple calculation methodology to find out strain energy in the presence of a
crack.
Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 5

(Refer Slide Time: 13:18)

We would find out the calculations.

We have already looked at the
form of the strain energy in the
earlier examples, where sigma
squared by 2 into Young's
modulus. This is multiplied by the
volume, and here volume of the
triangles. I have two triangles, so I
have two here, half BH, we have
assumed the height as lambda
times a, and lambda is used just
as a proportionality constant here,
it is not denoting a wavelength,
because we are accustomed to
seeing lambda as wavelength. This is just used here as a proportionality constant and when
you multiply this, you get an expression. 2 lambda a squared B sigma squared divided by E.
See we are only doing a relaxation analogy, the idea is to feel that strain energy would get
released from the system given to the cracks to form. To illustrate that we have looked at
relaxation.

We can be satisfied only when you do the mathematical calculation completely on the
displacements of the crack faces, as well as the stress field associated with it, that would
require some more time for us to do that. Right now, we only look at an analogy and we also
bring in one more aspect; we have taken a thin plate. We are really considering a plane
stress situation and for plane stress situation, lambda is taken to be pi by 2.

This also will get clarified, when we actually do the calculations based on the displacements
and the stress field. So, when I do this, I get this as Ua equal to pi a squared B sigma
squared divided by E. The earlier expression, we saw was for a plate of unit thickness, now
you have a thickness of B and what do you see here? The strain energy in the presence of a
crack is related to a squared. So, when the crack length changes, the energy would change
as a parabola, second degree curve.

And now we will look at for the formation of two new surfaces what way we require the
energy? If we look at these two natures we can understand, there could be a crack growth
phase and a catastrophic crack
propagation. We would try to plot
them.

(Refer Slide Time: 16:34)

So, this is what I have here. What

is the variation of surface and
strain energy in fixed grips. You
should not forget this, we are really
looking at fixed grips; fixed grips
means constant displacement. And
we have gamma s is the surface
energy per unit area of one
surface. Then the surface energy
required for a model of thickness B
Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 6

is 2 into 2a into B into gamma s. And this we denote as U suffix s. And if you look at this final
expression, it turns out to be 4a into B gamma s. It is a linear function of the crack length. As
the crack length changes this would vary linearly. And we would try to plot this variation in a
graph, where the X axis is taken as the length of the crack, and Y axis is taken as the
energy. This would be obviously a straight line. We will plot that.

So, I have this as a function of crack length you find that this as a straight line, this denotes
the surface energy. We have already got the result of what is the strain energy in the
presence of a crack, which is a second degree curve. You have already got that as U a
equal to pi a squared B sigma squared divided by E. This is plotted in this fashion.

Please note that, this is only a schematic its not drawn with actual values. It is only a
schematic to illustrate the nature of the surface energy, the nature of the strain energy.
Suppose, I find out what is the total, that would change in this fashion. That would change
like this. The slope would become 0 at a particular point, that demarcates the region from
stable crack growth to fracture. You get the critical crack length from that graph. This will
denote the critical crack length, so you have a region where it is incremental crack growth by
crack growth mechanism, and in this zone it is catastrophic crack growth by fracture
mechanism.

This is given pictorially, you know we would like to have this quantitatively. So we will look at
it from a mathematical stand point, but I would like you to make a neat sketch of this
diagram, this is very illustrative of what is really happening. So you are able to find out a
critical crack length beyond which the crack would propagate faster. And what we are really
looking at is the incremental change in crack length to occur. So we are really looking at that
the incremental energy requirement should be satisfied. Mathematically, how will you write
it? The change of strain energy in the presence of a crack should be equal to change of
surface energy with respect to the crack.

(Refer Slide Time: 20:17)

So, dou U a divided by dou a equal

to dou of surface energy divided by
dou a. This is the kernel of
Griffiths approach to fracture
mechanics. He said, the formation
of two new surfaces requires
energy and this energy since we
are looking at fixed grips comes
from the strain energy.

When this is satisfied, the crack

can grow. And now you have to be
very careful. See you may be
tempted to differentiate with
respect to dou a, what it really denotes is length of the crack, fine. In the numbers that we
have got, the expressions that we have got whether you would differentiate with respect to a
or differentiate with respect to 2a, because you are going to do it on the left and right hand
side, the end result would be same.

If, you differentiate it with respect to a, you will get this as pi sigma squared 2a divided by E
equal to 4 gamma s. If you differentiate with respect to 2a, you will get this as pi sigma
squared a divided by E equal to 2 gamma s. In fact, if you simplify this, you will come to this
Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 7

step. But dont differentiate with respect to a because I have already emphasized he
considered the problem of a central crack. The moment you look at a central crack you
should recognize, that it has two crack-tips. If you have to interpret these equations properly,
you have to differentiate with respect to 2a. So the caution is dont do it like this, do it like
this. Even in the definition of energy release rate, we will bring in the energy per crack-tip.
That is very important.

So, we get the final expression as pi sigma squared a divided by E equal to 2 gamma s. This
could be rewritten as sigma root a equal to 2E gamma s divided by pi. A very very important
expression. And symbolic step by Griffith in advancing fracture mechanics, because he has
been able to find out the fracture strength in the presence of a crack. You could also look at
this expression differently as sigma equal to root of 2E gamma s by pi a.

I have taken out a in this left hand side, so it appears as sigma root a. And you know once
you develop an expression like this it has to be experimentally justified. Otherwise you know
you will not be satisfied with the mathematical development. And what is the kind of
experiment with Griffith performed to establish this? Griffith carried out a series of
experiments on glass tubes and spherical vessels subjected to internal pressure, p.

(Refer Slide Time: 24:08)

And you should keep in mind

almost all the work of Griffith was
confined to brittle solids. So the
energy release rate concept was
originally developed for ideally
brittle materials. Later on this was
extended by Irwin and Orowan for
high strength ductile solids. Until
we do that all our discussion
confined to brittle materials. What
is the advantage of a brittle
material? We have already seen,
that crack could heal in a brittle
material. Such things have been
observed, we have also seen one such example. And the advantage is when you are
approaching from the energy point
of view, you are looking at a
conservative system. So energy
method is easily applicable. When
you say the glass tubes and
spheres, they were pre-cracked and
then annealed to eliminate residual
stresses, if any.

(Refer Slide Time: 26:20)

So, he had taken sufficient

precautions in performing the
experiment. And he increased the
stresses, and he estimated what is
the value of sigma root a, which was
found to be in the range of 0.25 to 0.28 MPa root meter. Have a look at this value he has got
Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 8

some value in the range of 0.25 to 0.28. And this he has got the left hand side expression.
Sigma root a has been able to get it. For you to get the right hand side expression you
need to get the value of surface energy; surface energy in glass has to be estimated. You
know you will have to congratulate Griffith for having taken the pains to perform such difficult
experimentation.

He considered that surface tension of glass, is a linear function of temperature. Good

approximation. You know, this is how engineers always approximate. We are very close to
straight lines. And we always want to take anything as linear, if it satisfies our requirement
we leave it at that. We want to avoid as much of nonlinearity as possible. And what he did,
he extrapolated surface tension values of glass fibers between 1110 degree centigrade and
745 degree centigrade to room temperature. From that exercise he could estimate the
surface energy of glass as 0.54 Newton per meter. And E for glass is about 62 GPa, and
when you want to evaluate this expression, root of 2 gamma glass into E divided by pi gives
you 0.15 MPa root meter.

Is it same as what he had got in his experiments? It is not same. When students perform
experiments these days and compare it with theory for fear of experiment being rejected,
they report only a difference in the second decimal place. I have been mentioning, when you
are performing an experiment you have to learn to report the results that you have got
faithfully. Experimentalists have to be extremely honest, because the values may carry more
information than what you had
originally thought of.

We have seen, in the case of

fatigue experiment lot of scatter.
That is a well performed
experiment. So, do not think that
when you compare theory and
experiment they will be very, very
close. Unless, you are taking a
benchmark problem, and theory is
very well understood, and also the
experiment is very carefully
performed you will not have that
kind of a correlation. Nevertheless,
if you look at the order of
magnitudes in this case, a correlation definitely exists.

(Refer Slide Time: 29:34)

This gives you a confidence the way you have developed the expressions. Sigma root a
equal to 2 gamma s E divided by pi is a meaningful step. But we will also look at in what
context Griffith had really approached this problem. He had not in his mind that there is
fracture mechanics that he has to develop. He had not approached the problem from that
perspective. His perspective was different. You know scientists were really looking at how to
estimate strength of the material from a theoretical stand point.
Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 9

(Refer Slide Time: 30:29)

So, you have an atomistic model.

You say it is a crystalline solid.
Make a sketch of it. And you have
this equilibrium position is given by
the lattice spacing b. And you want
to find out, at what value of this
stress would there be separation.
So, the idea is, estimate the
strength of a crystalline solid
based on its lattice properties. And
for this, you need to go for a force-
separation law. And how this is
coined; the inter-atomic force
separation law can be
approximated by a function exhibiting the following three properties.

You know, I am going to present this without much of mathematics we will only look at the
final results. In fact, people who work on cohesive zone modeling really play with different
types of law, and then also incorporate it in the finite element modeling and analyze, how the
plastic zone develops near the vicinity of the crack. They do such kind of activity but we will
only look at salient features of it.

Our interest is to see, what kind of result people have got based on lattice property
calculation. So, if I want to have the force-separation law I need to consider the law such
that, the initial slope corresponds to the elastic modulus E of the material, whatever the
material that you are looking forward to. And the second aspect is a total work of separation,
that is area under the curve corresponds to the surface energy gamma s.

So, what you will have to look at is, people who are making theoretical calculation have
already brought in, the role of surface energy in some other form. So, this kind of thinking
was there at that time. So you need to have the force-separation law to have an initial slope
that corresponds to the elastic modulus E.

Total work of separation that is the area under the curve corresponds to the surface energy
gamma s. And finally, you should have the amplitude a maximum value that represents the
inter-atomic cohesive force. So,
you can obtain a force-separation
law if it exhibits the following three
properties.

(Refer Slide Time: 33:00)

So, based on this the law appears

in this fashion. Sigma function of x
is given as E into gamma s divided
by b whole power half multiplied by
sine of E into b divided by gamma
s whole power half into x by b.
Where b represents the equilibrium
inter-atomic spacing thats what
you had seen in the sketch and x
Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 10

denotes the displacement from the equilibrium separation distance.

(Refer Slide Time: 33:58)

So, now we have an approximate

relation which gives you the force
necessary for separation. And
what is the maximum value? It is
given by this; so the maximum
value is nothing but the theoretical
strength. The theoretical strength
is given as E into gamma s divided
by b whole power half. See we
would be able to appreciate this
only when we look at the actual
values. From a lattice point of view
you can calculate the theoretical
strength. From experiments you
find the actual strength displayed by the material. If we had understood the material
behavior correctly my theoretical calculation should more or less coincide with experimental
observation.

We will have to see, whether this is so or not. People have also found out what is the value
of gamma s. For many materials it is found to be approximately Eb divided by 40. So when
you substitute this into this expression, you get a value for theoretical strength. And what is
the theoretical strength? Should be like E by 6. In fact it is very high. E by 6 is not a small
value. We will look at for mild steel, because mild steel everybody knows. Although we may
not apply fracture mechanics to mild steel, to dispute the theoretical strength calculations
are way off when you compare with experimental results, mild steel is a good example.

In the case of mild steel, you have Young's modulus as 200 GPa, so 200 by 6 would be
something like 33 GPa. Mild steel is no where near this. If you look at, its yield strength is
something like 220 MPa or so, its ultimate tensile strength would be something like 320 or
400 depends on the material composition. that is all in terms of MegaPascal whereas,
theoretical strength is in terms of GigaPascal, that means the theoretical strength
calculations are way off. Why this is so?

(Refer Slide Time: 36:42)

You know this is a puzzle, people

wanted to resolve this puzzle for a
very very long time. And we will
look at what are the kind of issues.
People based on experiments
coined this as a size effect; they
had no clue why the theoretical
strength is far higher than what is
observed in experiment. But in
experiment they saw something
very interesting, that can be traced
way back to Leonardo da Vinci. He
found an inverse relationship
between the strength and the length for wires of constant diameter.
Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 11

See, there are two issues that we will have to look at. In the case of conventional spring
design if you look at the design books, you will find the yield strength for the same material
when it is used as a wire is much higher than a bulk material, this is one aspect. There is
another aspect, which was noted by da Vinci. What he said was the length, if the length is
shortened, the strength increases. It is an experimental observation.

See we need an explanation for this. Remember experiment is truth. If the experiment is
carefully performed and the results are also honestly reported then, experiment is truth. You
have to find out reasons for such a behavior. And later you find there another person Le
Blanc in 1839, he also established long iron wires to be weaker than short wires of the same
diameter.

So, people are finding there is something to do with the size. And they generalized it as a
size effect, and they reported an increase in strength as the dimensions of the test piece
decreases. A too drastic conclusion. You know, this kind of a conclusion we cannot make it
now. In those days when they were struggling to find out, why the theoretical strength was
not achieved in actual practice. But they observed when the size of the specimen decreases
the strength increases. They have noted this. They needed an explanation for this. And this
was the aspect which Griffith was concerned about and he felt that, the size effect is actually
a crack size length effect.

(Refer Slide Time: 39:30)

So, what he did was he performed

result on glass fibers. Carefully
reduced their diameter and then he
plotted thickness of the fiber
versus the fracture strength. And
this is purely based on experiment.
And this was exponentially
decaying and it was asymptotically
reaching the strength of bulk glass
about 175 MPa.

When the thickness of the fiber is

reduced, it was approaching 11000
MPa that is a theoretical strength
of glass. And what is the difference? The difference between the two is, you have cracks of
various dimensions exist, and if you draw the glass thinner and thinner the cracks diminish
and glass becomes very strong. In fact, it is useful. See if you look at the development of
composite it definitely hinges on this useful property. Glass by itself is brittle, but glass fibers
embedded in a resin works as a fiber glass composite. Very good, you have fiber glass
boards, you have fiber glass mridangam and so on. So, you have utility of glass in fiber form
better, when embedded in a resin.

So, what Griffith looked at was, in the case of glass fiber the presence of cracks have
diminished strength. So he was able to provide a rational explanation why, when the size of
the specimen diminishes, the strength increases. I would like you to make a neat sketch of
this graph. It is very illustrative very important observation by Griffith. And you should also
look at in those days when they were developing the theories if they had a doubt they had
always verified it by suitable experimentation. However, complex it may be.
Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 12

(Refer Slide Time: 42:31)

They have taken the trouble to do

the experiment, because they were
struggling to find answers for the
questions that were burning at that
time. And we have already
developed this expression without
getting into much of the
mathematics.

The fractures strength by Griffith is

as follows. It is 2 by pi E into
gamma s divided by a whole
power half. And this is for a plane
stress situation. And what is the
other important aspect of this
expression? See this expression can be looked at from various point of view.

One is he has related the fracture because of crack length. Other way of looking at it is, in
the case of Inglis solution even a very very small crack with very very small external load,
the stresses would be so high the specimen will break into pieces. But that is not so in actual
practice. So, thats a paradox.

The Inglis solution was useful, but utility of that solution is questioned, because people are
unable to explain how actual structures remain solid with cracks. So that paradox was
resolved by Griffith, that strength of the plate is independent of the size of the crack. That
was the immediate observation from Inglis solution. Whereas Griffith said the crack length
does matter.

I can find out using this expression fracture strength, as well as critical crack length for a
given stress value. So this expression is very important. You know, we have looked at from
very simplistic analysis that, you get energy from strain energy for the formation of two new
surfaces, and we were able to get the expression for fracture strength.

(Refer Slide Time: 44:50)

So, what Griffith concluded was,

the apparent size effect was
actually a crack-size effect. And if
you look at the expression for
theoretical strength as well as
fracture strength I can find out the
ratio of sigma f divided by sigma
th, that is approximately equal to
b by a whole power half.

(Refer Slide Time: 39:30)

Video Lecture on Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 13

So, you can go and find out for a bulk glass what would be the size of the crack; the size of
the crack is of the order of 0.025 millimeter. Because a bulk glass has this crack length. Its
fracture strength is far below the theoretical strength. See, if you look at the graph we have
noted that theoretical strength is around 11000 MPa, whereas bulk glass strength is around
175 MPa. Because of the expression developed by Griffith, now, we are able to say that, the
crack length for the bulk glass would be around 0.025 millimeter. Now, you take a fiber of
glass thinner than this. What happens when it is much below the size of that crack? You
find the value approaches theoretical strength. Now, the story is complete. People were
struggling why, when the size of the specimen decreases, strength increases.

(Refer Slide Time: 47:09)

This was the problem with Griffith

was focusing while developing his
energy balance approach. He got
the fracture strength. And he could
relate that length of the crack also
plays a very important role. And
people have worked on this kind of
an approach and they have also
found out the fracture strength for
several other situations.

We have seen, fracture strength as

2 by pi into E into gamma s divided
by a whole power half, for a plane
stress situation. Suppose, I look for a plane strain situation how does the expression look
like? There is only a small variation in the expression. What you find here is instead of
Young's modulus you get this as Young's modulus divided by 1 minus nu square. In fact,
you would come across this kind of modification in fracture mechanics. You will develop the
expression for plane stress then, you can convert it to plane strain by changing Young's
modulus by Young's modulus divided by 1 minus nu square, vice versa. We can do that and
you get this for plane strain.

And people have also looked at another problem. You know when you have a three-
dimensional solid, where you have a penny shaped crack. The moment you have a penny
shaped crack, people found the expression only the numerical factor has changed. the form
remained same; you have Young's modulus, you have surface energy divided by crack that
remains same. only the numerical factor was changing, and this is a very important
observation utilized by Irwin to generalize crack problems.

So from the energy approach, Griffith identified formation of two new surfaces requires
energy to be given from the strain energy of the system helped him to get the expression for
fracture strength, which also explained the paradox of Inglis solution, also explained why
actual strength of the material is far below the theoretical strength, and later on when people
carried out for variety of practical situations, they found a similarity. So, it helped to develop
fracture mechanics in the years to come. that is a very important contribution by Griffith mind
you all his analysis were focused to ideally brittle solids. You should never forget that, thank
you.