Engineering Fracture Mechanics

Prof. K. Ramesh
Department of Applied Mechanics
Indian Institute of Technology, Madras

Lecture No # 30
Dugdale Model

(Refer Slide Time: 00:21)

See, in the last class we have looked at what is the motivation of Dugdale model; we will
have a brief look at that and proceed further.

An elastic analysis predicts singular stresses at the crack-tip, which is unrealistic. In an

effort to eliminate such an unrealistic prediction, Dugdale in 1960 and Barenblatt 1962
independently introduced yielded or cohesive strip zones extending from the crack-tip.

So, you will have to keep in mind these are all models. People were attempting to solve a
very complex problem, so they have proposed from their understanding, which way to go
about. The opening of prospective fracture surfaces ahead of the crack-tip is assumed to
be opposed by a cohesive stress, and what did Dugdale do? Dugdale took that stress to be
the yield stress of the material, so it is an assumption. Essentially, he has taken this kind
of an assumption and proceeded with the model.
(Refer Slide Time: 01:35)

The next step is to find out the cohesive zone length and how this is determined? It is
determined by the condition, that the stresses be nonsingular. When you look at the
mathematical development, this statement would become quite clear and you will also
have to note, that there are obvious similarities between the approach of Dugdale and
Barenblatt, which has led researchers to refer it as Barenblatt-Dugdale crack theory.

But you will also have to note, that these have different physical basis. In fact, all of this
we had discussed towards the end of last class, this is only for recapitulation.

The physical basis of Dugdale’s model is macroscopic plasticity; on the other hand,
Barenblatt theory is based on molecular cohesion.
(Refer Slide Time: 03:02)

Now, we will take up the mathematical development of Dugdale’s approach and you
need to make neat sketches of this, and one is an idealization of the crack, other one is an
actual crack with the plastic zone developed.

So, what you find here is, you have a crack of length 2a; there is an extent of plastic
zone, which is depicted as delta. And you would also see the experimental result and
reconvince our self, that this kind of picturization is what is seen in the actual
experimentation. A figure like this is taken up for mathematical analysis and you will
have to keep in mind, that the Dugdale’s approach is suitable for thin plates. This is the
restriction number 1 and the theory also assumes, that the materials respond to Tresca
yield criterion. So, only for those materials the results will be exact.

And Dugdale also made another assumption. You know, we had earlier looked at what is
the shape of the plastic zone near the vicinity of the crack-tip, we had seen a finite shape
in front of the crack-tip for the Dugdale’s model. He assumed that plastic deformation
concentrates in a line. See, this goes with the other two assumptions. When you have a
thin plate and the material respond to Tresca yield criterion, this is also valid.

In fact, we would see later, when Dugdale reported the results people did not accept it
immediately, people were waiting for the experimental result of Hahn and Rosenfield;
we will see that also. And because you have plastic deformation concentrates in a line,
the model is also termed as strip yield model.

So, what you have is, you have an extension of plastic zone, that distance is given as
delta and we will have to find out how to get the extension of length delta.

(Refer Slide Time: 05:51)

And make a neat sketch of this, you have a center crack in an infinite strip and what you
have here is, the length of the crack is 2a and for this zone, which is plastically
deformed. The stress level is taken as sigma ys, this is an assumption made by Dugdale
in developing the model. And what is the effective crack length now becomes? It
becomes a plus delta.

And from the discussion we have done earlier, what we are really looking at is, we want
to see K sigma plus K delta should go to 0. You have a center crack in an infinite strip,
because of the applied stress sigma you will have a singularity developed. And what
Dugdale assumed is, ahead of the crack-tip, there would be a plastic zone and the
associated stress level, he had taken that as yield. The strength of the material sigma ys,
this will also introduce stress singularity. And we have to find out the length delta in
such a fashion, the summation of these two goes to 0. And what is K sigma? K sigma is
sigma square root of pi into a plus delta.
We are taking a fictitious extent of crack length by a distance delta and based on that
only you have to calculate the stress intensity factor K sigma. And as I mentioned earlier,
K delta is a singularity due to pressure sigma ys.

See, the singularity, because of the applied stress, is given in a very simple fashion and
can you find some similarity here to find out how to evaluate the stress intensity factor in
the presence of the sigma ys? The crack phases are symmetrically loaded, that you can
recognize and in fact, we have done evaluation of stress intensity factor for symmetric
concentrated load acting on the crack surface. Earlier, in fact, you could invoke that and
write an expression.

(Refer Slide Time: 08:53)

The difference is you are having a distributed loading where we had looked at a
concentrated loading. So, I can take a distance s from the center of the crack and look at
an incremental distance, ds. Please make a neat sketch of this graph, the crack loaded
with stresses. I can also look at simultaneously on the negative x-axis at distance s,
another strip of load.

And what we are going to write is we will write dK delta. We would just invoke the
expressions that we have developed for a symmetric load and write this for an
incremental value of stress intensity factor. Then, you integrate for the extent of the load
sigma ys applied; you will be able to calculate the stress intensity factor.
(Refer Slide Time: 10:10)

So, if you look at the form of the stress intensity factor for a symmetric wedge load, you
have this load P per unit length, you have this as 2 by pi P into root of pi a divided by
root of a squared minus s squared. So, I can invoke this and write the incremental value
of stress intensity factor in the current problem.

(Refer Slide Time: 10:38)

So, you have dK delta equal to minus 2 by pi. That expression was for a load trying to
pull the crack surface; here, the load is trying to close the crack surface. So, I have a
minus sign here and the stress level is known, this is given as sigma ys and incremental
distance is ds. So, sigma ys ds denotes the load P and root of pi a and pi a is taken as pi
into a plus delta and you have divided by square root of a plus delta squared minus s
squared. And rest of it is simple mathematics, you know, we have been handling
integration of such quantities and we would get those expressions.

So, what I have here is, it becomes K delta equal to minus 2 sigma ys multiplied by pi
into a plus delta divided by pi between the limits. You integrate a to a plus delta 1 by
square root of a plus delta square minus s squared into ds.

You know, we have solved such expressions earlier; you can look at your notes a few
pages back and then, see, what would be the integrated value? You are essentially using
the table of integrals, so you can easily find out what would be the integrated value, then
apply the limits. In fact, I am going to skip that step; I will apply the limits and write the
final expression. The intermediate step, I would like you to do it, we have already seen it
earlier while discussing evaluation of stress intensity factors, so you can use that part of
the notes. So, I am skipping the intermediate step of putting the limits.

After putting the limits, whatever the expression I get, could be simplified to cos inverse
a by a plus delta and this is pre multiplied by minus 2 sigma ys square root of pi a plus
delta divided by pi.

So, what we have done? We have used our knowledge of stress intensity factor for
symmetric load acting on the crack phase, extended it for finding out the stress intensity
factor when I have sigma ys on either side of the crack-tip, and we have got the
expression for K delta. What we have to do next? We have to write the expression K
sigma plus K delta equal to 0, that is what I am going to do.
(Refer Slide Time: 13:51)

And that is what is written here. So, what I have here is, sigma square root of pi a plus
delta, this is the K sigma expression and K delta expression we have recently calculated,
that is, minus 2 sigma ys square root of pi a plus delta divided by pi multiplied by cos
inverse a by a plus delta equal to 0.

See, there is a subtle difference in the development of the expressions itself. See, if we
have looked at what way we developed Irwin’s model, we had a plastic zone, the plastic
zone length was given as r p, but the value of delta to modify the crack length is taken as
1 half of r p that was taken as delta. But here, in the development itself, we always take
the complete length of the plastic zone as a correction required for the crack length. This
is seen at every step of Dugdale’s development, so keep a note of this.

So, the next simplification is, if I take out square root of pi a plus delta, the expression
turns out to be sigma minus 2 sigma ys divided by pi multiplied by cos inverse a by a
plus delta equal to 0. At every stage you are correcting the crack length with the
expression delta.

So, finally, when we look at the plastic zone length and there correction necessary for
crack length, we have to handle it carefully in Dugdale’s model as well as Irwin’s model,
there is a subtle difference between the two.
And I can take this to right hand side, I can knock off this. So, essentially, I get sigma
minus 2 sigma ys divided by pi multiplied by cos inverse a by a plus delta equal to 0. So,
this will help me to get an expression for delta. Let us see what we get? So, I get a by a
plus delta. On simplification you can take it on the right hand side and you will get this
as pi sigma 2 sigma ys, so you get this a by a plus delta as cos of pi sigma divided by 2
sigma ys. We have to solve this further.

See, whenever we have an expression like this, as engineers we have to bring in certain
approximations. Only by bringing certain approximations you would be able to get a
simplified expression and what is approximation we will do? When you have sigma and
sigma ys appearing here in most of the fracture problems, the value of sigma will be far
below the value of yield strength of the material. See, that was the worry; structures were
failing as stress is far below the yield strength value.

(Refer Slide Time: 17:32)

So, I can invoke this kind of simplification by saying sigma is far below sigma ys and
also make another observation, that the extension of plastic zone is much smaller
compared to the crack length. You know, these are very important, only when you make
approximations like this we would be able to simplify and proceed further.

So, what you have is 1 minus delta by a equal to 1 minus pi square sigma square divided
by 8 sigma squared ys. So, from this I would be able to calculate what is the expression
for delta and delta turns out to be pi sigma squared pi a divided by 8 sigma squared ys.
Since we are handling a center crack in an infinite plate, I can bundle this and write this
as pi by 8 K 1 by sigma ys whole squared. In fact, in all problems dealing with plastic
zone length you will have an expression K 1 by sigma ys whole square. It will be pre
multiplied by a factor here. If the factor is pi by 8, we will also have a summary of these
quantities and see how this can be utilized.

What is the kind of comparisons that we can make from simplistic model, Irwin’s model
and Dugdale’s model? See, you will have to note, when Dugdale developed his model,
he has got the extent of plastic zone length. In fact, in his original paper, he had just
compared from certain experiments, what is the value of this extent of plastic zone
experimentally, as well as predicted by his theory this matched quite well, but what he
had not done is, he had not provided the pictorial support for his observations. We will
also have to note that.

(Refer Slide Time: 20:03)

So, Dugdale verified his prediction of the plastic zone size with experiments on sheets.
So, he was working on thin specimens containing both, internal and edge cracks, and the
comparison was quite good because it is only a model. See, you are charting a difficult
problem and you have to idealize it, so that you get some useful result for you to proceed
further. So, his idea was, there would be a plastic zone, it will be in the form of a strip
and he was also able to verify his prediction based on experiments.
As Dugdale did not provide pictorial evidence in his classic paper, people doubted that
whether yielding could occur this way and hence, the work was not accepted
immediately. See, we will have to keep it in mind, you know, we had done a circus to
find out, approximately, what is a shape of the plastic zone near the crack-tip. I said, only
in the case of mode three situation, you have, this plastic zone is like a circle, in all other
cases it has some shape. And someone comes and says, in the case of thin plate you have
this as a strip, you know, people just could not digest his contributions. So, people are
doubting, is there anything wrong in his approach?

Actually, if you look at the result, that he had quoted in the paper, the experimental and
theoretical prediction, the comparison was quite good, he had not provided the pictorial
representation. This was done by Hahn and Rosenfield in 1965, who provided this
pictorial evidence, then people started believing, yes Dugdale’s model is a valid
approach for evaluating the plastic zone length.

You know, this is very important, like, as students you get doubts scientists also got
those doubts in those days, so it had to be verified by experiments. So, whatever the
theory, that you develop, the predictions have to be substantiated by experimental
observation only, then the theoretical approach can be taken on its face value. It is very
important.

(Refer Slide Time: 22:42)

So, what is the experimental work of Hahn and Rosenfield? They demonstrated the
existence of plastic zone as postulated by Dugdale’s model. And what they did was, they
used 3 percent silicon steel for the specimen and the surface of the test pieces were
electro-polished and etched, and this etching is a very special process. This has helped to
reveal the shape of the plastic zone and essentially, they were metallurgists, they were
handling these kinds of issues earlier, so they had used this for crack problem.

The method of etching ensured preferential attack of individual dislocation. So, whatever
is the plastic zone, it is because of dislocation pile up and the etching process ensures,
that there is a preferential attack and it makes the plastic zone reveal in some form. The
etching has resulted in gradual darkening of the surface as the strain is increased from 1
to 2 percent.

See, you will have to note down, in the context of strain, 1 to 2 percent is very high
because if you really look at material testing, the offset strain, that you take when the
yield strength is not sharply defined, you take that as 0.2 percent offset. So, 0.2 percent
itself is sufficient for the material to yield. So, when you have 1 to 2 percent, the material
has definitely deformed plastically.

Beyond 2 percent strain the etching response has diminished and above 5 percent strain,
no attack at all. These are certain minor details on how the experiment was done. What is
important to us is by the process of etching, because of preferential attack, you are in a
position to obtain the shape of the plastic zone that is what you have to look at. You are
in a position to get the shape of the plastic zone by a particular procedure adopted in the
experiment.
(Refer Slide Time: 25:34)

The etching technique thus revealed both, the extent of the plastic zone and to some
degree the distribution of strain within the zone because I have already mentioned, that 1
to 2 percent, it is able to do well; beyond 5 percent, there is nothing has happened. So,
you also get some crude values on what is a level of strain, shape is very interesting;
level of strain is additional information, though it is crude, it provides some such
additional data.

And what they had done was to reveal plastic zone at various depths, the specimen is
reground to various depths, polished and re-etched. So, you have a via media, whether
the plastic zone remained constant over the thickness of the sheet of the specimen, all
that you have to investigate because we are going to see, how the plastic zone changes
across the thickness. And we will also see, how the plastic zone, say, changes as the
crack propagates, both we will have to look at it.

So, whatever the experimental method of Hahn and Rosenfield, also ensured, not only
you see the plastic zone on the surface of the specimen, but also at various depths, you
have that kind of an advantage. And you have a sample of the plastic zone shape of the
surface and mid-section of a sample and that is given as follows.
(Refer Slide Time: 27:35)

On the surface the plastic zone is like this, at a mid-section the plastic zone is like this.
See, this is a very highly magnified picture, you can for all practical purposes consult this
as a line.

So, this happens on a surface, I would like you to make a sketch of this, you know, you
need to have, these are all very valuable information and only after seeing such a plastic
zone people believe Dugdale’s model. So, from that point of view this is very important
result, and make a neat sketch of this. This happens on the surface and we would also
spend sufficient time on this a few slides later, so make a neat sketch.

So, this is from Hahn and Rosenfield from the paper on local yielding and extension of a
crack under plane stress, published in Acta Metallurgica and what you have to notice is
the mid-section plastic zone resembles the shape of plastic zone predicted by the
Dugdale’s model. So, this provided a comfort in accepting Dugdale’s model as a valid
approach.

And really, what is the use of finding out the plastic zone? See, we have seen earlier
when we looked at SSY approximation. You have to take the effective crack length as a
plus the extent of plastic zone in Dugdale’s model. In the case of Irwin’s model, it was a
plus 1 half of plastic zone; that is what we had looked at. And we will also have to find
out what is the resulting stress intensity factor that would be useful information because
we want to graduate from high strength alloys behave in a brittle fashion to alloys, which
also have some small trace of plastic zone. You could apply fracture mechanics to that,
so you have a simple via media of modifying the crack length.

(Refer Slide Time: 30:06)

And before we proceed to the calculation of stress intensity factor, we would have a look
at the plastic zone length that we have calculated by various methods. We will also look
at what way the results of plane stress and how the results change for plane strain. You
had the simplistic model, where you did not consider redistribution of load, the plastic
zone length is 1 by 2 pi K 1 by sigma ys whole square and in the case of plane strain this
is much smaller, you have this as 1 by 18 pi K 1 by sigma ys whole square.

Then, you have the Irwin’s model, for plane stress it is 1 by pi K 1 by sigma ys whole
square and when you go to plane strain, it is 1 by 3 pi multiplied by K 1 by sigma ys
whole squared; see, this is a very important value. Later on, we are going to study what
should be the selection of the specimen thickness for fracture toughness testing; there
you would see the use of such expressions.

Then, finally, we had seen the Dugdale’s model and we will have to keep in mind this is
applicable only for a plane stress situation. He had taken only thin specimens, essentially
thin plates; he had not calculated the plastic zone size for plane strain situation. And in
Dugdale’s model, the plane stress plastic zone length is pi by 8 multiplied by K 1 by
sigma ys whole squared.

And we had noted that Irwin’s model is an elasto-plastic analysis and this, we have set
this as an elastic analysis. He had considered redistribution of load, whereas Dugdale
considered a strip of plastic zone. You have no singularity because of that, on that
premise he was able to calculate the plastic zone length.

(Refer Slide Time: 32:38)

And we will also look at what is a correction that we have to do for crack lengths.

In the case of simplistic model, you take that as, length as r p for plane stress; for plane
strain, you know, the calculations we have done for nu equal to 1 by 3, you take that as r
p. However, when you go to Irwin’s model, you take that length as r p by 2 because only
on this basis, you know, Irwin has arrived at even the estimation of the plastic zone
length, it is embedded in the development of the equation itself. Finally, when you come
to Dugdale’s model, you take the total length of the plastic zone as the correction for
crack length.

You know, these are all subtle differences. When people handle complex problems, they
do by different approaches. So, when you go to actual application, if you find it satisfies
your experimental observation, you carry on with it.
Now, with the corrected crack length we should also improve the value of stress intensity
factor because the whole idea is to get the stress intensity factor. When you have small
amounts of plastic zone, an intermediate step is, consider the crack as slightly longer
than the actual crack length; that is physics here.

(Refer Slide Time: 34:16)

What we will do is we will calculate the stress intensity factor for an infinite plate
considering the plastic zone size.

K 1 is sigma into pi into a plus delta whole power half. See, you have to note, that delta
is dependent on K, but because the problem is simple, I do not have to do iteration, I can
find out an expression for K 1. For this specific example, a closed form expression is
possible.

So, I will substitute the expression for delta from Irwin’s model. I have this as sigma pi
power half multiplied by a plus K 1 squared divided by 2 pi sigma squared ys whole
power half. And this could be further simplified as sigma root pi a plus sigma root pi K 1
divided by root of 2 multiplied by root of pi sigma ys, these are all intermediate steps.
(Refer Slide Time: 35:47)

And we would finally get the expression for K 1 as, in this fashion, K 1 equal to sigma
root of pi a divided by 1 minus 1 by 2 sigma by sigma ys whole square whole power
half.

See, you note down, the ratio sigma by sigma ys appears in all these calculations. So, we
can also comment, when the stress levels are far below the sigma ys, the correction is
almost negligible. We can also estimate, when sigma is closer to sigma ys, to what extent
the values of K 1 can change. We would see that also in the case of an infinite plate with
a center crack.

If you consider the plastic zone, the K 1 becomes an expression, other than sigma root pi
a you have a denominator, which will be less than 1. So, this value will be eventually
higher.
(Refer Slide Time: 36:56)

And you can also extend this kind of a correction for a finite plate. The moment you go
to your finite plate you cannot avoid iteration; in general, it has to be determined
iteratively, there is no other go. So, I would write this expression as K 1 equal to sigma
multiplied by pi a plus delta whole power half and a function of a plus delta by w. So,
this function also will change when you do the iterative process and it is possible to write
software to do this job.

So, I will get K 1 equal to sigma pi power half multiplied by a plus K 1 square divided
by 2 pi sigma squared ys. You are essentially replacing the delta in terms of Irwin’s
result. I have the function rewritten as a plus K 1 squared divided by 2 pi sigma squared
ys whole divided by w.
(Refer Slide Time: 38:18)

And what are the steps for an iterative evaluation? In the 1st round of iteration, K 1 on
the right hand side is taken based on the actual crack length a.

Evaluated value of K 1 is then fed on the right hand side in the 2nd round. It is very
simple; it is not complicated, straight forward.

The iteration procedure is repeated until two successive values of K 1 are within a
prescribed percentage difference, then you stop.

(Refer Slide Time: 39:13)

See, what is important at this stage is what is the influence of correction of crack length?
I have already mentioned, when the applied stress is far below the yield strength of the
material, plasticity correction is negligible because we have seen in all the expressions, a
ratio of sigma by sigma ys was appearing.

Now, what we will have to know is what way K can change? If the value of applied
stress is closer to sigma ys, K will increase by 40 percent in plane stress and 10 percent
in plane strain, so that relative increase of K is significant. What happens in plane stress
and what happens in plane strain? And it is also worthwhile to look at the expression for
Dugdale’s model written out in a different fashion. When you rewrite, you can write this
as pi by 8 as 0.393. In the case of Irwin’s model, the extension is given as 1 by pi K 1
divided by sigma ys whole square; this 1 by pi is actually 0.318.

So, if you compare, they are not really very different. Irwin has approached the problem
from one methodology, Dugdale has approached in another methodology; both seem to
have got the extension of plastic zone within limits. But the way how it is used in
modifying in the crack length is slightly different in Irwin’s model and Dugdale’s model.
And you have to keep in mind, you know, the problem is very complex, people wanted
to proceed further, so they wanted to have simplifying assumptions and go ahead. From
that point of view, the contribution of Irwin is very significant.

(Refer Slide Time: 41:29)

Now, having looked at all this, what we will look at is what is a variation of plastic zone
shape over the thickness of the specimen? See, you have to look at the mathematical
development. We try to proceed in a way as simple as possible, so you evaluate at
approximate shape, but approximate shape is not the one, which is going to be seen in an
experiment, that was for specimens of different thickness.

On the other hand, Dugdale came and said, for thin plates it is like a strip. That is other
extreme, but for a specimen of given thickens, you have to know how the plastic zone
actually is. So, whatever the discussion I am going to present is based on what is seen
from finite elemental solution or from experiment. Whatever the discussion we had
earlier was approximate analytical approaches and you have to keep in mind the shape of
the plastic zone is not same over the thickness of the specimen, a variation exists.

You had that evidence even in your verification of Dugdale’s approach. You had seen
how the plastic zone was on the surface, how it was there in the middle, even for a thin
plate. Now, if you are having a thickness specimen, you are going to have… And we also
had a discussion, what way we call plane stress and plane strain in the case of fracture
mechanics. If I have a thick specimen, I will consider the surface to behave like a plane
stress and interior when you go, it behaves like a plane strain. So, that kind of an idea I
would use in looking at the plastic zone shape also. So, what you will have to look at is
the plastic zone shape differs over the depth of the specimen; so, that is what is
summarized.

The surface of the specimen is ready to contract and the plastic zone shape can be
approximated to be as that for plane stress case at the surface. And I have already
mentioned the plane stress and plane strain are loosely defined in the case of fracture
mechanics. So, on the surface, the shape is what is seen for plane stress case; in the
interior of the specimen the shape can be approximated to be that for the plane strain
case.
(Refer Slide Time: 44:19)

And what I would like you to do is make a reasonable sketch of this. So, the 1st set of
result, what is presented here is if no redistribution of loading is considered, the variation
of plastic zone is as follows. So, you have to note, that the crack is like this, you have a
thickness of the specimen as B. On the surface, the shape is something like this; interior
to the surface, the shape is like as a butterfly. I have this slipping takes place, I have this
length of the plastic zone, along the crack axis is given as like this and whatever you see
here, is similar to your mode one isochromatics.

See, people will have also done finite element calculation, including the second term,
that is, t stress. They have plotted a very similar to this, so this gives you an
understanding over the depth of the specimen. Over the thickness of the specimen the
plastic zone varies and views the approximation, that surface behaves like a plane stress
situation. On interior it acts like plane strain situation and this is very similar to your
mode one isochromatics.
(Refer Slide Time: 46:11)

Suppose, I consider the redistribution, what way these zones change? You just observe
this, then copy it. What happens is, the inner places, it increases and the surface shrinks.
So, what I will do is, I will just redo the whole thing.

(Refer Slide Time: 46:29)

(Refer Slide Time: 46:47)

(Refer Slide Time: 46:50)

If no redistribution of loading is considered, you have this and this is what happens at the
slip surface and you should also note, that these are known as butterfly shape plastic
zone because it looks like this, corroborated by experiments, that is very important,
similar to mode one isochromatics. And if you consider redistribution, just observe the
animation, this will shrink and that will increase.

So, drawing it on your notes maybe quite difficult because the variation, it may not be
easy for you to show, but nevertheless, you have seen, how this changes when you have
redistribution. So, we have seen over the thickness of the specimen how the plastic zone
changes, now we will have to see, when the crack propagates how does the plastic zone
change? That is also people have studied.

(Refer Slide Time: 47:35)

You know, in the case of Hahn and Rosenfield, we have seen the plastic zone in this
shape. This people have given explanation, so we will have to go and see how the plastic
zone is dictated near the crack-tip and how it is dictated at distances away. So, what I
would appreciate is, you observe this animation today and in the next class, may be you
can make a sketch.

At low stress levels, one observes a hinge type plastic zone; that is what is seen here. At
high stress levels, the plastic zone is projected in front of the crack in the direction
parallel to the crack plane; that is what happens. And what you will have to now
remember is the plane where the maximum shear stress occurs, is different in plane
strain, as well as plane stress. See, this is very important, that influences.
(Refer Slide Time: 48:43)

Now, I take a situation, which is very close to the crack-tip and this is the plane of
maximum shear, it is different. And because of this, what happens is, you have a slip
occurs like this, you have a slip occurs like this, you have planes like this, it occurs like
this. So, you have the steps shown here and slip occurs between these planes and a
tearing action can take place and the crack will propagate.

So, what I will do is, I will just repeat the animation sequence, you just observe. The
plane is different, it is in the x-y plane itself and you will have slip occurring like this.
This is magnified here, you have slip occurs here and because of this you will have crack
propagation, crack extends by tearing.
(Refer Slide Time: 49:56)

Now, let us see what happens at distances away. I am looking at a plane like this, where I
have essentially plane stress. The plane of maximum shear is really out of plane. So, do
not think when I say plane stress everything remains in the plane, it is not so. That is why
we have looked at stress tenser as well as strain tenser.

And you can see the animation here and when you look at the side of the specimen, you
have slip takes place at 45 degrees because this is the plane where you have the
maximum shear stress and this is how you anticipate plastic zone to occur.

(Refer Slide Time: 50:50)

And in fact, if you go ahead and then see how the specimen looks like, this you had seen
it earlier from the context of delineating, the crack surface and fracture surface. So,
initially a crack propagates by tearing action, finally you have shear lip. Now, the
question is, do we have an experimental evidence for this? I have to show, at the end of
the specimen, you should have slip taking place at 45 degrees.

(Refer Slide Time: 51:23)

So, if you have that as an objective, you see the experimental result, you find, on the
surface it is like this, close to the crack-tip. At the ends, you indeed see slip planes at 45
degrees. This was again, reported by Rosenfield, Dai and Hahn. So, very important
result, you know, this gives you an understanding, that people have been able to find out
what happens at the crack-tip.

So, in this class we have essentially looked at Dugdale’s model followed by what is an
extension of crack length you have to take. Then, we evaluated stress intensity factor
with the modified crack-length, then we had looked at how does the plastic zone changes
over the thickness of the specimen and as crack propagates, initially it is dictated by
plane strain followed by plane stress.

The plane of maximum shear stress is different in plane stress and plane strain; that is a
reason why, the plastic zone appears like this.

Thank you.