Why Do Ceramic Materials Exhibit Different Brittle Fracture Points?

[ali297]

Hi

I have 5 different experimental data of a ceramic undergoing compressive stress, and I've graphed them. There are some significant differences among these graphs compared to other materials I've compared (aluminium, pvc etc). Each has multiple fracture points etc...my question is what could be the reason for these discrepancies?

Thanks

 

[Studiot]

To steal a phrase

Google is your friend

http://www.efunda.com/formulae/solid_mechanics/failure_criteria/failure_criteria_brittle.cfm

http://www.efunda.com/formulae/solid_mechanics/failure_criteria/failure_criteria_crossover.cfm

 

[ali297]

I've already been to that page, but didn't understand it. I'd appreciate if you can very briefly explain the reasons.

 

[Studiot]

Well your initial post was a tad short and I didn't fully understand what you are trying to show.

What do you mean by "I've graphed them" ?

The strength of ceramics is quite different from that of ductile metals. Ceramics genrally contain many flaws (microcracks) so failure is generally controlled by fracture mechanics and statistical considerations, not plasticity considerations as with metals.

In particular the failure mechanism for compression is different from that of tension or bending. Why are you doing compression and what other measurements did you do or do you have to compare with?

 

[ali297]

It was an experiment we had where we put a piece of ceramic in a machine which slowly crushed it while it recorded data. The graph is the compressive stress vs strain.

But I think I get your point that the microcracks in ceramics differ from one another, which is primary cause for the differences in the graphs produced? (each have different fracture points etc)

 

[Studiot]

Do you know what fracture toughness is?

 

[ali297]

No, does it matter?

Here is the graph.

 

[Studiot]

Only that I will avoid using fracture toughness in the explanation.

All ceramic materials contain many small random cracks.

These cracks are randomly distributed
They are of random length
They are of random orientation.

So when we stress a ceramic sample all three of these random elements come into play.

For tension and compression the stress is distributed evenly across the sample so it does not matter where the crack is, all locations will be equally affected.

Cracks propagate when they are pulled apart (obvious really). The larger the crack and the more the crack is oriented perpendicular to the stress the easier it is to propagate.

So wherever it is in the sample the larges crack at right angles to the tension will propagate if the tension is large enough.
For different samples it the starting point in relation to length and orientation will vary randomly so the failure tension will vary randomly.

For bending the stress varies across the sample so the tension is maximum at the edges and zero somewhere in the middle.
So the largest crack is less likely to appear at thee dges so the failure tension in bending will be larger (up to 5 times larger).

For compression, cracks at right angles to the stress will be closed up not pulled apart so will not fail.
Cracks parallel to the stress will elongate slowly due to poisson (bursting) effects. So a failure crushing zone of small cracks will gradually extend in compression until the whole zone fails.
So the compressive strength is dependent on the average of a large number of cracks.
It can be 15 times greater than the tensile failure strength.

Sorry this was rather rushed

 

[xxChrisxx]

I've got nothing to add to the thread really but I've got to say. Studiot, that is a damn fine explination of fracture mechanics without jargon.

 

[ali297]

Thanks for the info, appreciate it

 

[gteuae]

Superb Info... thanks much..

Regards..

 

[Studiot]

Nice to know it was useful, thanks for the feedback.

 

[cloa513]

While studiot's description is reasonable, it really comes back to manufacturing variability its harder to control production of ceramics to ensure the same strength.
It needs ceramic's total crack characteristics
(number, length etc.) are the same- long cracks are eliminated by general control but the important small ones are hard to control. Metals only need the general processing to be the same to get close strength properties- short cracks have little effect as they aren't critical.