In almost all loading conditions, metals fail in compression by plastic flow.*  The compressive strength is then simply the yield stress of the metal.  There are many tables of yield strengths for alloys prepared in different ways, so the data are easy to come by.

*The most common exception to this is buckling of thin columns or sheets, which is governed by the Young's modulus of the material.  This is usually not considered a material failure, because it is easy to design against.

Compressive strength will then be improved by anything which increases the yield stress of the metal.  For aluminium, common methods include precipitate hardening (often with copper) and grain size refinement.

Ceramics are a different matter, as they commonly fail by cracking or crushing.  In this case, it is more important to minimise the number and the size of flaws, pores and cracks present in the ceramic.  Again, tables of compressive strengths of ceramics are available.

Polymers are different again.  Below their glass transition temperatures, they tend to be brittle and fail in the same manner as ceramics.  Above their glass transition temperatures, they may fail by yielding, in a similar way to metals.  However, the stiffnesses of polymers tend to be much lower and so they are more prone to elastic failure modes such as buckling.

In almost all loading conditions, metals fail in compression by plastic flow.*  The compressive strength is then simply the yield stress of the metal.  There are many tables of yield strengths for alloys prepared in different ways, so the data are easy to come by.

*The most common exception to this is buckling of thin columns or sheets, which is governed by the Young's modulus of the material.  This is usually not considered a material failure, because it is easy to design against.

Compressive strength will then be improved by anything which increases the yield stress of the metal.  For aluminium, common methods include precipitate hardening (often with copper) and grain size refinement.

Ceramics are a different matter, as they commonly fail by cracking or crushing.  In this case, it is more important to minimise the number and the size of flaws, pores and cracks present in the ceramic.  Again, tables of compressive strengths of ceramics are available.

Polymers are different again.  Below their glass transition temperatures, they tend to be brittle and fail in the same manner as ceramics.  Above their glass transition temperatures, they may fail by yielding, in a similar way to metals.  However, the stiffnesses of polymers tend to be much lower and so they are more prone to elastic failure modes such as buckling.

Dear Philip R. Howie

Thank you for your answer

using grain size refinement what is the limit that can be reached (even if expected) for aluminum (alloys or composites) compression strength?

can it approach 1 GPa? 

I'm no expert on aluminium metallurgy, so perhaps others will be better placed to answer.

The effect of grain size refinement is described by the Hall-Petch relationship.  For pure aluminium, the Hall-Petch coefficients are given by, for example, Wyrzykowski & Grabski, Phil. Mag. A 53, 505-520 (1986).  That paper finds a maximum yield stress of around 150 MPa for pure aluminium.

For alloys, the situation is much more complicated, as the properties depend on composition, as well as processing route.  Commercial alloys seem to top out around 500 MPa (see, for example, the graph on p37 of this lecture: http://www.alueurope.eu/talat/lectures/1501.pdf), so I'd suggest that 1 GPa is unlikely.

measuring Compressive strength is not that easy. You have specimen size requirement to be taken care of - especially the aspect ratio - height to diameter of the specimen. If you respect threquired standard, you can measure the " true compressive strength " Otherwise you may end up with values affected by barreling and buckling as mentioned above !

Regarding the compression strength of brittle materials it is an experimental fact that the crushing stress is between 10 to 15 times greater (absolute value) in compression than in tension. This occur because tensile stress opens cracks wheras compressive close them. Furthermore the experimental values of tensile or compressive or bending stresses follow a Weibull statistic. See on that matter the excellent (and probably the best on this subject) book of Meyers and Chawla : Mechanical Behavior of Materials (Cambridge university press).

In order to increase the crushing stress of brittle material you need to reduce defects density (internal and surface cracks), surface roughness, pores etc... In other word try to obtain a "perfect" material....

Note also that (see Weibull analysis) decreasing the size of the sample augments the fracture stress because the probability of finding a critical defect is smaller in a small volume.

Dear all;

Thank you so much for your valuable answers, grateful to you guys:)