Remolded clays strength

[Okiryu]

I did some unconfined compression tests in undisturbed and remolded highly overconsolidated clays and the remolded strenghts are much higher than the undisturbed strengths.

Properties for these clays are:

Clays: 60.5%
Silt: 32.7%
Sand: 6.8%
Gravel: 0%
Wn: 34%
LL: 67
LP: 30.4
Ip: 36.6

Results are:

qu (undisturbed) = 142 kPa
E50 (undisturbed) = 16 MPa

qu (remolded) = 223 kPa
E50 (remolded) = 24 MPa

Interesting thing is that I did also some UU triaxial tests for the undisturbed clays: c=108 kPa, phi=8. Clays were above water surface.

Clays were remolded under natural water content so it simulates 85% compaction (Modified Proctor OMC was 22.5%). I was surprised with the results. Means that these clays may perform well as engineering fill since their relative high remolded shear strenght. Assuming 1% of the fill height for settlement estimations, these clays can work well as engineering fill under structures or medium height embankments (althouhg I would use these clays only under small-light structures).

Some thing I realized: based on UU triaxial test results, qu's are similar for the undisturbed and remolded clays (so this gives me an idea about the conservatism of the unconfined compression tests).

What do you think about the results? Are any other parameters that I may need to verify if I want to use these clays as engineering fill?

Thanks !!

 

[oldestguy]

For use as compacted structural fill I'd advise thinking about possible expansion if there is available free water sometime in the future. Heavy compaction at lower than natural moisture content can result in severed structural damage if free water reaches these compacted clays and the confinement does not exceed the expansion potential.

 

[Okiryu]

Hi OG, thanks for the comments. Appears that the increased shear strength of the remolded clays does not have too much relevance to you...perhaps as you mentioned there are other more important issues to note here... Actually, I think that compaction at lower than natural moisture will take a lot of effort so local contractors normally do not that. Instead I am assuming that the clay will be compacted at natural water content so the maximum compaction that can be achieved is 86% (based on Modified Proctor test). Possible expansion is a good point, and although I have not heard issues about expansive clays at my area, I will double check this.

 

[oldestguy]

Okiryu:

As to remolded shear strength, your tests are interesting, but I have not delved into this much than noting the common weakening effect when working on slope stability issues.

 

[GeoPaveTraffic]

My first impression is that the clays are not heavily overconsolidated. How sure are you of the overconsolidation?

Mike Lambert

 

[Okiryu]

GeoPaveTraffic, I did also consolidation tests. Pc was 990 kPa and the sample was taken at GL-3m.

 

[GeoPaveTraffic]

Ok, so that settles the overconsolidation question.

Were the samples blocky or did they have any visible structure?

Mike Lambert

 

[Okiryu]

Hi GeoPaveTraffic, thanks for the comments. The samples were blocky, very stable clay cores...

 

[GeoPaveTraffic]

In that case, then I expect that the undisturbed samples failed along existing planes of weakness leading to relatively low strengths when compared to the remolded samples that did not have the weak planes.

Mike Lambert

 

[Okiryu]

Mike, that's very good point. Your comment gave me the impression that you don't normally expect strength values for remolded samples greater than the undisturbed ones. Is there any special reason for that?

Also I was trying to understand your previous comment about you were thinking that the clays were not highly overconsolidated. I was thinking that once you disturb the clays you bring them to a normally consolidated conditions, that's why you did not expect normally consolidated (remolded) conditions to be higher than overconsolidated (undisturbed) conditions....Maybe this is the reason??

 

[fattdad]

if you want to use the clay as engineered earth fill, I'd think you'd want to know the drained strength. I'd also think the drained strength could fatigue over time. Often this leads geotechnical engineers to consider fully softened shear strength as freezing-thawing and wetting-drying cycles can affect cohesion and the like.

So, for undrained strength you've seen some interesting data and received some good thoughts. For drained strength, I'd consider normally-consolidated, reconstituted drained direct shear testing. I'd use the peak value (if you get one) as that'd be fully-softened. Residual values would be for residual design (i.e., if there's some latent failure).

f-d

ípapß gordo ainÆt no madre flaca!

 

[Okiryu]

Thanks f-d. I was not thinking in drained strengths at all... I have read some papers from Professor Stark's research in fully softened and residual shear strengths and found that it can be a good reference....

 

[GeoPaveTraffic]

When comparing remolded undrained strengths to undisturbed undrained strengths, the remolded could be higher or lower depending on the original clay. If the inplace clay is heavily overconsolidated, then the undisturbed strength might be higher than the remolded strength. Remember that by remolding the clay in many ways you are overconsolidating it. I've never seen a well compacted clay that did not have a much higher undrained strength than a normally consolidated clay.

As for drained strengths, I am beginning to use Stark's fully softened strengths for many of my designs. Although I still prefer CD traix tests.

Mike Lambert

 

[Okiryu]

I was thinking that when you remolded samples you make them to be at normally consolidated state. The fully softened strength corresponds to a normally consolidated specimen. That's why I was thinking remolded strengths are expected to be lower than the in-site/undisturbed conditions.

 

[GeoPaveTraffic]

It will depend on how you remold them and what confining pressure you test them at. For unconfined compression tests, any remolded sample is going to be heavily overconsolidated.

Mike Lambert

 

[Okiryu]

Mike, thanks for your comments. This is an intersting topic. I think that it also depends on the characteristics of the insitu clays. For example, if you have insitu overconsolidated soils, are you expecting also that remolded samples can become heavily overconsolidated?

 

[GeoPaveTraffic]

Once you remold the material, the original stress state no longer matters. So once the material is remolded it only matters what confining pressure you put on the sample during testing. Since you are running unconfined tests, the confining pressure is zero. Therefore any compacted sample is going to be overconsolidated.

Mike Lambert

 

[fattdad]

to GPT. The act of compaction is consolidation and will show up as stress history. After compaction, the confining pressure clearly has an effect, but it may or may not be greater than the pre-consolidation pressure of the remolding stage.

f-d

ípapß gordo ainÆt no madre flaca!

 

[GeoPaveTraffic]

fattdad,

I agree that a UU test with a confining pressure may or may not be greater than the effective pre-consolidation pressure induced by the remolding/compaction process.

However, if an unconfined compression test is ran, then relative to the confining stress of the test (i.e. 0 psf); the sample is overconsolidated. You agree?

Mike Lambert

 

[Okiryu]

Maybe we also have to separate this discussion in drained vs. undrained conditions. There was a previous post from fattdad which gave good inputs about this.

Also, regardless the type of test, I got that remolded samples become normally consolidated from Professor Stark notes. I think that Skempton in his Rankine lecture (1964) also talks about this. Remold of clays will affect the orientation of the clay particles (will be re-orientated parallely) and therefore shear strenght will be reduced. So I am understanding this as you are are creating a "new clay" which can be considered as a normally consolidated condition.