Session 28 "Modeling and Verification"

In trying to find a grout, or some kind of stabilizer that would work in this application, we established some performance criteria. Of those, what I controlled the most was viscosity, then I had a whole list of other criteria. I needed a long, controllable gel time, durability and the ability to set and stabilize the soil. So I looked through a lot of different stabilizers, some really crazy ones and also the typical chemical grouts that are used. The typical chemical grouts were eliminated because of long-term durability concerns, the need for having extended gel times, and several other concerns. I would be happy to discuss that with you further after the talk if you're interested. But, generally it was based on viscosity, gel time, and the need for extended gel time. In terms of the relative density of the sand many other grouts could have been used. But the main issue was trying to develop this way to do low pressure grouting that would not cause invasive disruption of the site. We couldn't go in and grout it, put centers, so most of the other grouts wouldn't work for various reasons. The forty percent relative density samples were prepared by plubiation. That was both during the cyclic testing stage of the research, which happened a few years back, and also during the box model and the laminar box models at RPI. In terms of what criteria were used to determine the success of the grouting, when I initially started this work what I did was take the grouted samples and I subjected them to cyclic triaxial testing. What I did was take a look at the cyclic deformation resistance of different concentrations of grouted colloidal silica treated sand. I looked first at 20% concentration because I wanted something that I knew worked and then I gradually decreased to 15%, 10%, and 5%. What I was most concerned with was that it did not liquefy and that it was able to hold together it and experience some deformation. But most importantly it needed to not liquefy. So, that was the biggest issue in determining the success of the grouting. In terms of the centrifuge modeling it was done by looking at several components: from looking at the pictures of the bending moments, looking at the acceleration records and how the acceleration records were transferred through the soil, looking at the reduction in the lateral displacement and those kinds of issues. So let me show you the difference.