Abstract

Currently, there is an increasing interest in providing more sustainable solutions for mitigating liquefaction and interdisciplinary work has started to introduce the use of nano-materials for this purpose. These materials have some advantages when compared with traditional techniques. For instance, they could be injected at low pressures or delivered into the natural ground water flow to treat a target area through passive remediation, which leads to less carbon emission production compared with normal grouting. In addition, after they gel, the nano-materials can provide increased shear strength and cohesion. This study focuses on the application of laponite, a synthetic nano-clay with the same structure as natural clays. Laponite suspensions have thinning behaviour characteristic in which its viscosity decreases with increase in shearing rate while keeping its gel properties. Moreover, after the shearing load is removed, the suspension recovers its viscosity in a self-healing process. In this study, the shear resistance of laponite suspension is studied by rheological measurements and the effect of the addition of 1% laponite (by weight) on the liquefaction resistance of the host sand is evaluated through cyclic simple shear tests. Results indicate that the number of cycles required to reach liquefaction is increased considerably for low shear stress conditions. In addition, by re-testing samples that have undergone liquefaction, the ability of the mixture to “heal” after liquefaction is confirmed. The results reaffirm the potential of laponite as an environmentally-friendly material for ground remediation purposes.

1 INTRODUCTION

To provide more sustainable solutions to mitigate soil liquefaction, interdisciplinary work has started to introduce nano-materials for soil remediation. Some of them, e.g. colloidal silica, bentonite and laponite, have been proven to be effective in increasing the liquefaction resistance of soils.

Colloidal silica is a chemical grout than can provide cementation and can restrain shear strain occurrence (e.g. Gallagher and Mitchell, 2002; Gallagher et al, 2007). Some researchers have performed cyclic triaxial tests (Gallagher and Mitchell, 2002), resonant column tests (Spencer et al., 2007), and centrifuge model tests (Conlee et al., 2012) and they found that colloidal silica suspensions injected into clean sand can increase the shear modulus and reduce strain deformation. These tests have reported that 10% by weight of colloidal silica suspension is enough to reduce the susceptibility to liquefaction. Colloidal silica seems to be very promising in passive site remediation application near existing structures.

On the other hand, laponite and bentonite are nano-clays that can modify the pore fluid (e.g Rugg et al., 2011; El Mohtar et al., 2013; Santagata et al., 2015; Ochoa-Cornejo et al., 2016) i.e. convert it to a solid-like fluid and delay the generation of excess pore water pressure. Bentonite and laponite suspensions make use of the hypothesis that highly plastic clays increase the resistance to liquefaction (Santagata et al., 2015).

Laponite has received attention in recent years due to its thixotropic properties that make it promising in mitigating liquefaction. Figure 1 schematically describes this phenomenon. Laponite’s thixotropic behaviour is related to the flat-disc shape of its crystal (Figure 1 a, b) and its chemical composition that naturally makes its surface to be negatively charged and its edges positively charged (Barnes, 1997). When dry, its particles conglomerate with each other and form an aggregate of silt/clay size. In suspension, the particles disperse and arrange themselves in a kind of house-of-cards formation (Figure 1 c), and with time, this formation become stronger. Under very small strain, laponite suspension forms a gel that behaves like an elastic solid. When the applied shear stress is larger than a threshold value (or yield shear stress), i.e. τ≥τ_ys, the network breaks and the material flows with a decreased viscosity (Figure 1 d, e). When the shear stress is supressed and the suspension goes back to rest, the structure is recovered and the resistance to flow increases again (Barnes, 1997). This is an indicator of a self-healing material that can be stable to resist new cyclic shear stress episode.

Figure 1: Scheme of thixotropic behaviour on laponite suspensions

Some researchers have studied this material and its potential to mitigate liquefaction. Ochoa-Cornejo et al. (2014) performed cyclic triaxial tests to study its effect on clean Ottawa sand, and they found that 1% laponite by dry mass of sand could increase the number of cycles to liquefaction from about 100 to 600 under similar shear stress. Santagata et al. (2015) performed dynamic oscillatory measurements to compare bentonite with laponite suspensions, and resonant column tests to study their effect on pure Ottawa sand’s dynamic properties. They observed that sand mixed with 3.25% laponite suspensions had similar dynamic response to one with 10% bentonite. However, this material is still in evaluation stage and more research is needed to define its applicability in mitigating liquefaction.

In this study, the effect of adding 1% laponite to pure sand (by weight) is evaluated through cyclic simple shear tests and rheological characterization. The main objective is to assess the self-healing capacity of this material and its contribution to the liquefaction resistance of the host sand.

2 MATERIALS USED AND METHODOLOGY

2.1 Test materials

The host sand used in this study is the Mercer sand, sourced from the Waikato River. Figure 2 shows the grain size distribution of this material. It is a very uniform soil (coefficient of uniformity, C_U=2.1). To assess liquefaction, this material has been studied in loose condition.

Figure 2: Grain size distribution curve of the host sand

On the other hand, laponite is a synthetic nano-clay with a structure similar to natural clays. It possesses no hazard to humans or to the environment. Laponite has a wide range of application, for example, it is usually used in cosmetics, as surface coatings, in some environmental applications as a barrier to trap gasses or contaminants, as well as in biomedical applications (e.g. Kwak, 2010; Tritschler et al., 2016).

In this study, Laponite RD was used, which is a general purpose rheology modifier. When a rheology modifier is dispersed in a fluid, it changes the way in which this fluid deforms when a shear load is applied. In this case, laponite changes the way the water flows when the cyclic loads are applied. To quantify the flow characteristic of laponite suspensions, rheological measurements were performed to measure the shear resistance of laponite suspension and to evaluate the gelling process. In addition, to study the effect of laponite on the liquefaction resistance of sand, undrained cyclic simple shear tests were performed.

2.2 Methodology

The soil samples were prepared by a modified slurry deposition method (Ishihara et al., 1978; Khalili and Wijewickreme, 2008). This method is useful in preparing highly gap-graded samples or, in this case, in preparing uniform sample consisting of granular material and nano-clay in saturated conditions. This method is very replicable and samples with a sand skeleton void ratio of e_sk=0.7581 can be prepared, with less than ±2% of variation. The laponite treatment was fixed at 1% by weight of sand. Then, the amount of water required to have 100% saturation was computed, leading to a laponite concentration by weight of water as 3.4%.

The procedure to prepare the samples is illustrated in Figure 3. First, laponite suspension with a concentration of 3.4% was prepared by pouring the laponite powder in deaired-deionized water, and mixed with magnetic stirrer for 20 minutes at about 1100 rpm. Then, enough suspension was poured into the dry soil, and the mixture was stirred until a uniform slurry was produced. The remaining suspension was poured into the simple shear base, and the slurry previously prepared was poured inside the mould with a spoon (Figure 3c). In some cases, gentle tamping was applied to fit the sample within the target volume. Finally, the excess amount was removed and collected in a container to compute the amount of treatment, which varied from 0.99% to 1.15% of laponite by weight of sand.

Figure 3: Sample preparation for specimens tested on CSST

2.3 Rheological measurements

Suspensions of 3.4% laponite by weight in water were prepared as described above (Figure 3a) and stored in a sealed container. The rheology tests were performed using a Physica UDS 200 rheometer with a cone-plate measuring device (Figure 4a). To avoid errors due to evaporation, the tests were conducted to last no more than 1 hour, and only a small portion of the suspension was used at each time with the rest of the batch stored in the sealed container. The first tests were performed almost immediately after the suspension was prepared, and then the tests were repeated with different specimens from the same suspension every 10 hours for 2 weeks.

Figure 4: Rheological measurements

Each test consisted of 3 steps (Figure 4b). In the first step, small strain amplitude oscillations (γ=0.1%) were applied at a constant frequency 0.56 Hz. The purpose of this step is to assess the current shear strength of the sample, so it did not last for more than 10 minutes. In the second step, a high shear rotation γֹ=100 s^-1 was applied, with the purpose of breaking the internal structure of the developing gel. Finally, the sample was allowed to recover under small strain oscillations γ=0.1% at f=0.56 Hz, and the recovery was evaluated in terms of the time required to reach the same shear resistance that the sample had at the beginning of the test (i.e. in Step 1).

2.4 Cyclic Simple Shear Tests (CSST)

Specimens with a diameter of 63 mm, and height of 24 mm were tested in cyclic simple shear machine. To study the capability of laponite to recover, sand samples treated with laponite were tested in three phases, as depicted in Figure 5. First, the samples were consolidated with vertical load of σ′_v0=100 kPa (Figure 3e) for 72 hours. This was followed by the cyclic shearing stage, in which the cyclic shear stress ratio (CSR=τ ⁄ σ′_v0) was kept constant at 0.1. These two steps were repeated two more times (Phases 2 and 3).

Figure 5: Scheme of the phases in testing laponite-treated sand samples

The sand skeleton void ratio at the beginning of each phase was computed, and pure sand samples were prepared with similar void ratios and compared with the response of laponite samples at each phase. The pure sand samples were consolidated for about 2-3 hours and then tested at CSR=0.1.

3 RESULTS AND DISCUSSION

3.1 Rheological measurement results

Figure 6a shows the shear modulus of the suspension versus the elapsed time since it was first prepared. From this result, it can be observed that at least 72 hours are required for the gel to have at least 80% of its final shear strength. This observation is consistent with those obtained by other researchers (e.g. El Howayek, 2011; Santagata et al., 2014; Ochoa-Cornejo et al., 2016).

On the other hand, Figure 6b shows the results only from the recovery step (described in Section 2.3). The markers’ colour is proportional to the elapsed time since the suspension was first prepared. The results indicate that after enough time has elapsed (T_e >3.5 h), the recovery time is almost instantaneous, i.e. within minutes, the shear strength has recovered. Consistent with Figure 6a, for samples just prepared, the time to reach a steady value is about 72 hours.

Figure 6: Time effect on laponite suspensions: (a) elapsed time since suspension preparation; and (b) recovery step

3.2 CSST results

Figure 7 shows the results from cyclic simple shear test for CSR=0.1, in terms of increase in excess pore water pressure ratio and double amplitude shear strain development. In the figure, the size of the marker is proportional to the relative density after consolidation, while the colour of the marker is related with to the time that elapsed since the sample was first prepared, i.e. white colour for less than 3 hours and black colour for more than 216 hours. The first column represents the first phase of the tests in which the samples were prepared with a skeleton void ratio of e_sk=0.758±0.015. The samples treated with laponite shows a gradual development in pore water pressure ratio until the 10^th cycle, afterward the rate of development of pore water pressure increases, reaching liquefaction (i.e. r_u1) just a little later than the pure sand samples. Columns 2 and 3 show the results of re-testing Laponite samples two times after liquefaction. The initial void ratio before consolidation was estimated to be e_sk=0.710±0.015 and e_sk=0.681±0.003, respectively, for Phases 2 and 3. These samples were compared with pure sand specimens prepared with similar void ratios and the effect of time and self-healing is clear in the graphs.

Figure 7: CSST results for CSR=0.1

These results are consistent with the rheology results shown in Figure 6, in which laponite suspension recovered very fast (within an hour), and the suspension continued to gain shear strength with time. Thus, in Phase 2, the number of cycles required for liquefaction to occur increased by about three orders of magnitude with the treatment; in Phase 3, the sample underwent more than 10⁵ cycles with shear strain less than 1%. Moreover, during Phase 3, the excess pore water pressure oscillated more at the end of the test, possibly indicating that the sample was recovering while the test was in progress.

4 CONCLUSIONS

This paper presented the results of rheologic measurements on laponite suspensions and cyclic simple shear tests on laponite-mixed sands at CSR=0.1, in which the self-healing characteristic of laponite was confirmed. The results from the rheometer tests indicated that at least 72 hours are required after the suspension has been prepared for the fluid to provide enough resistance and to behave as a gel; after that period, if shear stress was applied, the suspension yielded, but it recovered within one hour. These results were consistent with what was obtained in the cyclic simple shear tests, where the same sample was subjected to 3 phases of consolidation and cyclic shear application. Even though the void ratio of the sample decreased after each phase, the improvement was considerable when compared with pure sand samples with similar relative density. Therefore, the samples did not only recover, and being capable to resist a new set of cyclic stresses, it endured even more cycles because the laponite suspensions continued to harden.

These results indicated that with only 1% of laponite by weight of sand, it was possible to increase considerably the liquefaction resistance of pure sand. However, the results of CSST suggested that more than 72 hours may be required for effective treatment. After a sufficient time had elapsed after the sample preparation (more than 144 hours), the sample underwent more than three orders of magnitude of cyclic loading application than pure sand samples with similar relative density to liquefy. However, given the thinning behaviour of laponite suspension (i.e. decrease viscosity with increase in shear), such effectiveness observed for CSR=0.1 many not occur for higher CSR; more tests are currently planned to study the response at different levels of CSR, and different elapsed times and relative densities.

Laponite is currently being used in preliminary stages as a soil stabilizer, but the results presented in this paper reaffirmed their potential for ground remediation purposes. In practice, laponite could be injected or deposited in the ground and allowed to flow in passive remediation, where it could provide a more environmentally-friendly alternative to stabilize the soil since it poses no hazard to people and environment, and could help reduce carbon emission.

REFERENCES

Barnes, H. A. (1997) ‘Thixotropy – A review’, Journal of Non-Newtonian Fluid Mechanics, 70(97), pp. 1–33. Conlee, C. T. Gallagher, P. M., Boulanger, R. W. and Kamai, R. (2012) ‘Centrifuge modeling for liquefaction mitigation using colloidal silica stabilizer’, Journal of Geotechnical and Geoenvironmental Engineering, 138(11), pp. 1334–1345.

Gallagher, P. M. and Mitchell, J. K. (2002) ‘Influence of colloidal silica grout on liquefaction potential and cyclic undrained behavior of loose sand’, Soil Dynamics and Earthquake Engineering, 22, pp. 1017–1026.

Gallagher, P. M., Pamuk, A. and Abdoun, T. (2007) ‘Stabilization of liquefiable soils using colloidal silica grout’, Journal of Materials in Civil Engineering, 19(1), pp. 33–40.

Ishihara, K., Sodekawa, M. and Tanaka, Y. (1978) ‘Effects of overconsolidation on liquefaction characteristics of sands containing fines’, Dynamic Geotechnical Testing. ASTM International, pp. 246–264.

Khalili, A. and Wijewickreme, D. (2008) ‘New slurry displacement method for reconstitution of highly gap-graded specimens for laboratory element testing’, Geotechnical Testing Journal, 31(5), pp. 424–432.

Kwak, J. (2010) ‘Layered silicate particles filled polymer nanocomposite for barrier applications’, PhD Thesis, University of Florida.

El Mohtar, C. S., Bobet, A., Santagata, M. C., Drnevich, V. P. and Johnston, C. T. (2013). ‘Liquefaction mitigation using bentonite suspensions’, Journal of Geotechnical and Geoenvironmental Engineering, 139(August), pp. 1369–1380.

El Howayek, A. (2011) ‘Characterization, rheology and microstructure of laponite suspensions’, Master Thesis, Purdue University.

Ochoa-Cornejo, F. Bobet, A., Santagata, M. and Johnston, C. and Sinfield, J. (2014) ‘Liquefaction 50 years after Anchorage 1964: How nanoparticles could help prevent it’, Proceedings of the 10th National Conference in Earthquake Engineering.Ochoa-Cornejo, F. Bobet, A., Johnston, C., Santagata, M. and Sinfield, J. (2016) ‘Cyclic behavior and pore pressure generation in sands with laponite, a super-plastic nanoparticle’, Soil Dynamics and Earthquake Engineering, 88, pp. 265–279.

Rugg, D. A. Yoon, J., Hwang, H. and El Mohtar, C.S. (2011) ‘Undrained shearing properties of sand permeated with bentonite suspension for static liquefaction mitigation’, ASCE Proceedings of the Geofrontiers. Dallas, Texas, pp. 677–686.

Santagata, M. C. Clarke, J.P., Bobet, A., Drnevich, V. P., El Mohtar, C.S., Huang, P.T. and Johnston, C. T. (2014) ‘Rheology of concentrated bentonite dispersions treated with sodium pyrophosphate for application in mitigating earthquake-induced liquefaction’, Applied Clay Science. Elsevier B.V., 99(9), pp. 24–34. Santagata, M. C. Bobet, A., El Howayek, Ochoa-Cornejo, F., Sinfield, J.V. and Johnston, C. T. (2015) ‘Building a nanostructure in the pore fluid of granular soils’, Geomechanics from Micro to Macro. Edited by K. Soga et al. CRC Press, pp. 1377–1382.

Spencer, L., Rix, G. J. and Gallagher, P. M. (2007) ‘Dynamic properties of colloidal silica gel and sand mixtures’, 4th International Conference on Earthquake Geotechnical Engineering, (1324).

Tritschler, U., Zlotnikov, I., Fratzl P., Schlaad, H., Grüner S. and Cölfen H. (2016) ‘Gas barrier properties of bio-inspired Laponite – LC polymer hybrid films’, Bioinspiration & Biomimetics. IOP Publishing, 11.