Journal of Engineering Geology

---

Introduction
Grouting is one of the most widely used methods of soils improvement in which pressurized grout penetrates in the voids, of the soil. In the grouting method, in addition to reducing the permeability, shear strength and stiffness of the soil increase significantly. However, application of this method in projects such as dam construction and soil improvement requires the use of a large volume of grouting materials in order to satisfy the design criteria. In more recent years, due to the economic and environmental issues, in the case of cement-based grouts, replacing the whole or a portion of Portland cement with other materials has been experimentally investigated. A special type of kaolinite clay called metakaolin has recently been used in concrete, which has yielded interesting results. However, few studies has been conducted on the use of metakaolin in cement-based grouts. Such that, its effect on the mechanical behavior of the grouted soil are not well understood. Accordingly, in the present study, the mechanical behavior of a type of sandy soil grouted with different combinations of Portland cement and metakaolin was experimentally investigated in laboratory scale.
Material and methods
Different materials used in the present study including sand, cement, metakaolin, bentonite and water were selected based on the standard criteria and with the aim of better interpreting the test  result and their differences due to considered influencing factors. Sand was obtained from Malayer Shushab river bed. According to the Unified Soil Classification System (USCS), it is classified as SP. Ordinary Portland cement was used in this study regarding to its widespread application in the practical works. The metakolin is classified as class N pozzolan according to the ASTM C618. Another constituent material of grout is bentonite which is produced by Iran Barit factory as sodium-calcium bentonite. Its liquid limit and plastic index are 296 and 262 percent, respectively. The water used to prepare the grouts was provided from Hamedan drinking water, which according to ASTM C94 has the required quality for grouting operations in laboratory.
The device for grouting specimens was developed at the Soil Mechanics Laboratory of Bu-Ali Sina University during the present study. It equipped with grouting pressure control system and tool for keeping grout in homogeneous conditions during the grouting operation into specimen. The samples were prepared with 0, 5, 10, 15, 20 and 25 percent substitution of cement with metakaolin. Curing time of grouted samples was considered as14 and 28 days.
In order to investigate the factor affecting stress-strain behavior of the grouted sand, the samples were sheared using advanced triaxial apparatus. After passing considered curing time, the samples were sheared considering three levels of confining pressures of 50, 100 and 200 kPa and by applying axial strain rate of 1 mm per minute. For each test, the maximum deviator stress and its corresponding axial strain were recorded. In addition, for studying post peak behavior of grouted soil, for every one of tests, average ratio of deviator stress to the axial strain as softening modulus, was calculated from the deviator stress-axial strain curves.  The moisture content of the samples was also measured according to ASTM D2216 at the end of tests. In the following, the role of different factors influencing stress-stain behavior of grouted sand including; confining pressure, ratio of water to the mix of cement and metakaolin, percentage of metakaolin, curing time and moisture content were investigated.
Results and discussion
Figure 1 shows the effect of metakaolin as alternative of a portion of cement on maximum confined compressive strength and its corresponding axial strain.
For the samples confined by pressure of 50 kPa the maximum confined compressive strength is almost constant by replacing cement with metakaolin up to 10%. However, the amount of axial strain corresponding to the maximum compressive strength of the specimens increases by 6% (Figure 2). For 25% replacement, the maximum confined compressive strength of the samples decreases by 17% compared to the initial state (pure cement). In contrast, the axial strain value related to peak state of most samples has been increased by 4% in comparison to the initial state.
 
In the case of confining pressure of 100 kPa, by replacing up to 10%, the mean confined compressive strength of the specimens was almost constant. However, the amount of axial strain corresponding to the peak state of the specimens has been increased by a maximum value of 18%. For 20% replacement percentage the compressive strength of the specimens has been decreased by about 15% compared to the initial state. However, in the range of 20 to 25 percent, the reduction process has slowed down, which can be due to various factors such as the effects of sample densification during further increase of metakaolin. According to Figure 1, it can be seen that in the range of 20 to 25% substitution, the amount of strain at failure state increased by an average of 40%, which indicates that the sample is more deformable.
In the case of confining stress of 200 kPa, by replacing 10% of the cement with metakaolin, maximum confined compressive strength and its corresponding axial strain, has been increased by approximately 5 and 14%, respectively. With increasing cement substitution up to 25%, the resistance of the specimens decreased by 8% compared to the result of sample grouted with pure cement. Although, axial strain at peak state has been increased by 28%. From the Figure 1, it is obvious that increasing in confining pressure yeilds a considerable increase in the maximum compressive strength of grouted soil. Besides, post peak behavior of grouted soil is also affected significantly by confining pressure. Such that an increase in confining pressure leads to decrease in softening modulus. On the other word, grouted soil displays a more deformable behavior. It should be noted that these aspects of grouted sand cannot be described by unconfined test. However compressive strength of the grouted soils in the majority of case, has been evaluated based on the unconfined test results. 
Conclusion
The aim of this study was to investigate, in laboratory scale, the mechanical behavior of sand grouted with cement-based grout and considering different percentage of metakaolin as an alternative for a portion of cement. The soil samples were grouted using a specific device developed during present study. After passing curing time the samples were sheared using triaxial apparatus by considering three levels of confining pressures. The main findings of this experimental research are as follows:
- Replacing 10% of cement with metakaolin, increases deformability of grouted soil, without reducing compressive strength. Deformability of grouted soil increases with adding more percentage of metakaolin however, in this case compressive strength decreases.
- By increasing confining pressures, more values of metakaolin can be used instead of cement in the grout.
- Increasing confining pressure, increases compressive strength, increases deformability and deceases softening modulus at post peak behavior.
- Shear strength parameters of grouted sand is affected by adding metakaolin into the grout. Increasing the percentage of metakaolin results in small changes in the internal friction angle of the grouted sand, however, the amount of cohesion decreases.