Effects of Curing Time on the Mechanical Property and Microstructure Characteristics of Metakaolin-Based Geopolymer Cement-Stabilized Silty Clay

Research Center of Mine Underground Engineering, Ministry of Education, Anhui University of Science and Technology, Huainan 232001, China School of Civil Engineering and Architecture, Anhui University of Science and Technology, Huainan, China State Key Laboratory of Mining Response and Disaster Prevention and Control in Deep Coal Mine, Anhui University of Science and Technology, Huainan, Anhui 232001, China


Introduction
Cement soil reinforcement technique has been accepted worldwide in the areas of foundation engineering, slope reinforcement, road construction, and rammed earth wall engineering for many years [1][2][3][4][5]. It has been proved as an effective reinforcement technique with many advantages, such as high usage rate of origin soil, low cost, and convenient construction [6,7]. However, in the cement production process, about 900 kg CO 2 will be emitted into the atmosphere per tonner of cement which contributed to global warming and climate change [8][9][10][11]. erefore, some supplementary cementitious materials (e.g., metakaolin, fly ash, slag, and nanosilica) have been used for replacing part of cement to reduce CO 2 emission associated with cement production [12][13][14].
As a typical pozzolanic material, metakaolin (MK) is an amorphous aluminosilicate compound produced by calcinations of pure kaolinite at 500°C to 550°C [15], which has high activity and has attracted increasing attention in recent years that could partly replace ordinary Portland cement [16]. Compared with other supplementary cementitious materials, MK has many advantages, including higher activity, larger specific surface, and finer particle size [17]. Studies have emerged that offer contradictory findings about the effects of MK content on physical and mechanical properties of cement soil and concrete materials and acquired abundant accomplishment [15,16,[18][19][20][21][22][23]. Wu et al. [15] studied the effect of MK contents on the strength performance of cement-modified soil, and test results reveal that both the unconfined compression and splitting tensile strengths are improved, and the suitable ratio of MK to cement ranges from 1/3 to 1/2. Moreover, the addition of MK led to a higher quantity of hydration products and denser microporosity distribution according to X-ray diffraction (XRD), scanning electronic microscopy (SEM), thermogravimetric analysis (TGA), and mercury intrusion porosimetry (MIP) test results, and a similar series of experiment results are also found [19]. Wang et al. [22] investigated the effects of curing time and coal-metakaolin content on the properties of cemented sandy soils, and test results indicate that the unconfined compression strength at 28 d was improved by 1.68-2.18 times by incorporating coalmetakaolin. Furthermore, similar tests were performed on cemented silty soil, and the optimum mass ratio of coalmetakaolin to cement is between 1 : 6.5 and 1 : 4 [23]. Experimental and theoretical results mentioned above show that MK can significantly improve the strength property of concrete and cemented soils at the appropriate proportion. Meanwhile, the optimum mass ratio of MK to cement is affected by the soil and MK types and test conditions including purity and grain sizes.
Curing time has a significant influence on chemical reaction degree between cement, MK, and water [24,25], which affects the strength and deformation properties of MK-based geopolymer cemented soil. Specifically, MKbased geopolymer cemented soil requires sufficient time to achieve high mechanical properties and elevated durability. However, by summarizing the theoretical and experimental studies achievements, it can be found that researches mainly concentrate on the MK content and its improvement mechanisms, but there are still insufficient data for effect of curing time, especially at an early curing age. In addition, studies focused on microstructure characteristics of MK cement soil under different curing times are relatively insufficient. Hence, the objective of this research is to investigate the effects of curing time on mechanical property (i.e., unconfined compressive strength, elasticity modulus, and stress-strain relationship) and microstructure characteristic of cement-stabilized soil incorporate with different MK contents. e mass ratio of binder (cement and MK) to dry soil is 15% in this test. Finally, SEM and nuclear magnetic resonance (NMR) techniques are used to study the porosity and microstructure characteristic of MK cement silty clay with 26.5% water content, to analyze the internal microstructure change of MK-cemented silty clay under different MK contents (i.e., 0%, 2%, 4%, and 6%) and curing times (i.e., 1 d, 3 d, 7 d, 14 d, and 28 d).

Materials.
e soil used in this study was collected from a foundation reinforce engineering in Huainan (Anhui, China). e physical properties of the soil are shown in Table 1. According to the Chinese National Standard GB 50021-2001 (2009 version) [26], it was categorized as "silty clay." e ordinary Portland cement (OPC) and MK in this experiment were produced by Bagongshan cement plant and Shanghai HaoFu chemical company, respectively. e chemical compositions of OPC and MK are listed in Table 2.
e average particle size, specific surface area, and activity index of MK were 1 μm, 2000 m 2 /kg, and 122, respectively. From Table 2, it was clearly noticed that the main components of MK were SiO 2 and Al 2 O 3 .

Sample Preparation.
e undisturbed silty clay was put in a dryer at 105°C for more than 24 h to constant weight and then crushed and sifted larger size (2 mm) from the sample. ereafter, water equal to the original water content (26.5%) was added continuously to the mixture and mixed for about 30 min, and then the silty clay was stored in a closed container for more than 24 h to make the soil saturated. In this experiment, 15% binder (mass ratio of binder to dry soil) was adopted to stabilize the silty clay according to the design guidelines for cemented soil pile established by standard JGJ 79-2012 [27]. e mass ratios of MK to OPC in this test were 0 : 15, 2 : 13, 4 : 11, and 6 : 9, respectively, and the corresponding group were designated MK0, MK2, MK4, and MK6, respectively. Sample preparation procedure was as follows: (1) put MK into the silty clay and mix evenly; (2) mix sodium silicate and NaOH solution (50% concentration) with the mass ratio of 5.8 to 1 for the preparation of alkali activator [28], and then spray them into the mixture equably with a sprinkling can; and (3) add the calculated cement into the mixture and the corresponding water. e action of alkali activator used in this test can provide alkaline environment for MK to perform a three-dimensional network structure which is composed of silica tetrahedron and alumino-octahedron, as shown in Figure 1.

Unconfined Compressive Strength (UCS) Test.
e final mixture was divided into five portions and put into a cubic mold with dimensions of 70.7 mm × 70.7 mm × 70.7 mm according to specification for the mix proportion test of cement-mixed soil established by standard JGJ/T 233-2011.
To conduct the UCS test, the top of each layer was scratched before adding the next layer to promote suitable bonding. e MK-based geopolymer cemented silty clay samples were removed from the molds after 24 h, and each sample was cured in a standard concrete materials curing room with temperature of (20 ± 2)°C and humidity of (95 ± 2)% [30]. ese samples were tested after curing 1 d, 3 d, 7 d, 14 d, and e UCS tests were conducted by an RWS-200 microcomputer controlled electronic universal testing machine at a loading rate of 1 mm/min. Furthermore, three parallel samples were prepared to improve the test reliability, and test results were considered suitable for presenting if their average error was less than 5%.

NMR Test.
NMR technology has been widely used to obtain the pore structures of cement-based materials [31][32][33][34][35], which was conducted to measure the porosity of MKbased geopolymer cemented soil under different test conditions in this test. In NMR tests, dimension of Φ50 mm × 50 mm is the maximum size for NMR device. e maximum particle diameter for MK-based stabilized clays is the soil particle with 2 mm, and the larger the specimen size, the precise the test results. In addition, considering the possible damage in sample transport process, specimens with the size of Φ50 mm × 25 mm under different curing times and MK contents are prepared for NMR tests.

SEM Test.
Quanta FEG 650 pattern environmental scanning electron microscope was used to investigate the microstructure characteristic of MK-based geopolymer cemented soil under different MK contents and curing times, all of the samples were obtained from the central part of complete samples [36,37]. ese samples were coated with gold for improving their conductivity before imaging.
e SEM instrument was operated at 10 kV.

Stress-Strain Relationship and Failure
Mode. e typical stress-strain curves of MK-based geopolymer cemented silty clay obtained from UCS tests are shown in Figure 2. e aforementioned figure reveals that with the increase of curing time, the peak stress increases while the stress decreases sharply after peak strain, and the corresponding failure modes change from plastic failure to brittle failure. Figure 2(b) illustrates that at 1 d, the stress of MK6 group increases slowly with increasing axial strain compared with that for MK0, MK2, and MK4 group. With the increase of axial strain, the increase rate of MK2 group is higher compared with other groups.

UCS and Elasticity
Modulus. UCS and elasticity modulus are two significant parameters to evaluate the resistance ability of MK-cemented silty clay subject to external loading. In this test, the peak stress is defined as the UCS, and the elasticity modulus (E 50 ) is defined as the slope between the original point and 0.5 UCS. Test data of UCS and E 50 of MKbased geopolymer cemented silty clay under different MK contents and curing times are shown in Figure 3. Figure 3 reveals that both the UCS and E 50 increase with the increase of curing time, and the increase rate shows obviously different at different curing time periods, specifically, the increase rate of UCS and E 50 from 1 d to 7 d is faster compared with that from 7 d to 28 d. In addition, with the incorporation of MK, both the UCS and E 50 show first increase and then decrease, and the average values of UCS and E 50 of MK2 and MK4 groups are larger compared with that of MK0 group under the same curing time. However, both the UCS and E 50 of MK6 group are smaller than those of MK0 group, indicating that too much MK content weakens the property of MK-cemented silty clay. For example, at 28 d, the average values of UCS and E 50 of MK2 group are 1.32 and 1.30 times compared with the control group MK0, respectively, agreeing with the previous study results [22]. However, these values of MK6 group are only     0.53 and 0.59 times compared with MK0 group. Hence, the optimal mass ratio between MK and cement is 2 : 13 in this experiment, and the corresponding strength of MKcemented silty clay is the highest. As a contrast, Wang et al. [22] found a similar optimum mass ratio (1 : 6.5-1 : 4) for CMK cemented silty soil, and 1 : 4 mass ratio was found for sandy soil [23]. Wu et al. [15] reported that the optimum mass ratio ranges from 1/3 to 1/2 for silt sand. e reactivity degree of geopolymer and cement is affected by test conditions. Hence, the difference between the above test results may be caused by the soil and MK types [38,39]. After a detailed evaluation of the changes in the UCS and E 50 data, to evaluate the effect of curing time on the mechanical property of MK-cemented silty clay, strength increase coefficient (ζ s ) and elasticity modulus increase coefficient (ζ e ) are defined in this research and can be calculated as follows: where UCS n and UCS 28 are the unconfined compressive strength of MK-cemented silty clay at curing time of n d and 28 d, respectively. E 50n and E 50,28 are the elasticity modulus of MK-cemented silty clay at curing time of n d and 28 d, respectively. Figure 4 presents the calculated results of ζ s and ζ e at different curing times. It is clearly noticed that (1) compared with MK0, MK2, and MK4 groups, MK6 group shows relatively small values of ζ s and ζ e from 1 d to 7 d because there have been no sufficient hydration products to react with geopolymer. For MK6 group, the values of ζ s at 1 d and 3 d are 21.9% and 40.6%, respectively, which are much lower than those of other three groups; (2) the values of ζ s and ζ e for MK2 and MK4 groups are higher than those of MK0 group at 1 d; however, the values of ζ s and ζ e are close to MK0, MK2, and MK4 groups from 3 d to 28 d. In addition, ζ s can reach 75.8% to 77.8% at 7 d for MK0, MK2, and MK4 groups, and the corresponding values of ζ e range from 79% to 80.8%, indicating that these three groups show better early strength and deformation resistance ability.
In ground treatment and foundation supporting design engineering practice, the ratio between E 50 and UCS is an important evaluation parameter which should lie in the range of 100 to 120 according to the standard JGJ 79-2012. Hence, the relationship between E 50 and UCS at different curing times in the present study and other test results [40] are collected and shown in Figure 5. Yilmaz and Ozaydin [40] found that the ratio between E 50 and UCS of belite cement (BC) clay and ordinary Portland cement (OPC) clay were 79 and 107, respectively. e ratio between E 50 and UCS lied in the range of 60 to 145 in the present study. In addition, most of the test results at 28 d satisfy the standard request.

NMR Results.
e porosity of MK-based geopolymer cemented silty clay with different test conditions are measured by NMR. e obtained NMR data are presented in Table 3. e test samples with 6% MK content at curing time of 1 d, 3 d, and 7 d fall to pieces during saturated process because of their low strength. Moreover, the relationships among porosity, UCS, and E 50 are shown in Figure 6. Table 3 reveals that at the same curing time, MK2 group shows the smallest porosity values compared with other three groups. Moreover, the porosity decreases with increasing curing time at the same MK content, which is attributed to the pozzolanic reaction between geopolymer, cement, and water [41]. Figure 6 illuminates that both UCS and E 50 of MK-based geopolymer cemented silty clay decrease with the increase of porosity and exhibit a linear correlation. In this test, the porosity of MK-based geopolymer cemented soil increases from 21.22% to 26

SEM Analysis.
e microstructure can reflect the variations of physical and mechanical properties of MK-based geopolymer cemented silty clay [15,22,23]. SEM tests are conducted to observe the microstructure characteristic and hydration products in the bulk binder phase. Figure 7 shows the SEM images of samples at curing time of 28 d with different MK contents at 800x amplification.
MK content has a significant influence on the microstructure characteristic of MK-based geopolymer cemented soil. Figure 7(a) is the representative SEM images of MK0 group. It can be found that the microstructure of mixture without MK is loose and not uniform, and some cracks and pores are observed between soil particles. As a contrast, Figure 7(b) demonstrates that with the incorporation of MK, the microstructure of MK2 group is relatively denser and uniform, while the number of cracks and pores clearly decrease. is finding primary attributes to accelerate cement hydration caused by secondary reaction between geopolymer and hydration product of cement [29]. For the MK4 group, there exist more cracks and pores compared with that of MK2 group, while better than that of MK0 group. Furthermore, widely distributed cracks and pores are found for MK6 group, which also exhibits the loosest microstructure. Hence, 6% MK content has a negative effect on MK-based geopolymer cemented silty clay, which reduces its UCS and E 50 . By comparing the microstructure characteristics of four groups, it is obvious that the microstructure of the sample modified by 2% MK is most homogeneous and dense. Figure 8 represents SEM images of MK2 group with different curing times at 800x and 12000x amplification, respectively.
e SEM images at 800x amplification can easily observe the cracks and pores inside, while the microcracks, micropores, and hydration products such as C-S-H, which improve the property of MK cemented soil [41], can be observed at 12000x amplification. As shown in Figure 8     Advances in Materials Science and Engineering quantity of C-S-H has covered most of the soil particles and connected with each other. Figure 8(e) illustrates that the structure of MK2 at 28 d is denser, while both macropores and micropores are filled with C-S-H, thereby improving its macromechanical properties, such as strength and elastic modulus.

Conclusions
(1) Both the UCS and E 50 increase with the increase in curing time, and the increase rate from 1 d to 7 d is faster compared with that from 7 d to 28  Data Availability e datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this paper.