Keywords

1 Introduction

Cemented soil is an artificial mixture of materials by mechanical mixing or jet punching the foundation of the natural soft soil and cement slurry (or powder) stirred together, mainly a mixture of cement and soil [1]. Due to low cost, wide range of materials, simple construction technology, good performance and so on, cemented soil is widely used in foundation treatment and road engineering [2]. In recent years, cement production has caused a lot of energy problems and environmental problems. To reduce the demand for cement consumption and further improve its mechanical properties, many scholars used a certain amount of additives to replace the cement in the cemented soil [3]. Lang et al. used cement and steel slag powder to treat mucky soil and found that steel slag powder can effectively alleviate the strength loss caused by humic acid in mucky soil [4]. Furlan et al. studied the effect of fly ash on the mechanical properties and microstructure of cemented soil, and found that fly ash can increase its compressive strength and improve its pore structure [5]. Avci et al. studied the influence of sodium silicate on the mechanical properties and permeability of cemented sandy soil, and found that sodium silicate can significantly improve its compressive strength and permeability [6]. Wang et al. studied the effect of nano-SiO2 on the mechanical modification coastal of cemented soil in the early age, and found that nano-SiO2 can improve its splitting tensile strength and elastic modulus, but it would aggravate its brittleness [7].

Coal metakaolin (referred to as CMK) is the coal kaolin ( \(Al_2 O_3 \cdot 2SiO_2 \cdot 2H_2 O\), referred to as AS2H2) that calcined dehydration at a certain temperature (500 ℃–900 ℃) to form anhydrous calcium aluminate (\(Al_2 O_3 \cdot 2SiO_2\), referred to as AS2) [8]. Coal kaolin, also known as coal gangue, is a by-product of coal. This mineral is mainly distributed in the north China region, which is a unique and precious resource in China [9]. CMK is a highly active mineral admixture with high pozzolanic activity. It is mainly used as a concrete admixture and can also be used to make high-performance geopolymer, but it is rarely studied in cemented soil. Therefore, the research on cemented soil with metakaolin has some positive social and economic value.

In this paper, the compressive strength of cemented soil with different CMK contents and ages was performed by the indoor simulation experiment. Combined with the phase composition and microstructure of the hydration product, the strength mechanism of cemented soil was revealed by means of X-ray diffraction (XRD) and scanning electron microscopy (SEM) image analysis. The relationship between the compressive strength and the age of cemented soil was analyzed. The compressive strength prediction formula of cemented soil by age was set up.

2 Materials and Testing Methods

2.1 Experimental Materials

The soil used in the experiment was sandy soil, sourced from a construction site in Taiyuan, Shanxi Province, China. The soil was dried, crushed, and passed through by a 2 mm sieve. The particle composition of the soil is shown in Table 1, according to Chinese standard GB50123-2019. The cement is P.O 42.5 ordinary Portland cement produced by Taiyuan Lionhead Cement Co., Ltd.. The CMK was HP-90B coal metakaolin produced by Shanxi Jinyang Co., Ltd.. It was white powders, the particles of less than 2 μm (6250 mesh) accounted for (90 ± 3)% of the total mass. The chemical composition of cement and CMK is shown in Table 2.

Table 1. Particle composition of soil.
Full size table
Table 2. Chemical composition of cement and CMK.
Full size table

2.2 Sample Preparation

The specific mix proportions of cemented soil were shown in Table 3. The cement and CMK were cementitious materials. The amount of cementitious material added was 15% of the dry soil mass. In the mix proportions, CMK was used to replace part of the cement to prepare cemented soils. The content of water required for mixing was 20% of the dry soil mass according to Chinese standard JGJ 79-2012.

Table 3. Mix proportions.
Full size table

The dried soil, CMK and cement were weighed into the mortar mixer and stirred for 2 min. After the solid particles were fully and evenly stirred, the corresponding quality tap water was added to stir for another 2 min to make the solid-liquid mixture well mixed. The mixture was divided into a 70.7 mm × 70.7 mm × 70.7 mm steel mold. The whole steel mold was put on vibration shaking for 1 min to ensure bubbles discharged. Then the surface of the samples was flattened, and the steel mold was wrapped with plastic wrap. After 24 h, the steel mold was dismantled. All cemented soil samples were numbered and placed in a water tank for curing.

2.3 Sample Preparation

The compressive strength of cemented soil samples was tested by a microcomputer controlled electronic universal testing machine (WDW-100) at the age of 3 days, 7 days, 14 days, 28 days, 60 days and 90 days. The maximum load of this testing machine was 100 kN, the error of load and displacement was better than ± 1%, and the loading rate was 0.1 kN/s. In the analysis of experimental data, the compressive strength was calculated by six parallel samples according to Chinese standard JGJ/T 233–2011.

After 28 days of compressive strength testing, selected about 10 g specimen from each sample, placed them in an ethanol solution for dehydration, and ground them through a 0.075 mm sieve. The specimens were measured by an XRD-6000 X-ray diffractometer produced by Shimadzu Co., Ltd.. The scanning angle was 4°, the ending angle was 60° and the speed was 6°/min. The phases were analyzed by JADE6.5 software.

The samples were cut into 20 mm × 20 mm × 5 mm specimens, then polished and cleaned. An SBC-12 ion sputtering instrument was used to plate gold film on the surface of the specimens. Then the specimens were put into TM-3000 scanning electron microscope produced by HITACHI Co., Ltd.. The acceleration voltage of the scanning electron microscope is 15 kV, and the magnification is 2000 times.

3 Results and Discussion

3.1 Analysis of Compressive Strength Test Results

The relationship between the compressive strength of cemented soil and the CMK content at all ages is shown in Fig. 1. In general, the compressive strength of cemented soil mixed with CMK was significantly improved compared with that of unmixed cemented soil. When the CMK content was less than 3%, the compressive strength of cemented soil increased with the increase of the CMK content. When the CMK content was more than 3%, the compressive strength of cemented soil decreased with the increase of the CMK content. When the CMK content was 3%, the compressive strength reached the maximum. This indicates that the use of CMK to replace a certain amount of cement contributes to improving the compressive strength of cemented soil.

This may be because the cement content decreases with the increase of the CMK content. When the CMK content is small, the cement can not only ensure the strength produced by its early hydration, but also can produce an appropriate amount of Ca(OH)2 to react with the active ingredients (Al2O3, SiO2) in the CMK [10]. This makes further improving in the compressive strength of cemented soil. When the CMK content increases to a certain level, the Ca(OH)2 produced by the early cement hydration has completely reacted with the CMK. At this time, the CMK is not involved in the reaction and production of hydration products to provide strength [11]. However, the role of cement hydration to provide strength is weakened due to the reduction of the cement content. Finally, the compressive strength of cemented soil begins to decrease.

Fig. 1.
figure 1

Compressive strength of cemented soil at all ages.

Full size image

3.2 Analysis of XRD Phase Results

Figure 2 shows the XRD phase analysis of cemented soil with five contents of CMK at the age of 28 d. It can be seen from the Fig. That the main phase components of cemented soil were quartz, calcium aluminate hydrate (CAH), calcium hydroxide (CH), calcium aluminosilicate hydrate (CASH), calcium silicate hydrate (CSH), ettringite (AFt), anorthite (CAS) and other compounds. According to the hydration reaction equation of cement and CMK, it can be basically determined that CAH, CH, CASH, CSH and Aft were the main hydration products. Specifically, during the cement hydration process, tricalcium silicate (C3S), dicalcium silicate (C2S), tricalcium aluminate (C3A), and tetracalcium aluminoferrite (C4AF) reacted to form CSH, CH, tricalcium aluminate hydrate (C3AH6), calcium ferric aluminate hydrate (CFH), etc. The main reaction formulas are as follows [12]:

$$ C_3 S + 2H \to CSH + CH $$
(1)
$$ C_2 S + 2H \to CSH + CH $$
(2)
$$ C_3 A + 6H \to C_3 AH_6 $$
(3)
$$ C_4 AF + 7H \to C_3 AH_6 + CFH $$
(4)
$$ 3CS + C_3 A + 32H \to C_6 AS_3 H_{32} $$
(5)

When the CMK was added, it can accelerate the hydration effect and pozzolanic effect of cement. It is mainly that the active substances (SiO2 and Al2O3) in the CMK were activated by the alkaline environment where CH was generated by the cement hydration. At this time, the interface between the cemented soil slurry and the aggregate would accumulate a large amount of CH and become a weak link. CMK had a large number of disconnected chemical bonds, which would rapidly undergo a secondary hydration reaction with CH to generate calcium silicate hydrate. It makes the content of CH reduced and the orientation of CH changed. According to the content ratio of CMK to CH in the hydration process of cement, the hydration products were mainly CSH, tetracalcium aluminate hydrate (C4AH13), tricalcium aluminate hydrate (C3AH6) and anorthite aluminate hydrate (C2ASH8). The specific reactions are as follows [13]:

$$ {{AS_2 } / {CH}} = 0.5,AS_2 + 6CH + 9H \to C_4 AH_{13} + 2CSH $$
(6)
$$ {{AS_2 } / {CH}} = 0.6,AS_2 + 5CH + 3H \to C_3 AH_6 + 2CSH $$
(7)
$$ {{AS_2 } / {CH}} = 1.0,AS_2 + 3CH + 6H \to C_2 ASH_8 + CSH\; $$
(8)
Fig. 2.
figure 2

XRD phase analysis of 28d cemented soil.

Full size image

It can be seen from Fig. 4 that the diffraction peak intensities of CH and AFt are relatively higher in M0C15, and the diffraction peaks of CAH, CASH and CSH are relatively higher in M3C12. This is due to the high cement content in the cemented soil without mixing CMK, and the hydration reaction is concentrated in the cement reaction. When CMK is used to replace cement, the hydration reaction of CMK will consume CH to generate more CAH, CASH and CSH [14]. However, it does not mean that the more cement is replaced by coal measure metakaolin, the more CAH, CASH and CSH are generated. In this study, when the mass ratio of CMK to cement was 1:4, the hydration reaction was complete and the hydration products were the most in the cemented soil. A reasonable amount of cement and CMK can help promote to generate hydration products in the cemented soil. This is the reason why CMK can effectively improve the compressive strength of cemented soil.

3.3 Analysis of XRD Phase Results

Figure 3 shows the microstructure of cemented soil with five contents of CMK at the age of 28 d. As can be seen from the Fig., the main hydration products of cemented soil were a large number of flocculent CSH/CASH/CAH gels, acicular AFt and platy CH crystals. In Fig. 3 (a), the amount of CSH/CASH/CAH in cemented soil was relatively small, the gap between soil particles was large, the overall structure is loose. This is the reason why the strength of cemented soil without CMK is relatively low. In Fig. 3(b)–(c), the CSH/CASH/CAH gels were cross-connected with Aft crystals after CMK was added, which filled the pores with each other and wrapped the soil particles more densely. Especially, when the CMK content was 3%, the number of cemented soil pores was significantly lower than that of cemented soil without CMK. In Fig. 3(d)–(e), when the CMK content exceeded 3%, the CSH/CASH/CAH gels and Aft crystals began to decrease obviously, and the pores between soil particles increased. This indicates that the overall structural strength of cemented soil is reduced.

Fig. 3.
figure 3

Microstructure of cemented soil magnified 2000 times at 28d.

Full size image

According to the microstructure analysis, it can be explained that CMK has a filling effect in the cemented soil, mainly in two aspects. On the one hand, the particle size of CMK is about an order of magnitude smaller than that of cement (the CMK selected is 6250 mesh). The space between cement particles and soil particles is filled by the CMK particles. On the other hand, the pores between cemented soil frameworks are filled by the main hydration product of CMK and cement (CSH/CASH/CAH gels, AFt crystals) [15]. The pore structure is optimized to reduce the internal porosity. The compactness of the structure is increased to promote strength enhancement.

3.4 Prediction of Long-Term Compressive Strength

The relationship between the compressive strength of cemented soil and the age is shown in Fig. 4. When other conditions were the same, the compressive strength increased with age, but the rate of increase decreased with age.

Fig. 4.
figure 4

Relationship between compressive strength and age.

Full size image

The results show a general linear relationship between the compressive strength of cemented soil and age under the same conditions [16]. For the different CMK contents, the prediction formula of the cemented soil compressive strength and age is established as follows:

$$ q_{u,t} = A_t \cdot q_{u,t_0 } $$
(9)

where qu, t0 is the known compressive strength of cemented soil at age t0, qu, t is the prediction compressive strength of cemented soil at age t, At is the prediction coefficient.

Based on the data results, the compressive strength of different ages was regarded as the known compressive strength qu, t0. The relationship between the compressive strength of each age and the known compressive strength was analyzed. Figure 5 shows the relationship between the compressive strength at different ages and the compressive strength at 7d (t0 = 7d), 14d (t0 = 14d) respectively. The linear relationship between the compressive strength of each age was linearly fitted and the linear slope of each age was obtained. The corresponding slope was the prediction coefficient At, and the fitting correlation coefficient was Rt. The prediction coefficients At and the fitting correlation coefficient Rt is listed in Table 4.

It can be seen from Table 4 that the relationship between qu, t and qu, t0 was determined mainly by the prediction coefficient At. Figure 6 shows that the variation of the prediction coefficient At and the age t can be fitted by the power function.

$$ A_t = at^b $$
(10)

where a, b is the fitting parameters, R is the corresponding fitting correlation coefficient. For different ages t0, the fitting parameters are shown in Table 5, and b is a fixed value (b = 0.395).

Fig. 5.
figure 5

Relationship of cemented soil between qu, t and qu, t0.

Full size image
Table 4. Prediction coefficient At and fitting correlation coefficient Rt.
Full size table
Fig. 6.
figure 6

Relationship between At and t.

Full size image

The relationship between the fitted parameter a and the age t0 is shown in Fig. 7, and the obtained equation is:

$$ a = t_0^{0.135} $$
(11)
Table 5. Fitting parameters a, b of different t0.
Full size table
Fig. 7.
figure 7

Relationship between a and t0.

Full size image

Equation 11 is put into Eq. 10:

$$ A_t = t_0^{0.135} \cdot t^{ - 0.135} $$
(12)

When the specific age t0 is known, the cemented soil compressive strength qu, t can be estimated by the compressive strength qu, t0, the equation is as follows:

$$ q_{u,t} = t_0^{0.135} \cdot t^{ - 0.135} \cdot q_{u,t_0 } $$
(13)
Table 6. Predicted value and measured value for cemented soil qu,60, qu,90.
Full size table

In actual engineering, the compressive strength of cemented soil is generally selected at the age of 28d to predict the compressive strength of 60d and 90d. In this paper, qu,28 was used as the known compressive strength, and qu,60, qu,90 were calculated by Eq. 14. The predicted values of 60d and 90d were compared with the measured values, the specific results are shown in Table 6. It is found that the predicted values fall within 6% of the deviation of the measured values. This proved that the formula proposed to predict the later compressive strength of cemented soil is reasonable in this paper.

$$ q_{u,t} = 1.568t^{ - 0.135} \cdot q_{u,t_0 } $$
(14)

4 Conclusions

In this paper, the influence of the CMK content and age on the compressive strength of cemented soil was investigated. The solidification mechanism of the cemented soil was explored by microscopic testing technology. The main conclusions can be drawn.

  1. (1)

    The compressive strength of cemented soil can be significantly improved by using part of CMK instead of cement. When the total amount of cementitious materials was 15% and the CMK content was 3%, the compressive strength of the cemented soil reached the maximum.

  2. (2)

    CMK played a role in accelerating the cement hydration effect, pozzolanic effect and filling effect, which reacted with CH generated by cement hydration to generate CAH, CASH and CSH gels. These hydration products and cement hydration products enhanced the compactness of soil particles through bonding, filling, and compaction, forming a dense whole, thereby improving the strength of cemented soil.

  3. (3)

    The CMK and sandy soil used to study were in Shanxi Province, China. The two raw materials had obvious regional characteristics. The engineering properties and material compositions of raw materials in different regions were different. Based on this, a statistical formula between compressive strength and age of cemented soil (\(q_{u,t} = t_0^{0.135} \cdot t^{ - 0.135} \cdot q_{u,t_0 }\)) was proposed in this paper.