Effects of Granular Gradation on the Compressibility and Permeability of Lime-Stabilized Slurry with High Water Content

Abstract

Lime stabilization is one of the main methods to achieve efficient treatment and resource utilization of waste slurry. This study investigated the compressibility and permeability of lime-stabilized slurry with different granular gradations based on the ultra-low stress consolidation/permeability test and identified the stabilization mechanism of lime-stabilized slurry with high water content by mercury intrusion porosimetry (MIP) and scanning electron microscopy (SEM) tests. The test results indicated that (i) lime-stabilized slurry with high water content showed obvious evidence of remolded yield stress and (ii) the remolded yield stress increased with the increasing lime dosage. Lime was found to induce the flocculation of clay to form aggregates through ion exchange, further stabilizing them through the volcanic ash reaction, thus increasing the remolded yield strength of the stabilized slurry. The remolded yield stress of the Hangzhou stabilized slurry with a 1% lime dosage was shown to increase from 0 kPa to 5.71 kPa, while the compression index C_S1 decreased by 68.8%. In addition, the pore volumes and diameters of the soil increased once the flocculation was completed, leading to increased permeability of the stabilized slurry. It was, however, observed that the stabilized slurry permeability did not increase infinitely with the increasing lime dosage, but on the contrary decreased once the lime dosage exceeded a certain threshold value. The permeability of the Hangzhou stabilized slurry was found to be one order of magnitude higher than that of the remolded slurry at the optimal dosage. Whereas for slurry with high clay content, the recommended lime dosage was established to be 2% to reduce its compressibility or enhance its permeability; for slurry with high silt content, the recommended lime dosage was ascertained to be 3%.

1. Introduction

With the accelerating urbanization of China over recent years, the amount of waste slurry generated from projects such as port channel dredging, municipal dredging, and pile foundation construction has increased dramatically [1,2]. According to statistics, hundreds of millions of cubic meters of waste slurry are produced in China annually, the extremely high water content of which presents significant challenges in terms of achieving solid/liquid separation. Notwithstanding this difficulty, waste slurry would severely contaminate the surrounding environment if discharged untreated [3]. Therefore, effective slurry treatment measures are critically needed [4,5], with chemical flocculation currently ranking among the main methods to achieve efficient treatment and recycling of waste slurry [6]. Adding flocculants to the slurry modifies the surface properties of the suspended clay, which induces particle flocculation into flocs, thus accelerating sedimentation [7].

Currently, lime-based inorganic flocculants are often utilized in engineering to treat waste slurry [8,9,10]. A series of physicochemical reactions, including ion exchange reactions, volcanic ash reactions, and carbonation reactions, occur once lime is mixed with slurry [11,12,13]. Significant differences were observed in some studies in both the compressibility and permeability of the slurry before and after lime-based treatment and its dual effect, which were found directly to affect the subsequent dewatering process [14,15]. In recent years, various researchers studied the compressibility and permeability of lime-stabilized soils. Sudhakar and Shivananda [16] used lime to stabilize marine clay on the Madras coast of India, and their test results showed that lime could effectively reduce the compressibility of clay, reducing its coefficient to between only half to one-third of that of untreated soil samples. Wang [17] compared and analyzed the compressibility of different granular gradations of the slurry with a 2% lime dosage. His results established that the clay particle content significantly affected the compressibility of the soil samples after lime stabilization. Nagih et al. [18] preliminarily explored the effects of water content and lime dosage on the permeability of soil samples. Locat et al. [19] investigated the permeability of lime-stabilized soil samples with initial water contents of 122% and 650% by consolidation/permeability tests. It was shown that the optimal lime dosage for soil samples with both initial water contents was 3%. Quang and Chai [20] used a flexible wall permeameter to explore the permeability of lime-stabilized dredged slurry with an initial water content ranging from 110% to 160%. They established that the permeability of lime-stabilized soil decreased once the lime dosage exceeded 4% and found that the incidence in pores smaller than 1 μm increased significantly after lime treatment. Cuisinier et al. [21] analyzed the soil pore change trend before and after lime stabilization by comparing pore sizes with mercury intrusion porosimetry (MIP). Their test results showed that a large number of new pores smaller than 0.3 μm had formed in lime-stabilized soil, which were not present in the remolded soil samples. Al-Mukhtar et al. [22] examined the structure of lime-stabilized flocs by scanning electron microscopy (SEM) and found a significant increase in the intergranular accumulation layer. The mechanism of the effects of lime on the compressibility and permeability of a high water content slurry has not been elucidated in existing studies, leading to engineering problems such as wide variations in stabilized slurry performance in different areas and difficulties in determining the optimal content of curing agent.

In light of the above, the present study investigated the compressibility and permeability of lime-stabilized slurry with different granular gradations. Four types of slurry with high water content in the southeastern coastal area of China were selected as the research subjects, and the variation law of compressibility and permeability of the slurry with different lime dosages and granular gradations was analyzed with an ultra-low stress consolidation/permeability tester. Based on the MIP and SEM test results, the stabilization mechanism of lime-stabilized slurry with high water content was revealed at the microscopic scale.

2. Materials and Methods

2.1. Test Materials

The four types of slurry used in the test were collected from (i) the reclaimed area of Wenzhou Oufei Shoal, (ii) the site of the hydraulic fill foundation treatment works in Taizhou Agglomeration Area, (iii) the foundation pit project in Zhoushan Dinghai District, and a pile foundation construction site in Hangzhou Yuhang District. Their basic physical indices are shown in

Table 1. Basic physical properties of the four types of soil samples.

Sampling Site		Wenzhou	Taizhou	Zhoushan	Hangzhou
Initial water content w0 (%)		104	122	94	90
Liquid limit wL (%)		56	40	31	38
Plastic limit wP (%)		32	27	18	27
Relative density GS		2.55	2.63	2.72	2.62
Plastic index IP		24	13	15	11
Granular gradation	<0.002 mm	2	2.5	23.1	25.9
	0.002–0.075 mm	54.7	66.1	32.5	60.7
	0.075–0.425 mm	43.2	25.3	41.8	13.4
	0.425–0.2 mm	0.1	6.1	2.6	0

below. Among these, the Wenzhou and Taizhou soil samples were classed in the category of high liquid limit silt, each containing over 50% of silt and some fine sand, whereas the Hangzhou and Zhoushan soil samples belonged to the silty clay category, each comprising in excess of 20% of clay. The clay content of the four soil samples was, in descending order, Hangzhou slurry, Zhoushan slurry, Taizhou slurry, and Wenzhou slurry.

The lime used in the test was laboratory-grade slaked lime produced by Sinopharm, China, which was a uniform crystalline powder with a calcium hydroxide content of over 95%.

2.2. Test Protocols and Devices

In this study, 48 groups of one-dimensional consolidation/permeability tests were established to clarify the effects of granular gradation and lime dosage on the compressibility and permeability of lime-stabilized slurry with high water content, as shown in

Table 2. Proportion scheme in the test.

Soil Type	Initial Water Content w0 (%)	Liquid Limit wL (%)	Lime Dosage (%)
Wenzhou	168	56	0, 1, 2, and 3
Hangzhou	114	38	0, 1, 2, and 3
Taizhou	120	40	0, 1, 2, and 3
Zhoushan	93	31	0, 1, 2, and 3

below. The soil sample had a diameter of 61.8 mm (area of 30 cm²) and a height of 40 mm. Due to the extremely high initial water content of the test soil samples, if a conventional one-dimensional consolidation instrument with an initial loading stress of 12.5 kPa had been used, the loading would easily have led to the destruction of the soil sample and affected the test accuracy. Therefore, an ultra-low stress consolidation/permeability tester with a minimum consolidation pressure of 1 kPa was used for this test, as shown in Figure 1 below [23]. The test loading sequence was as follows: 1 kPa, 2 kPa, 3 kPa, 5 kPa, 7 kPa, 9 kPa, 12.5 kPa, 25 kPa, 50 kPa, 100 kPa, 200 kPa, 400 kPa, 800 kPa, and 1600 kPa. After the completion of each loading level and stabilization for over 24 h, the specimens were considered stable once they were settling by no more than 0.01 mm per hour, and the settlement of the soil sample was measured.

The permeability test was carried out after each level of loading was completed [24]. The water head height was strictly controlled throughout the permeability test. When the vertical effective stress was below 12.5 kPa, the initial water head height was 500 mm. When the vertical effective stress was greater than 12.5 kPa but less than 200 kPa, it was 1000 mm. When the vertical effective stress was greater than 200 kPa, it was 2000 mm. During the test, the change of water head with time was measured at a certain time interval, and the permeability coefficient was calculated based on this.

After consolidation/permeability tests, the compacted soil samples were made into the size required for MIP tests and SEM tests. The sample size of the MIP test and SEM test were 10 mm $\times$ 5 mm $\times$ 5 mm and 10 mm $\times$ 10 mm $\times$ 20 mm (length $\times$ width $\times$ height), respectively. All soil samples were immersed in liquid nitrogen for 10 min and freeze-dried for 24 h [25]. For the MIP test, the soil samples needed vacuum treatment before injecting mercury. The instrument settings in this MIP test are as follows: the surface tension of mercury was 485 dynes/cm, the contact angle was 130°, and the maximum pressure was 60,000 psia. For the SEM test, the soil samples were observed by a scanning electron microscope after spray-gold treatment.

3. Results and Analysis

3.1. Compressibility of Lime-Stabilized Slurry

The log(1 + e)–log p curve for the lime-stabilized slurry is shown in Figure 2 below [26]. The p is the applied vertical effective stress and the e is the void ratio. The compression curve of the remolded slurry was a straight line, while the compression curve of the lime-stabilized slurry was represented by two straight lines with different inclines. The horizontal coordinate of the intersection of the two lines marked the remolded yield stress of the soil sample [27]. Figure 3 below shows the compression indices of each group of lime-stabilized slurry at different dosage levels, where C_S1 and C_S2 are the compression indices of the stabilized slurry before and after yielding, respectively. The test results showed that increasing the lime dosage causes the C_S1 of the stabilized slurry to decrease and the C_S2 to increase. Taking the Wenzhou slurry as an example, the C_S1 of the stabilized slurry at a 1% lime dosage was only 32.3% of that of the remolded slurry, and C_S1 was found to decrease gradually with the increase in lime dosage. When the vertical effective stress exceeded its remolded yield stress, the compression index of the stabilized slurry was higher than that of the remolded slurry, and the compression index increased gradually with the increase in lime dosage.

Meanwhile, the granular gradation was observed to significantly affect the compression index of the lime-stabilized slurry. With the increase in clay content, the C_S1 and C_S2 of the stabilized slurry were noticed to gradually reduce. Consequently, the effect of lime stabilization on the slurry with high clay content was more significant than that on the slurry with high silt content.

As shown in Figure 4, the presence of significant remolded yield stress in the-lime-stabilized slurry was interpreted to indicate that lime stabilization could enhance the structural properties of the slurry with high water content. Meanwhile, the remolded yield stress of stabilized slurry was observed to increase with the increase in lime dosage, and the compressibility of stabilized slurry with a 3% lime dosage was shown to be the lowest.

When comparing the remolded yield stress values of the slurry with different gradations, the effect of granular gradation on yield stress is similar to the compression index. Taking the Hangzhou slurry and Zhoushan slurry as examples, the remolded yield stress of the Hangzhou slurry was substantially greater than that of the Zhoushan slurry with a 1% lime dosage. With the further increase in lime dosage to 2% and 3%, whereas the remolded yield stress of the Hangzhou slurry did not change significantly, that of the Zhoushan slurry still maintained an increasing trend. Therefore, the clay content directly influenced the remolded yield stress value of stabilized slurry, and the Hangzhou slurry formed a stable structure with a 2% lime dosage; thus, the further increase in lime dosage did not significantly reduce its compressibility. Therefore, more significant changes were noted in the compressibility of the slurry with high clay content than that of the slurry with high silt content after lime stabilization.

3.2. Permeability of Lime-Stabilized Slurry

Figure 5 below shows the slurry log k–log p curves for each group with different lime dosages. It can be observed that although lime stabilization effectively improved the permeability coefficient of the slurry with high water content, this permeability coefficient did not increase indefinitely with the increase in lime dosage. Instead, the permeability of the stabilized slurry was found to decrease when the lime dosage exceeded a certain threshold value. Taking the Hangzhou slurry and the Wenzhou slurry as examples, while the permeability coefficient of the Hangzhou slurry reached its peak with 2% lime dosage, it decreased significantly after the lime dosage was further increased; in contrast, the permeability coefficient of the Wenzhou slurry consistently increased along with the increase in lime dosage within the range of 1% to 3%.

The granular gradation was furthermore observed to affect the permeability coefficient of the lime-stabilized slurry. The permeability coefficient of the stabilized slurry was more substantially enhanced with the increase in clay content. Under a vertical effective stress of 100 kPa, the permeability coefficients of the Hangzhou and Wenzhou slurries rose by 178.0% and 55.1%, respectively, with a 1% lime dosage. It can thus be seen that clay proved more sensitive to lime than silt and that the slurry with high clay content achieved a greater improvement in permeability after lime stabilization than that with silt.

3.3. Microstructure of Lime-Stabilized Slurry

3.3.1. MIP Test Results

Figure 6 and Figure 7 below show the cumulative mercury injection curve and the pore volume distribution curves of the stabilized slurry with different lime dosages, respectively. The pore distribution of the soil samples in each test group showed an obvious “unimodal” structure, and the pore diameters were concentrated between 0.1 μm and 1 μm. After lime stabilization, both the total pore volume of the slurry and the pore diameters corresponding to the peak of the pore volume distribution curve increased. In the case of the Hangzhou stabilized slurry, both pore volume and peak pore diameter increased with the increase in lime dosage, reaching a peak with a 2% lime dosage, and both decreased when the lime dosage was further augmented. Based on earlier consolidation/permeability test results, it was established that the law of change in pore volume and pore diameter was consistent with that of the permeability coefficient of cured silt. As the seepage flowed, the pore volume was observed to directly influence the permeability of the soil samples. When the lime dosage was lower than the optimal dosage, the pore volume and pore diameter of the stabilized slurry increased, as did the permeability of the soil. When the lime dosage exceeded the optimal dosage, the pore volume and pore diameter of the stabilized slurry was found to decrease, which was also believed to explain the decrease in permeability of the Hangzhou stabilized slurry when the lime dosage was further increased in the macroscopic test.

By comparing the pore volume distribution curves of each group of stabilized slurry with the same lime dosage, the pore structure of the slurry with high clay content changed more significantly than that of the silt slurry. When the lime dosage was 1%, the pore volume of the Hangzhou stabilized slurry showed the greatest tendency to increase compared to the remolded slurry, reaching 31.2%, while that of the Wenzhou, Taizhou, and Zhoushan stabilized slurry only increased by 12.3%, 11.2%, and 12.0%, respectively. In addition, the pore diameter corresponding to the peak of the pore volume distribution curve of the Hangzhou stabilized slurry reached 0.43 μm, which was also greater than the pore diameters of the other three slurry types at the same point. In summary, firstly, the clay content proved to be affected by pore volume and pore diameter after lime stabilization, a finding that was consistent with the results of the macroscopic consolidation/permeability tests, and secondly, the best permeability was achieved by treating the Hangzhou soil samples with a 1% lime dosage.

To analyze the variation pattern of pore structures of the lime-stabilized slurry, the pores were classified into four categories: (i) micropores (pore equivalent diameter ≤ 0.01 μm), (ii) pinholes (0.01 μm < pore equivalent diameter ≤ 0.1 μm), (iii) mesopores (0.1 μm < pore equivalent diameter ≤ 1 μm), and (iv) macropores (pore equivalent diameter > 1 μm), as shown in Figure 8 below. The pores in the slurry of each test group mainly consisted of pinholes and mesopores, with a total combined percentage exceeding 80%. After lime stabilization, the percentages of micropores, pinholes, and macropores in the slurry of each test group were found to decrease, while the percentage of mesopores increased and reached its peak with the optimal lime dosage. Taking the Zhoushan slurry and Hangzhou slurry as examples, the percentage of pores in the Zhoushan slurry increased with the increase in lime dosage and reached its peak at a 3% lime dosage, while the percentage of pores in the Hangzhou slurry reached its peak at a 2% lime dosage. Whereas the percentage of micropores and pinholes was observed to increase, that of mesopores decreased when the lime dosage was further increased. The reason was hypothesized to be that the excessive filling of the pores with cementing materials reduced the diameter of the pore throat.

3.3.2. SEM Test Results

Figure 9 below shows the SEM images of the Zhoushan slurry before and after lime stabilization. The remolded slurry soil samples were mainly flaky and granular, and the contact mode between the particles was face-to-face contact, forming a relatively dense soil skeleton. Whereas the pore distribution was dominated by intergranular pores which were mostly round or elliptical in shape and small in diameter, the microstructure of the soil samples had significantly changed after lime stabilization, when the contact mode between particles had become edge-to-face and edge-to-edge contact. The loose soil skeleton was found to have led to an increase in the number of overhead pores which were irregular in shape and relatively large in diameter. With the increase in lime dosage, the particles developed from local edge-to-face contact to overall edge-to-face contact, with a gradual rise in the number of overhead pores.

The SEM images of the four types of stabilized slurry with a 1% lime dosage are shown in Figure 10 below. The number of clay mineral aggregates and pore diameters after lime stabilization of the slurry with high clay content was larger than those of the slurry with high silt content. This was hypothesized to indicate, firstly, that the stabilization had a more significant effect on the slurry with high clay content than that with high silt content for the same lime dosage. Secondly, the pore diameters had obviously increased, which manifested macroscopically as noticeably enhanced permeability.

4. Discussion

After lime stabilization, the slurry with high water content revealed remolded yield stress, with significantly reduced soil compressibility prior to yielding. It has been shown that the remolded yield stress of soils is closely related to their structural properties [28]. In this study, the mixing of lime and slurry induced the flocculation of fine particles into flocs through ion exchange, forming a relatively stable soil structure. Meanwhile, the volcanic ash reaction produced various cementing substances, such as calcium silicate hydrate (CSH) and calcium aluminate hydrate (CAH), which further bound the flocculated structure. Under the combined action of ion exchange and volcanic ash reaction, the compressibility of the stabilized slurry was shown to substantially decrease. When the vertical effective stress exceeded the remolded yield stress, the structure of the stabilized slurry was destroyed, and its compressibility was close to that of the remolded slurry with a sharp rise in compressibility, which was consistent with a previous study [24].

Due to the large specific surface area of clay, ion exchange reactions occurred easily, resulting in a greater stabilizing effect on the slurry with high clay content with a 1% lime dosage [29]. Meanwhile, clay proved more sensitive to the lime dosage than silt. When the optimal lime dosage was exceeded, the lime that did not participate in the reaction instead reduced the bonding force between soil particles, thus leaving little difference between the respective remolded yield stress values of the slurry with high clay content with 2% and 3% lime dosages [30].

Lime was also found to enhance the permeability of the slurry with high water content, due primarily to the increase in pore volume and diameter. After lime stabilization, the percentage of mesopores was observed to have risen while the percentages of micropores, pinholes, and macropores had all reduced. Song et al. [31] proposed that the percentage of mesopores should be regarded as the key factor in determining the permeability of the lime-stabilized slurry, and that although its variation trend is consistent with that of the permeability coefficient of specimens, micropores and pinholes have virtually no effect on the soil sample permeability due to the small pore throat size. In the existing studies, it was usually assumed that the reduction in mesopores and macropores in lime-stabilized soils was associated with a volcanic ash reaction, while the drop in micropores and pinholes was related to both ion exchange and a volcanic ash reaction [32,33]. Further studies have established that the combined effects of ion exchange and volcanic ash reactions flocculate soil particles to form aggregates, resulting in reduced percentages of micropores and pinholes. Conversely, the cementation products generated by the volcanic ash reaction fill the pores, leading to smaller pore diameters. The above two reactions have furthermore been shown to take place simultaneously, and their primary and secondary effects should be deemed to relate to the lime dosage.

The recommended lime dosages for actual engineering projects have been set to (i) 2% for slurry with high clay content in order to reduce its compressibility or enhance its permeability and (ii) 3% for slurry with high silt content.

5. Conclusions

This study investigated the compressibility and permeability of lime-stabilized slurry with different granular gradations based on the ultra-low stress consolidation/permeability tests and revealed the stabilization mechanism of lime-stabilized slurry with high water content by MIP and SEM testing. The main conclusions obtained were as follows:

(1)

Lime-stabilized slurry with high water content showed obvious evidence of remolded yield stress, which increased with the increase in lime dosage. Meanwhile, it was established that lime increased the permeability of the slurry with high water content, but that its permeability did not rise infinitely with the increase in lime dosage, while the permeability of the stabilized slurry decreased once the lime dosage exceeded a certain threshold value.

(2)

The granular gradation affected the lime stabilization impact on the slurry with high water content, where clay proved more sensitive to lime stabilization than silt. When the lime dosage was 1%, the remolded yield stress of the stabilized slurry gradually rose with increased clay content, while its compressibility gradually decreased and permeability gradually improved.

(3)

The mixing of lime and slurry induced the flocculation of fine particles into flocs through ion exchange, forming a relatively stable soil structure and generating further bonding under the volcanic ash reaction, thus augmenting the remolded yield stress of the stabilized slurry. The pore volume and diameters also increased as part of this process, thus enhancing the permeability of the stabilized slurry.

(4)

For the slurry with high clay content, the recommended lime dosage to reduce its compressibility or enhance its permeability was 2%, whereas for the slurry with high silt content, it was 3%.