Keywords

1 Introduction

Previous studies revealed that changing of pore water chemistry (such as cation concentration, cation valence and dielectric constant etc.) has significant influences on physical and mechanical behaviour of clays, especially on swelling clays (Yukselen-Aksoy et al. 2008; Abdullah et al. 1997; Bolt 1956; Di Maio et al. 2004; Kaya and Fang 2000; Oren and Kaya 2003). Bolt (1956) firstly stated that the compressibility of pure clay suspensions could be accounted well by considering the interaction between double electric layers around clay particles. Di Maio et al. (2004) presented that an increase in pore solution concentration may cause a reduction in compressibility of compacted bentonite. Gajo and Maines (2007) investigated the mechanical behaviour of active clays affected by pore fluid acidity and alkalinity. Zhang et al. (2016) and Ye et al. (2017) systematically investigated the influences of infiltrating solutions on volume change behaviour and hydraulic conductivity of highly compacted GMZ bentonite. However, the salinity and clay mineral effect on their consolidation behaviour of pure clays is still few concerned.

This paper focuses on the pore water salinity and clay mineral effect on the consolidation behaviour and micro-structure of four soils at high initial water contents (0.6–1.5 times liquid limits). To simulate this variety of mineral composition and porewater salinity, a mixture of commercial kaolin, bentonite and NaCl solution was employed in this investigation. The salinity effect on the compression index Cc and the ending time of primary consolidation were examined. With the help of micro-structure observations, a reasonable mechanism of salinity effect on macro consolidation behaviour was suggested.

2 Experimental Programs

2.1 Materials

To prepare specimens in laboratory, commercial kaolin and bentonite from Jiangsu province in China, were mixed with different proportions based on the dry weight ratio. It should be noted that the composition of mixtures was selected based on the typical marine clay deposited in Lianyungang area in Jiangsu Province. The basic physical properties of materials are shown in Table 1. Liquid limit (LL) and plastic limit (PL) was determined using Casagrande method according to British standard BS1377-2 (1990). Besides, the specific surface area (SSA) was determined using the Ethylene Glycol Monoethyl Ether (EGME) method (Cerato and Lutenegger 2002; Fan et al. 2014). By the X-ray diffraction analysis, the mineral compositions were presented in Table 2.

Table 1. Physical properties of the considered materials
Full size table
Table 2. Mineral composition of the kaolinite and bentonite
Full size table

2.2 Specimen Preparation

The pure kaolin or mixed clays were added with distilled water or sodium chloride solutions at a mass concentration of 1%, 3%, 5% and 10% (equal to 0.17 mol/L, 0.51 mol/L, 0.86 mol/L and 1.7 mol/L). For the artificial mixed clays, three mass ratios of kaolin to bentonite at 19/1, 9/1 and 4/1 were designed. A preset mass of dry powder and solutions were thoroughly stirred for at least 10 min using a paddle mixer. Thereafter, the slurries were cured in a sealed plastic chamber for a minimum of 24 h to homogenize the distribution of the pore liquid. The indexes of mixtures were summarized in Table 3. The selected water contents were set as 0.6 to 1.2 times of LL. Hereafter, for K5%B95% as an example, ‘B’ and ‘K’ in sample identification is shorted for ‘bentonite’ and ‘kaolin’ clay, and the value represents the percentage of bentonite and kaolin clay in the mixture.

Table 3. Initial state of samples
Full size table

2.3 Oedometer Test

A total of 24 specimens were prepared by mixing kaolin and bentonite at different proportions. These mixtures were divided into 6 groups: K, B5%K95%, B10%K90%, B20%K80%, B20%K80%Micro and KMicro. The first 4 groups were used to conduct oedometer tests, and each specimen was mixed with NaCl solution at concentrations 0%, 1%, 3%, 5% and 10%. The mixtures were carefully installed to the oedometer ring. Then, samples were first saturated by soaking with the equivalent concentrations of sodium chloride for 48 h. For the slurry samples (w0 > LL), vertical loads 1 kPa, 1.8 kPa, 3.2 kPa and 5.8 kPa were supplied step by step in order to avoid squeezing the soil from the ring (Hong et al. 2010). For sample B20%K80% (w0 < LL), an initial load 12.5 kPa was applied avoiding the swelling influence. Each load was kept for at least 24 h until the displacement satisfied the standard requirement (ASTM-D2435 2011). Additionally, the organic plastic oedometer cell was applied in order to keep the concentration in the cell and prevent the cell erosion.

2.4 Micro-structural Investigation

As mentioned above, the last 2 groups (B20%K80%Micro and KMicro) were used to perform SEM (Scanning electronic microscopy) tests, aiming to clarify the mechanism of salinity effect on the intrinsic compression behavior of soft clay. Samples were consolidated by stepwise applying the vertical load from 1 kPa to 100 kPa, and then removed from oedometer cells. Thereafter, all samples were carefully trimmed to appropriate sizes and then immersed in liquid nitrogen for instant freezing (Zhang et al. 2014). Finally, they were lyophilized and vacuum coated with a layer of gold to conduct SEM tests.

3 Results and Discussions

3.1 Salinity Effects on Compression Index Cc

As a common knowledge, the compression index Cc was defined by the slope of straight portion of compression curve. The Evolutions of compression curves, LL and Cc changing with concentrations of NaCl solution were presented in Fig. 1(a)–(b). It shows that the LL and Cc of B5%K95%, B10%K90% and B20%K80% all decrease with pore water salinity, whilst those of kaolinic clay are relatively stable. Note that the variation trend of LL shown in Fig. 1(b) is consistent with the results in previous publications (Abdullah et al. 1997; Di Maio et al. 2004; Kaya and Fang 2000; Oren and Kaya 2003; Sridharan and Rao 1975; Yukselen-Aksoy et al. 2008). Among them, Sridharan and Rao (1975) proposed a mechanism on the LL of the clay, i.e., for montmonrillonite, LL is predominantly controlled by the thickness of double diffusion layer. Thickness suppression of the double diffusion layer caused by increasing the salt concentration, leads to LL’s reduction. The similar variation trends of Fig. 1(a) and (b) indicate that the mechanism accounting for pore water salinity effect on compression behavior of clays is probably related to the suppression of double diffusion layer.

Fig. 1.
figure 1

Liquid limits LL and compression index Cc vs. NaCl molarity: (a) Compression curves; (b) Cc vs. NaCl molarity and LL vs. NaCl molarity

Full size image

3.2 Salinity Effects on Consolidation Behavior

The primary consolidation stage is defined in terms of the change in effective vertical stress. In the absence of pore water pressure measurements, the end of primary consolidation stage is defined using settlement observations. The elapsed time tp represents the time required for the completion of primary consolidation. The typical relationships between void ratio e and elapse time t of four clays are shown in Fig. 2(a)–(d). As illustrated in Fig. 2(a), the primary consolidation ended in 10 min for kaolinite mixed with distilled water and 10% NaCl solutions, and the ending point of primary consolidation tp were not significantly affected by pore water salinity and applied stresses. In contrary, the results as shown in Fig. 2(b)–(d) indicate that the period of primary consolidation behavior of clays containing montmorillonite are apparently influenced by water salinity. The tp of B5%K95%, B10%K90% and B20%K80% decreased by nearly an order of magnitude when the saline concentration increased to 10%. According to Terzaghi’s one-dimensional consolidation theory, the decreasing of tp represents a shorter drainage time and a higher consolidation coefficient. The permeability coefficient k is directly proportional to consolidation coefficient Cv, and it can be concluded that the permeability of pure clays containing montmorillonite increased with pore water salinity increasing. The mechanism attributed to the above findings was further discussed by micro-structure observations.

Fig. 2.
figure 2

The e-lgt curves of specimens: (a) K with 0% and 10% under 200 kPa; (b) B5%K95% with 0% and 10% under 400 kPa; (c) B10%K90% with 0% and 10% under 400 kPa; (d) B20%K80% with 0% and 10% under 100 kPa

Full size image

3.3 Microstructure and Mechanism

The salinity effect on microstructures was examined by binary SEM images (magnified 5000 times) as shown in Fig. 3. The fabric element of specimen B20%K80% Micro-01 hydrated with distilled water (Fig. 3a) exists a shape of thin platelet; whilst that of specimen B20%K80%Micro-02 salinized with 0.86 mol/L NaCl solution (Fig. 3b) presents a shape of bulky pellet. On the other hand, for specimens only composed of kaolin (Fig. 3c and d), the similar arrangement and shape of bulky pellet are identified whatever with or without porewater salinity. Hence flocculation mechanism could be introduced to interpret the salinity effect on normalized compression behavior for reconstituted clay, especially for that with rich montmorillonite.

Fig. 3.
figure 3

SEM photos (magnified 5000 times): (a) B20%K80%Micro-01; (b) B20%K80%Micro-02; (c) KMicro-01; (b) KMicro-02;

Full size image

4 Conclusions

To better understand the salinity and clay mineral effects on the consolidation behaviour of clays, oedometer tests and SEM tests were conducted on four artificial clays. The main conclusions can be drawn as belows:

  1. (1)

    It shows that the LL and Cc of pure clays containing bentonite all decreased with pore water salinity, whilst those of kaolinic clay are relatively stable.

  2. (2)

    The ending points of primary consolidation of kaolinite were not significantly affected by pore water salinity. In contrary, the period of primary consolidation for clays containing montmorillonite decreased with pore water salinity.

  3. (3)

    The SEM observation shows that the fabric element of B20%K80% with a shape of thin platelet tend to be flocculated when the water salinity increased, whilst the micro-structure of kaolinite were not influenced by the water salinity.