Introduction

Collapsible soils are located in various parts of the world, including large areas of arid and semi-arid countries. These soils are characterized by their low density, high porosity, low water content, low plasticity index, high sensitivity, low particle bond strength, low cohesion and high content of fine particles [1,2,3]. Collapse and large induced settlements at the saturation state damage the structures built on them. Several methods can be used to minimize or eliminate soil collapse, including chemical stabilization, injection, pre-wetting, controlled wetting, soil replacement, compaction, and heat treatment, which depend on the depth, the type of structure to be constructed, cost, and the practicality of each method [1, 2, 4,5,6,7].

Chemical stabilization is commonly used to improve the mechanical properties of problematic soils. This method modifies soil properties through flocculation, cation exchange, bonding of grains, and pozzolanic reaction [8]. Lime and cement are conventional additives, which, due to approximate familiarity with their mechanism in relation to various soil types, are the subject of attention by engineers. The production process of these additives has several economic and environmental problems that need to be addressed with more importance. These problems are the emission of Nitrogen oxide and carbon dioxide, the need for mining and the air pollution which have led researchers to use alternatives with less negative environmental effects and lower costs. For this purpose, other materials such as calcium carbide residue (CCR), fly ash, ground granulated blast furnace slag, cement kiln dust, rice husk ash, and silica fume have been used [5, 9,10,11,12].

CCR is a by-product of the acetylene gas production. It mainly contains calcium hydroxide, with a small amount of calcium carbonate, carbon and silicate which makes it a preferred alternative for use as a stabilizing agent [5, 6]. Considering the lack of sufficient information related to investigating the collapsibility of stabilized samples with CCR in technical texts on the one hand, and the apparent increase in the strength of samples stabilized with CCR on the other hand [13,14,15,16,17], the authors tried to focus mainly on parameters affecting the collapse index (Ie) and collapse potential (Ic) of stabilized samples with CCR.

AlShaba et al. [18] investigated the behavior of collapsible soil stabilized with iron powder. They found that the initial dry unit weight and additive contents are the most important factors for improving the collapsible soils and reducing the settlement. Ayeldeen et al. [19] enhanced the collapsible soil characteristics with fiber and cement. The results showed that the collapse potential decreased significantly to less than 1% due to the addition of fiber to the cement-stabilized soil. Similar results were reported by Ogila and Eldamarawy [20] and Shah et al. [21]. They stated a significant reduction in the collapse potential of soil with cement kiln dust. Cardoso et al. [22] used the biocementation method to stabilize collapsible soils. They concluded that the biocementation method in reducing the collapse potential of soil was similar to the hydraulic binders. Ayeldeen et al. [23] evaluated the mechanical behaviors of collapsible soil using two biopolymers including guar gum and xanthan gum. Results proved a noticeable reduction in the collapse potential of soil stabilized with 2% biopolymer from 9% to about 1%. Haeri and valishzadeh [24] assessed the effect of various nano materials to stabilize the collapsible soil. They found that the collapsibility decreased from moderately severe to moderate. Similar results were reported by Tabarsa et al. [25] who reported the significantly effect of nano clay in reducing the collapsibility and dispersivity behavior of the loess soil. The results showed that the mixture of field soil with 2% nano clay had the best result in terms of reducing erosion.

As mentioned, collapse and large induced settlements at the saturation state damage the structures built on them. Therefore, measuring the collapse potential of these soils is essential for safe engineering works. The present study attempts to investigate the behavior of collapsible soil stabilized with CCR as an industrial waste for construction material in the earthwork applications, namely, roads embankment, sub-base materials, etc. Given the enormous volume of this waste and the existing limitations, it can be regarded as an appropriate option for conventional additives (i.e., cement and lime). Hence, it is necessary to evaluate the potential of this type of waste in terms of Ie and Ic for various civil engineering works. For this purpose, seven different contents of CCR, five curing periods, three different water contents, and two relative compactions were used.

Materials and methods

Soil

The collapsible soil used in this research was taken from the site of Golestan Earth Dam, located in the north of Iran. This dam is situated in Gonbad Kavous city, the location of which is shown in Fig. 1. Soil samples were taken from a depth of about 2 m.

Fig. 1
figure 1

Location of Golestan Earth Dam

Full size image

To identify the characteristics of soil, particle-size analysis by sieving and hydrometric methods, Atterberg limits, specific gravity, and standard compaction tests were performed. All of these tests were performed twice to reduce the error caused by the operator. The grain-size distribution of soil is shown in Fig. 2. Table 1 shows the average results of tests.

Fig. 2
figure 2

Grain-size distribution of soil

Full size image
Table 1 Physical properties of soil
Full size table

Additive

The CCR used in this research was provided from Noor Zarandieh Acetylene Gas Company, located in Zarand city of Kerman province, Iran. This company produces acetylene gas and the product of acetylene gas is CCR, which is initially in the form of slurry. After a few days of drying in the environment, it changes to a dry form. To use CCR, it is first placed in an oven at the temperature of 105 °C for 24 h. It is then sieved by a sieve no # 40 (see Fig. 3a). Scanning electron microscope (SEM) image of CCR is shown in Fig. 3b. Table 2 presents physico-chemical properties of CCR.

Fig. 3
figure 3

a Dried and passed CCR from a sieve no # 40, b SEM image of CCR

Full size image
Table 2 Physico-chemical properties of CCR
Full size table

Sample preparation

Cylindrical samples with a diameter and height of 50 and 20 mm (Fig. 4), respectively, have been prepared in three layers and at their optimum water content (ωopt), 2% less and more than the ωopt (i.e., ωopt−2 and ωopt+2), and two relative compactions (i.e., 80 and 85%). Static compaction method used to prepare stabilized and unstabilized samples. All stabilized samples were cured at a controlled temperature (23 ± 2 ℃) and humidity (95 ± 5%) for the desired curing period. In the present study, at least two samples were investigated under a similar condition to ensure the reliability of the test results.

Fig. 4
figure 4

A prepared sample

Full size image

It should be noted that the reason for using the relative compactions of 80% and 85% is to simulate the field conditions. According to the research of Noorzad and Pakniat [26], it was observed that the average maximum dry unit weight (MDUW) of the studied soil is 18.46 kN/m3. Also, the average dry unit weight of undisturbed samples taken from this site is 14.92 kN/m3. Thus, the relative compaction of soil at this site is equal to 80.82%. In this study, relative compactions of 80 and 85% were used.

Method

To evaluate the effect of CCR and some factors such as curing period, water content during sample preparation, and relative compaction on the magnitude of one-dimensional collapse, 133 tests were performed on the collapsible soil of the Golestan dam site. In addition, 16 standard compaction tests were conducted on unstabilized and CCR-stabilized soil to investigate the effect of CCR on the compaction properties. Table 3 illustrates the experimental program used in the present study.

Table 3 Experimental program used in the present study
Full size table

ASTM D5333 was used to measure the Ie and Ic of the samples [27]. The sample is placed in a consolidation apparatus (see Fig. 5) with filter paper and porous disks. The loading increments for Ie are 12, 25, 50, 100, and 200 kPa and for Ic are 12, 25, 50, 100, 200, and 400 kPa. Each loading increment is applied to the sample at 1 h intervals. After reaching the final loading increment and after 1 h, the sample is submerged for 48 h and the deformation is recorded. Distilled water is used for the Ie while water of Golestan dam is used for Ic.

Fig. 5
figure 5

Consolidation apparatus used in the present study

Full size image

Ic is the percent-relative magnitude of soil collapse at any stress level. This parameter is determined as follows:

$$I_{c} = \frac{d_{f}-d_{i}}{h_{0}}{100}$$
(1)

where df = dial reading at the appropriate stress level after wetting (mm), di = dial reading at the appropriate stress level before wetting (mm), and ho = initial sample height (mm). It should be noted that the Ie is the percent-relative magnitude of collapse determined at 200 kPa and calculated using Eq. (1). In the present study, the classification presented in ASTM D5333 according to Table 4 was used.

Table 4 Classification of Ie based on ASTM D5333
Full size table

Results and discussion

Effects of CCR on compaction properties

The compaction curves of soil with different contents of CCR are presented in Fig. 6 and Table 5. As can be seen, by adding CCR to the soil, the ωopt increases and the MDUW decreases. The reason for the decrease in the MDUW of CCR-stabilized soils is the lower specific gravity of CCR compared to the soil (specific gravity for soil and CCR were determined as 2.70 and 2.25, respectively). In fact, CCR particles are lighter in weight than soil grains. In addition, the immediate reaction between soil and CCR leads to an increase in the void between particles and hence, the MDUW reduces by the addition of CCR. Moreover, a portion of compaction effort is used to overcome the weak bond created due to the deposition of cementitious materials at the clay-particle interfaces.

Fig. 6
figure 6

Compaction curves of stabilized soil with different CCR contents

Full size image
Table 5 Average compaction properties (i.e., MDUW and ωopt) of stabilized soil with different CCR contents (Two tests were done)
Full size table

The reason for the increase in ωopt with increasing CCR content is related to the self-cementing nature of CCR. CCR hydrates rapidly even at a very short time. The hydration process requires water. In the other side, addition of CCR to the soil reduce soil workability, therefore by increasing CCR content, more water is needed to reach the highest unit weight.

With increasing delay time, the MDUW decreases and the ωopt increases (see Table 5). In fact, the delay time between sample preparation and the compaction process at the constant compaction energy reduces sample strength. Therefore, to improve the properties of stabilized soil, it is necessary to compact the soil with the least delay time.

Effect of various factors on I e and I c

CCR contents

Figure 7 and Table 6 show the effect of CCR contents and the curing periods on the Ie and Ic. It was observed that CCR contents significantly reduced Ie and Ic even after 1 day of curing and changed the degree of collapse from moderately severe to slight and non-collapsible one. For 1 day of the curing period, the instant enhancement can be related to the cation exchange and/or agglomeration of the soil particles. Agglomeration is a process that the flocculated clay particles commence to create weak bonds due to the deposition of cementitious material at the clay-particle interfaces [8].

Fig. 7
figure 7

Ie versus CCR contents at 1 day, and 7, 14 and 28 days of curing period

Full size image
Table 6 Ie and Ic for samples prepared with ωopt and different CCR contents and curing periods (relative compaction = 80%)
Full size table

For further curing periods, the main reason is the formation of cementitious products due to chemical reactions between clay particles and CCR as a calcium-based stabilizer. These products fill the empty spaces between the particles of the collapsible soil and stick them together, thereby creating the soil matrix denser and stronger. Similar results were reported by previous researchers [28,29,30,31,32]. Hence, stabilized samples can significantly reduce the degree of collapse to the allowable limit when exposed to the moisture. More cementitious products are formed using higher CCR contents and curing periods. For 5% and more CCR contents, an increase of the CCR content alone plays a significant role in reducing the degree of collapse, and the role of other factors such as curing period, relative compaction, and water content during sample preparation is less prominent.

It can be seen that from Table 6, the Ic is higher than the Ie, which is due to its more stress during the test. Similar conclusions were observed for all stabilized and unstabilized samples. The maximum stress for the Ic is 400 kPa, while the maximum stress for the Ie is 200 kPa.

According to the results, no significant changes were observed for the stabilized samples with CCR contents of more than 5%. Hence, to investigate the effect of relative compaction and water content during sample preparation on Ie and Ic, the CCR content was limited to a maximum of 5%.

Relative compaction

Figure 8 and Table 7 show the effect of relative compaction on the Ie and Ic for unstabilized and stabilized samples with different CCR contents. It is clear that by increasing the relative compaction from 80 to 85%, the Ie and Ic decrease. For example, in unstabilized samples prepared at relative compactions of 80 and 85%, the Ie was measured as 7.57 and 4.08%, respectively (in this comparison: water content = ωopt). A similar trend was observed for stabilized samples (see Table 7). According to the previous results and the significant effect of CCR contents and curing periods on the degree of collapse, stabilized samples with relative compaction of 85% were investigated only in two curing periods (i.e., 1 day and 7 days).

Fig. 8
figure 8

Ie versus CCR contents for different relative compactions and at 1 day and 7 days of curing period (in this figure, 80–1 = relative compaction of 80 and curing period of 1 day)

Full size image
Table 7 Ie and Ic for samples prepared at relative compactions of 80 and 85% with different CCR contents and curing periods (water content during sample preparation = ωopt)
Full size table

Increasing the relative compaction reduces the pore space between the soil particles, leading the denser structure. The denser the soil, the lower the initial void ratios, hence, there is less collapse upon wetting. Moreover, the soil’s dense state would limit the relative contribution of metastable forces to soil structure support. In stabilized samples, cementitious materials can help to form this dense and stiff structure by filling the pore space and sticking them together. In this structure, degree of collapse reduces due to the decrease in porosity and increase in interparticle contact area between the particles. Moreover, with the increase in the relative compaction and the particles getting closer to each other, a better interaction between the particles and the additive is formed to produce higher cementitious materials.

Water content during sample preparation

Figure 9 and Table 8 show the effect of water contents during sample preparation on the Ie and Ic for unstabilized and stabilized samples with different CCR contents. It is clear that by increasing the water contents the Ie decrease. According to Fig. 9, the Ie of unstabilized samples at relative compaction of 80% for ωopt−2, ωopt, and ωopt+2 were measured as 10.48, 7.57 and 4.37%, respectively. A similar trend was observed for Ic and stabilized samples. According to the results and the significant effect of water contents during sample preparation on the degree of collapse, stabilized samples with water content of ωopt+2 were investigated only in one curing period (i.e., 1 day).

Fig. 9
figure 9

Ie versus CCR contents for different water contents during sample preparation (relative compaction = 80%, curing period = 1 day)

Full size image
Table 8 Ie and Ic for samples prepared with ωopt−2 and different CCR contents and curing periods (relative compaction = 80%)
Full size table

Present concepts of compacted clay structure explain a random open flocculated structure at ωopt−2 and a close-packed dispersed structure at ωopt+2. The collapse is thought to be caused by the open-flocculated structure’s metastable nature. From another point of view, the greater initial water content decreases the metastable forces, and what stays from such forces to be entirely decreased by wetting is much less, resulting in less collapse. Similar results were found in previous studies [33, 34]. The stabilized samples prepared with ωopt−2 have a higher degree of collapse than those samples with ωopt and ωopt+2. In addition to the reason described about unstabilized samples, it can be stated that the chemical reaction between the soil particles and the CCR is an exothermic one. For ωopt−2, the water content is insufficient to begin and continue the chemical reactions. Therefore, the production of cementitious products is reduced to create a stiffer structure, leading to an increase in the degree of collapse.

Comparing the results with previous studies

The efficiency of CCR was compared with different additives used for the collapsible soil and was listed in Table 9. As expected, all additives decreased the collapse potential of the soil, but the level of improvement was different for each additive. This difference is related to the chemical compositions and the amount of each additive. It should be noted that this comparison is only based on collapse potential and further investigations are needed to decide on a suitable additive.

Table 9 Comparison of the results of the present study with some studies in the technical literature
Full size table

Conclusions

In this study, Ie and Ic as well as compaction properties of CCR-stabilized collapsible soil were investigated. In addition, the effect of various factors such as additive contents, curing period, relative compaction, and water contents during sample preparation on the Ie and Ic were analyzed and presented. In this section, a summary of the obtained results is given:

  • By adding CCR to the soil, the ωopt increases and the MDUW decreases. The reason for the decrease in the MDUW of CCR-stabilized soils can be attributed to the lower specific gravity of CCR compared to the soil, the immediate reaction between soil and CCR, and the formation of weak bond due to deposition of cementitious materials at the clay-particle interfaces. The increase in ωopt with increasing CCR content is related to the self-cementing nature of the CCR.

  • CCR contents significantly reduced Ie and Ic due to the formation of cementitious products. These products fill the empty spaces between the particles of the collapsible soil and stick them together, thereby creating the soil matrix denser and stronger. Hence, stabilized samples can significantly reduce the degree of collapse to the allowable limit.

  • Increasing the relative compaction reduces the pore space between the soil particles, leading to a decrease in Ie and Ic. The soil’s dense state would limit the relative contribution of metastable forces to soil structure support. In this structure, degree of collapse reduces due to the decrease in porosity and increase in interparticle contact area between the particles.

  • One of the most important factors in the investigations related to the collapsibility of soils is the water content during sample preparation. The effect of this factor is the same in stabilized and unstabilized samples, but they are completely different in terms of mechanism. In unstabilized samples, the greater initial water content decreases the metastable forces, resulting in less collapse, but in stabilized samples, the greater initial water content is sufficient to begin and continue the chemical reactions, leading to a decrease in the degree of collapse.

The present study showed that soil stabilization with CCR is considered an efficient method to reduce the collapse potential of problematic soils. Therefore, due to the high volume of CCR and existing limitations such as limited landfills and environmental effects, CCR can be used as a desirable option instead of traditional additives. On the one hand, this approach solves the disposal problems of industrial wastes and the environmental effects by arresting their hazardous materials. On the other hand, by reducing the consumption of traditional additives, it significantly reduces the costs of civil engineering works.