Effect of Freeze–Thaw Cycles on Black Cotton Soil Reinforced with Coir and Hemp Fibres in Alkali-Activated Binder


Expansive black cotton soil (BCS) is predominantly impervious and undergoes swelling and shrinkage when exposed to moisture fluctuation. This results in heaving of soil, causing it to lose its mechanical strength. Use of traditional cementitious binders like lime and cement has a significant impact on the environment and contributes almost 7% of the global CO2 emissions. In the present study, an attempt was made to improve the geomechanical properties of BCS using envirosafe alkali-activated binder (AAB) with naturally available coir (CF) and hemp fibres (HF). The coir and hemp fibres were chemically treated to improve their durability. AAB was prepared by blending an alkali activator solution of sodium silicate and sodium hydroxide with class F fly ash at 0.4 water to solid ratio. This study also investigated the effectiveness of coir and hemp fibre reinforced AAB–BCS at different freeze–thaw cycles. The influence of varying dosages of fibres and freeze–thaw cycles in AAB-treated BCS showed a significant improvement in soil tensile strength and durability. The microstructural and geomechanical results of treated fibres showed higher serviceability and tensile resistance compared to the untreated ones. Furthermore, nonlinear regression equations were also proposed to relate experimental test results with model-predicted results in terms of unconfined compressive strength and indirect tensile strength.


Black cotton soils (BCS) exhibit low volumetric stability in response to moisture imbalance [1, 2]. This dual existence (swell/shrink) is due to the presence of a high concentration of montmorillonite and smectite in BCS [3, 4]. Lime and cement are the most commonly utilized stabilizing agents to improve the geoengineering behavior of expansive soils and to effectively reduce the problems associated with swelling and shrinkage [5, 6]. Nevertheless, the production of these conventional cementitious binders has a significant environmental effect leading to production of carbon dioxide (CO2), nitrous oxide (N2O), and methane (CH4). Globally, it contributes 7–8% of CO2 annually. It is also stated that an excessive amount of lime or cement in soil may contribute to the development of carbonation shrinkage cracks. This leads to the rapid generation of sulfate reactions, which makes it difficult to achieve a target strength without addition of compounds [7, 8]. The use of low-carbon binding agents in soil is an alternative approach for increasing soil strength. The complete or partial substitution of industrial by-products as filler materials like fly ash [9, 10], solid waste such as bagasse ash, volcanic ash, cement kiln dust, marble dust and ground granulated blast furnace slag [11, 12] along with addition of fibres and geotextile is becoming increasingly widespread [13, 14]. Use of industrial by-products in soil stabilization techniques solves the issue of disposal and saves related costs through transport to landfill sites as well as mitigate emissions [8, 15, 16]. In view of the above issues, a new formulation of alkali-activated binders (AAB) is proposed as an alternate binder for ground improvement. This binder not only decreases the need for cement-based binders, but also promotes the use of industrial by-products. In contrast, AAB has similar mechanical behavior to the hardened cement binder [17,18,19].

The use of alkaline binders and fly ash in soil structural science has gained a great deal of interest primarily due to its abundant usability and low cost. As fly ash is a waste material obtained from thermal power plants, the cost of manufacturing AAB is also lower than that of portland cement. Considering the cost of sodium hydroxide and sodium silicate, the cost of AAB is approximately $40–43/m3, while that of ordinary portland cement (OPC) is approximately $45–50/m3 and that of lime is approximately $55–65/m3, depending upon the purity of raw materials. AAB is found to serve as an auxiliary cement binder with a higher degree of serviceability and low ecological impact. A broad and growing body of literature examined the use of pozzolanic precursors dependent on AAB in various soils as alternate cementitious material to improve geomechanical properties under specific loading conditions [19, 20]. Extensive study findings showed that the use of AAB as a construction material has major geoenvironmental benefits, such as increasing soil subgrade serviceability with high volumetric stability, improving the work-life of the canal liner, the stability of the slope and the railway foundation. This also helped to reduce greenhouse gasses and harmful pollution emissions as AAB replaces conventional cement binders [21,22,23]. Geopolymerisation mechanism in expansive soil through aluminosilicate precursors helped to improve compressive shear strength. Still, it cannot improve tensile strength, which is a crucial issue in the summer season when this soil expected to have tensile cracking. The problem of soil tensile cracking (shrinkage cracks) can be effectively addressed with the use of geotextiles or discrete fibres.

During the last few decades, fibres and geotextiles gained a great deal of interest in geoengineering applications due to their mechanical and tensile crack resistance properties, cost, and easy availability [24, 25]. Artificial fibres stay underground for a very long time and may have a detrimental impact on the environment. However, with increasing concern for sustainable development, researchers are encouraged to investigate alternative forms of reinforcement than the mainstream synthetic fibres. The efficiency of natural fibres is influenced by the presence of natural moisture, which leads to a decrease in serviceability. To solve the problems of fibre decomposition in soil, several researchers showed an interest in enhancing the resistance of fibres by chemical treatment [2, 15, 26]. Numerous studies proved the effectiveness of natural fibres in soil with chemical additives and considered it a good earth reinforcement material [27,28,29,30]. The fibres do not prevent cracking, but act indirectly to control spread of cracking and strengthen post-cracking mechanical properties. Soil–fibre reinforcement is an ancient technique used to enhance the adhesion and erosion resistance properties. Moreover, the combined addition of fibre with envirosafe AAB can control the soil dispersivity potential effectively. The reinforcing of subgrade soil with short fibres tends to have the greatest potential to prolong the service life of the pavement or to reduce the sub-base or base thickness. The freeze–thaw cycle is a weathering mechanism that effectively influences geomechanical behavior. Extensive work is focused on the impact of freeze–thaw processes on soil stability with various cement binders in seasonal frost areas. Some investigators documented the effect of freeze–thaw on the physical–mechanical and dynamic properties of fine-grained soil [28,29,30,31]. Results of freeze–thaw cycles on UCS, plasticity, and soil durability treated with discrete fibres and jute showed that fibre reinforcement could minimize the effects of freeze–thaw cycles on soil [32, 33]. Investigations on the impact of cement, lime, fly ash, fibre, and crumb rubber on the stabilized pavement under freeze–thaw cycles showed that the shear strength, CBR, and durable soil modules improved after ten freeze–thaw cycles [34,35,36]. Effect of the freeze–thaw cycles (0, 1, 5, and 10) on cement-treated expansive soil showed an improvement in UCS and shrinkage ratios in the fifth freeze–thaw cycle [37]. This work differs from earlier studies in the sense that aluminosilicate precursor-based alkaline binder in BCS tends to reduce dual activity (swell/shrink) effectively. The reinforcement of fibres in AAB-treated BCS can transform the brittle nature of the soil towards ductile through the interfacial friction mechanism of the contact between the fibre surface and the clay matrix. Chemically processed coir and hemp fibres provide a dual purpose by improving soil durability and avoiding early biodegradability of the material, which ultimately enhances its applicability. Freeze–thaw process in both CF and HF reinforced AAB–BCS can simulate material behavioral changes and evaluate tensile cracking upon temperature variations.

However, very few studies were reported on the strength characterisation of BCS treated with CF and HF-strengthened AAB. The process would contribute to the use of waste products such as fly ash and naturally available fabrics such as coir and hemp, thereby proposing a safe and environmentally friendly approach. Comprehensive work was also required to compare geomechanical performance between CF- and HF-strengthened AAB–BCS by examining the effects of freeze–thaw on soil resilience. The primary objective of this analysis was to improve the shear and tensile strength properties of black cotton soil by reinforcing it with coir (CF) and hemp fibre (HF) in alkaline binders at various freeze–thaw phases. The research also proposed regression equation models for the estimation of unconfined compressive strength (UCS) and indirect tensile strength (ITS) of CF- and HF-strengthened AAB-treated BCS at various fibre content and freeze–thaw cycles. In addition, the microstructural study was also carried out before and after the treatment of fibres to capture the chemical reaction and the bonding process of the fibre-reinforced AAB with the soil.

Materials and Methods


The black cotton soil used in this study was collected from the Telangana district of Nalgonda in the southern part of India. Sampling was carried out at a depth of 30 cm to avoid vegetation or roots along with collected soil. The soil was oven dried before the lumps were broken into pieces. According to the unified soil classification system (USCS), the soil was classified as highly compressible clay (CH) and found to have 76% of fines content (< 75 µm). The soil was rich in silica content, which is a prime requirement for chemical modification. The engineering properties of BCS and fly ash (FA) were listed in Table 1.

Table 1 Engineering properties of raw materials

Alkali-Activated Binder (AAB)

AAB was produced by the combination of fly ash and activator solution by maintaining water to solid ratio (w/s) as 0.4 in the binder. Activator solution was prepared by mixing sodium silicate solution with crushed pellets of sodium hydroxide, at least 24 h before using it, to react with fly ash. The chemicals were procured from HYCHEM Industries. Sodium silicate contained 55.9% water, 29.4% SiO2, and 14.7% Na2O. The mass ratio of fly ash to sodium hydroxide to sodium silicate was maintained at 400:10.57:129.43 [15, 32]. Table 2 summarised the amount of AAB required for treating per cu. m of BCS with varying water to solids ratio. Class F fly ash was collected from Ramagundam National Thermal Power Corporation (NTPC). Particle size analysis of untreated BCS, AAB-treated BCS, and fly ash was shown in Fig. 1. The main oxide compositions present in BCS, fly ash, and fibres (both treated and untreated) were obtained from X-ray fluorescence analysis (XRF) using the PANalytical Epsilon-1 spectrometer, as shown in Table 3.

Table 2 Amount of AAB required treating per cu. m of BCS with varying w/s ratio
Fig. 1

Particles size distribution curves of BCS and fly ash

Table 3 Elemental analysis of raw materials


Locally available coir fibre (CF) and hemp fibre (HF) were used as reinforcement material of BCS in combination with AAB. A constant length of 25 mm was adopted for both the fibres [25]. The basic physicochemical and engineering properties of CF and HF were tabulated in Table 4 (datasheet supplied by the manufacturer).

Table 4 Properties of coir and hemp fibers

Treatment of Fibres

In the present study, both CF and HF were chemically treated with a sodium hydroxide solution to improve the sustainability of the fibre in soil. The stepwise process of treatment of fibres was represented in Fig. 2. The entire procedure took 28 days for completion. In the first step, both CF and HF were soaked in water for 7 days at room temperature by ensuring the removal of waxy substances and natural oil covering the fibre surface cell wall. In the second step, the soaked fibres were dried in ambient conditions and normal room temperature (25 ± 3 °C) for 7 days. Dried fibres were soaked in sodium hydroxide solution for 7 days and washed with water and kept for drying for 7 days under normal room temperature. The elemental compositions for both fibres before and after chemical treatment were also shown in Table 3. In the present paper, TCF and THF denoted treated coir fibre and treated hemp fibre, and UCF and UHF denote untreated coir fibre and untreated hemp fibre, respectively.

Fig. 2

Process of fiber treatment: a raw coir and hemp fiber soaked in water for 15 days, b CF and HF soaked in water for 7 days, c drying for 7 days and keeping in NaOH for 7 days and d TCF and TGH after drying

Experimental Methodology

Sample Preparation

Expansive soil was uniformly blended with fly ash-based AAB (5% as an optimal soil mass binder) by retaining a 0.4 w/s ratio in the binder before mixing with treated fibre [16]. 0.4 w/s ratio and 5% AAB were selected as the optimal binder material based on response to workability, binding behavior, the rapid increase in soil shear strength and low alkali reactivity with minimal greenhouse gas emissions.[15, 32]. AAB-blended BCS was continuously covered with a wet jute bag for 48 h to eliminate any excess heat during geopolymerisation. Prior to random mixing of discrete TCF and THF (0%, 0.10%, 0.20%, 0.30% and 0.4% by mass of AAB-treated BCS), it was oven dried for 24 h, maintaining a constant temperature of 104 °C. A set of microstructural and geotechnical tests were performed on 5% AAB-treated BCS at varying dosages of fibres.

Microstructural Characterisation

A detailed microstructural analysis was carried out on both untreated and treated fibre-reinforced AAB-blended BCS using X-ray photoelectron spectroscopy (XPS), Fourier-transform infrared (FTIR), thermogravimetric analysis (TGA) and stereomicroscope. Such studies explain the chemical compositions, the formation of cement-based products and the bonding behavior of the soil–fibre matrix.

XPS surface analysis was obtained using a Thermo scientific apparatus with K-alpha rays produced at 100–1400 eV. The operating range of an X-ray spot size of 100 µm was maintained with 180° double-focusing hemispherical analyzer. Both treated and untreated fibre samples were made in the form of pellets and oven dried before the samples were analyzed. The test data result was plotted using Avantage XPS software. FTIR spectra conducted using JASCO FTIR 4200 setup with K.Br. pellet arrangement. The transmittance spectral range was chosen from 4000 to 400 cm−1 for all the samples. TGA was performed for both treated and untreated fibres using Shimadzu-DTG-60 setup at a heating rate of 10 °C per minute for a mass sample of about 15 mg up to 900 °C under a nitrogen-rich atmosphere. Thermal stability analysis of both untreated and chemically treated CF and HF were examined at different elevated temperatures. Stereomicroscopic images visualized the surface texture, and physical features of raw materials and fibre–AAB–BCS at varying magnifications were captured through Olympus SXZ7 setup with the least dimension of 20 µm.

Geotechnical Characterisation

A series of geotechnical tests were conducted on untreated BCS-, TCF- and THF-reinforced AAB-treated BCS at different freeze and thaw cycles. The effect of TCF and THF inclusion on geomechanical activity subject to freeze–thaw cycles was tested by performing a compressive shear strength and indirect tensile strength study. In addition, both soil–fibre matrix cracking and tensile resistance and their efficiency were also tested.

Unconfined compressive strength (UCS) (as per ASTM D-2166), indirect tensile strength (ITS) (as per ASTM D-4123) and soaked and unsoaked California bearing ratio (CBR) (as per ASTM D-2435) tests were conducted at various fibre dosages. Untreated BCS- and AAB-treated BCS specimens were tested with their maximum dry density (MDD) and optimal moisture content (OMC). UCS samples were molded with a diameter of 38 mm and a height of 76 mm under a static strain rate of 1.25 mm/min. ITS was conducted in a Marshall stability machine by attaching a loading strip of 12.5 mm on the load frame at a strain rate of 50.5 mm/min. Cylindrical samples were prepared by maintaining 80 mm height and 100 mm in diameter. UCS and ITS specimens were wrapped in a cover for ensuring no moisture loss and kept for freeze–thaw cycles for both CF and HF dosages. The ITS value can be calculated using the following formula

$$S_{t} = \frac{{2 P_{{{\text{ult}}}} }}{\Pi td},$$

where Pult is ultimate load at which failure of sample occurred in (N), t is thickness of specimen (mm), d is diameter of the specimen (mm).

Freeze–thaw cycle studies were conducted on AAB-treated BCS reinforced with chemically strengthened CF and HF at varying doses. All specimens were stored in a temperature-controlled freezer. Each soil specimen was maintained at a freezing temperature of minimum − 10 °C for 16 h and remaining time for thaw temperature of around 23 ± 2 °C in 24 h. To ensure maximum frost penetration in the fibre-reinforced soil matrix, the minimum freezing temperature of − 10 °C was set. Moreover, the Indian Road Subgrade has reached − 10 °C as the lowest temperature in the last few decades, according to the Indian Road Congress. In this analysis, freeze–thaw cycles were replicated at 0, 1, 3, 5, 7 and 10 times for both CF- and HF-strengthened AAB–BCS. The temperature variation for the time during the freeze–thaw periods was provided in Fig. 3.

Fig. 3

Variation of temperature with respect to time for one freeze–thaw cycle

Regression Analysis

In this study, regression analysis was applied to experimental results to quantify the effects of fibres and alkaline binders at different temperatures. Regression analysis commonly used to relate the dependent variables with an independent variable under given statistical conditions [33, 34]. The prediction of regression modeling helps to check on how effective limited field data are put to use in decision-making. The predicted parameters considered were unconfined compressive strength and indirect tensile strength of both TCF- and THF-reinforced AAB-treated BCS at different freeze–thaw cycles. Dosages of chemically treated coir and hemp fibre, the number of freeze–thaw cycles were selected as input parameters to predict UCS and ITS for AAB mixed BCS, respectively. The statistical package in commercially available software PYTHON was used to perform the regression analysis. The regression equation for a response variable (y) and factors (NFT, and DFibre) affecting it can be formulated as follows:

$$Y = K + a \left( {D_{{{\text{Fiber}}}} } \right) + b\left( {N_{{{\text{FT}}}} } \right) + C\left( {D_{{{\text{Fiber}}}} } \right)^{2} + d\left( {D_{{{\text{Fiber}}}} \times N_{{{\text{FT}}}} } \right) + e\left( {N_{{{\text{FT}}}} } \right)^{2} ,$$

where a, b, c, d, and e are regression coefficients. K is constants; DFibre = dosage of treated coir or hemp fibre in the AAB–BCS mixture; NFT = number of freeze–thaw cycle used for fibre-reinforced AAB soil.

Results and Discussions

Microstructural Characterisation

X-Ray Photoelectron Spectroscopy (XPS)

XPS provided the surface elemental compositions for both chemically treated and untreated fibres in Fig. 4a, b. As expected, carbon (C), oxygen (O), silica (Si) were found to be the significant components of UCF and UHF. Aluminum (Al) and calcium (Ca) were also found along with traces amount of potassium (K) peaks in UCF, as shown in Fig. 4a. Iron (Fe) and nitrogen (N) were found additionally in UHF with a minor peak of calcium (Ca) in Fig. 4b. In the treated coir, the peak content of calcium (Ca) was found to increase. Moreover, both the TCF and THF showed an additional peak corresponding to sodium (Na) near 1100 eV, which may be due to the dissolution of sodium hydroxide compound in the fibres.

Fig. 4

XPS pattern for untreated and chemically treated a coir fiber, b hemp fiber

Fourier-Transform Infrared (FTIR) Spectroscopy

Figure 5 displayed the standard features of the IR spectrum for both hemp and coir fibres before and after NaOH treatment. The absorbance spectrum curve of raw fibres characterised the cellulose with a sharp peak at 3450 cm−1 as O–H stretching hydroxyl [35, 36]. Also, the spectra for treated coir fibres showed a slight reduction of the intensity around 3430 cm−1, followed by broadband in the range of 2970 cm−1 corresponding to C–H stretching vibrations. The peaks at 1640 cm−1 and 1460 cm−1 in hemp fibre represented the C=O alkene and symmetric –CH2 lignin [36]. Moreover, the bands at 1340 cm−1 corresponded to the C–H group, which is the characteristic of hemicellulose in both UHF and UCF [15]. This carbonation reaction may be due to an excessive amount of Na attributed to fibres. Both treated and untreated fibres exhibited a sharp band at 1020 cm−1, which corresponded to the asymmetric vibration Si(Al)–O group. However, the transmittance peaks corresponding to amorphous silica appeared around 530 and 460 cm−1, which represented the stretching and bending of the Si–O–Si group. Thus the spectrum peaks from untreated fibres and chemically treated BCS fibres showed similar bonds with a chemical shift of about 10 cm−1.

Fig. 5

FTIR spectroscopy for untreated and chemically treated coir and hemp fibers

Thermogravimetric Analysis (TGA)

Thermogravimetric (TG) curves for both untreated as well as chemically treated fibres were presented in Fig. 6a, b. As the temperature increased to 900 °C, various components in the fibre started decomposing, leading to a loss in mass at the respective temperature. The TG curves for both untreated fibres and chemically treated fibre samples indicated an initial loss in mass around 100 °C due to the vaporization of hygroscopic water [4, 37]. The weight loss in UCF around 310 °C and 390 °C was a consequence of hemicellulose and cellulose decomposition, as shown in Fig. 6a. Moreover, it was also interesting to note in Fig. 6b that the percentage of mass loss was high for UHF (5–6%) when compared to treated HF (3–4%). Thus the reduction in dehydration peaks of TCF and THF might be because of the encapsulation of fibre by deposition of sodium hydroxide around the surfaces [15]. Furthermore, the TGA curves followed asymptotic behavior beyond 600 °C for all the fibre samples. Hence the chemically treated CF and HF contributed to improving the fibre stability under high temperature.

Fig. 6

TGA curves for untreated and chemically treated a coir fiber, b hemp fiber

Stereomicroscopic Analysis

Figure 7a–d displayed the typical surface images for both untreated BCS and chemically treated fibre-reinforced AAB–BCS. Figure 7a showed the raw BCS, which reveals dark brown-colored particles indicating the presence of iron and illite–montmorillonite groups [4, 38]. There were also some visible surface cracks on soil, which impacted the swelling and shrinkage behavior. Figure 7b showed the deposition of a thin layer of AAB paste around the clay surface by filling the pores and cracks generated. Fibre addition in the AAB–soil matrix (Fig. 7c, d) showed the TCF and THF addition in the AAB mixed BCS, which could act as a spatial thread groove network by interlocking the particles through interfacial friction [14, 16]. Hence the discrete fibre inclusions in the BCS aided to improve the soil stiffness by holding the particles firmly around the fibres as a bridge surface.

Fig. 7

Stereomicroscopy images of a untreated BCS; b AAB mixed BCS; c TCF-reinforced AAB–BCS; d THF-reinforced AAB–BCS

Geoengineering Characterisation

Specific geotechnical properties of fly ash-based AAB–BCS combined with various amounts of treated coir and hemp fibres in the soil were obtained through a series of laboratory experiments. Table 5 showed the maximum dry density (MDD), optimum moisture content (OMC), and linear shrinkage (LS) of both fibre-reinforced BCS. From Table 5, it was observed that MDD increased and OMC decreased on addition of fibre up to 0.4%. The increase in coir fibre content in the AAB-treated BCS aided to an increase in the confinement bonding and frictional resistance during loading. As a result, it was difficult for the clay particles surrounding the fibres to change their position from one point to another and thereby improved the interlocking density. Coir fibre was also able to spread effectively in the soil specimen and gained the ability to resist the stress produced around the surface matrix when the fibre dosage was relatively low. Addition of excessive fibre resulted in a rough interface that hindered easy compaction of soil. Hence the fibre-reinforced AAB-treated BCS reached their peak density at a threshold fibre content of 0.4%. Linear shrinkage decreased as a result of addition of fibre to AAB.

Table 5 Geoengineering properties of AAB-treated BCS at varying percentages of coir and hemp fibers

Unconfined Compressive Strength (UCS)

Figure 8a, b revealed UCS of two types of fibre-reinforced AAB-treated BCS with a different number of freeze–thaw cycles. The combined application of AAB and chemically treated fibres had a significant effect on the shear strength of BCS. The growth in UCS composite hemp fibre and AAB inclusions was even greater than AAB–coir fibre. It was also found that THF–AAB mixed UCS tests (Fig. 8a) improved by around 12% over the TCF–AAB matrix (Fig. 8b). This improvement was attributed to better interfacial adhesion with pozzolanic compounds around the surfaces of fibre. THF produced a higher surface roughness and contact area with the soil, as a result of which frictional resistance and confinement bonding were significantly improved during loading. TCF–AAB had a relatively low bonding contact and a smooth surface layer. The reduction of the TCF linkage effect across the clay particles had a major influence on the shear resistance properties. Dissolution of pozzolanic compounds and ion consumption by existence of active moisture in hemp fibre helped to form the spatial thread network by increasing the retaining power of the particles. Thus, the presence of THF in AAB-treated soil altered the morphology and molecular bonding interaction upon addition of fly ash effectively [39, 40]. This was in line with the compressive and tensile test results of lime-treated coir fibre-blended soft soil [26]. CF and HF reinforced AAB–BCS showed higher stiffness than unreinforced BCS. In most cases, soil deformation was resisted by fibre across large failure surfaces [41]. This result was consistent with the previously reported results for other stabilized clay soils with coir and hemp fibres [42, 43]. UCS of fibre–AAB-treated BCS without freeze–thaw cycles yielded a higher intensity than that obtained in the first freeze–thaw cycles. It was also interesting to note that after the first and third freeze–thaw cycles, the shear strength of both blended BCS fibres decreased and then reactivated after the fourth freeze–thaw cycle. Many investigators mentioned a similar trend in their results that the use of 0.2% polypropylene fibre with 3–6% cement improved the UCS after five freeze–thaw cycles [44]. Several researchers documented an improvement in shear strength after freezing, which was usually observed to decrease initially and then rise [45,46,47]. The variations in UCS due to freeze–thaw could be related to moisture redistribution and particle reorientation. Related observations had also been found in biochemical experiments on the impact of fibre–moisture variations [29, 48]. Continuous replication of the frozen–thaw-treated fibre–AAB–BCS could make a new fibre–soil structure matrix that effectively resists compressive force [31]. In comparison, excessive fibre (beyond 0.4%) inclusion in the soil might improve the smooth texture function, which did not allow the soil to compress. As an outcome, both the fibre-reinforced AAB-treated BCS hit their optimum intensity at a fibre content of 0.4% as a threshold. Comparing the UCS findings obtained during the fibre–AAB–BCS freeze–thaw process, it could be inferred that cement bonding and higher interfacial friction would effectively stabilize the soil during shearing.

Fig. 8

Variation of UCS of AAB-treated BCS reinforced with a coir; b hemp fibers at varying freeze–thaw cycles

Indirect Tensile Strength (ITS)

Figure 9a, b showed the typical tensile curves of AAB-treated specimens with different fibre dosages in the BCS. The trend of tensile resistance for TCF- (Fig. 9a) and THF (Fig. 9b)-reinforced BCS increased with an increasing percentage of fibre. The graph also revealed that the THF-reinforced BCS had a higher ITS value than the TCF. A major increase in tensile cracking resistance in THF–AAB–BCS was attributed to high surface roughness and contact area [49]. Comparing with earlier studies [26, 42, 48], it was observed that soil–fibre bonding efficiency depends on the roughness of the fibre and the degree of interfacial friction production along the length of the soil–fibre. This similarity of findings with those of the previous studies could be observed due to identical interactions of fibre in the soil. The use of fibre compounds in the BCS allowed reducing deformation and brittleness by the development of surface interfacial stiffness and bonding mechanisms. The results of this study were consistent with those researchers who reported that the incorporation of fibre in soil could increase the stiffness behaviour by reducing soil tensile cracks [13, 39, 48,49,50]. The recurrence of the AAB–BCS freeze–thaw cycle was, however, advantageous. If the number of cycles was greater than five, the soil stiffness often increases. As a result, the toughness on the soil surface and the bearing power ratio dramatically improved. It was also important to note that minor reductions were observed in ITS values for all AAB-treated soils after one and three freeze–thaw cycles. This low tensile resistance varied due to temperature variability and the creation of weak bond between soil–fibre interfaces [44, 49]. Thus, the blended combination of soil–AAB–fibre matrix suggested a high resistance to tensile cracking.

Fig. 9

Variation of ITS of AAB-treated BCS reinforced with a coir; b hemp fibers at varying freeze–thaw cycles

California Bearing Ratio (CBR)

The CBR tests of TCF- and THF-reinforced AAB-treated BCS were shown in Fig. 10 for both soaked and unsoaked conditions. The CBR of untreated BCS in soaked conditions increased from 1.96 to 3.8% after treatment with AAB. For unsoaked conditions, the CBR modified substantially (with and without AAB) for raw BCS. While the reinforcement of fibre in the soil was usually required to improve the stress bearing ratio (SBR). Some researchers reported an improvement in CBR due to fibre inclusions [21, 36, 51]. The application of fibres to AAB-treated BCS had improved both penetration resistance and durability effectively. This development might be attributed to the creation of a dense soil–fibre bridge network by rapid dissolution of pozzolanic compounds during the soaking process [16]. The CBR results of THF–AAB–BCS mixture improved significantly after the addition of 0.2% of the fibre and hit the peak after the inclusion of up to 0.4% of the THF as compared to TCF. THF also provided greater resistance to penetration of the plunger due to its high surface roughness and interfacial friction. Therefore the rise in CBR value might be attributed to the improvement in a definite interlocking fibre–soil matrix with geopolymerisation reaction.

Fig. 10

Variation of soaked and unsoaked CBR values of raw BCS-, TCF- and THF-reinforced AAB-treated BCS

Regression Analysis

The influence of fly ash-based AAB with different amount of fibres (either CF, or HF) on UCS, and ITS values at different freeze–thaw cycles were expressed through a nonlinear regression equation are as follows.

$${\text{UCS}}_{{\text{Prd } - \text{ CF}}} = 572.8872 + 590.9913 \times \left( {D_{{{\text{Fiber}}}} } \right) - 13.7628 \times \left( {N_{{{\text{FT}}}} } \right) + 239.2857 \times \left( {D_{{{\text{Fiber}}}} } \right)^{2} + 35.6449 \times \left( {D_{{{\text{Fiber}}}} \times N_{{{\text{FT}}}} } \right) + 1.869 \times \left( {N_{{{\text{FT}}}} } \right)^{2} \;{\text{with}}\; R^{2} = 0.975,$$
$${\text{UCS}}_{{\text{Prd } - \text{ HF}}} = 575.773 + 387.6769 \times \left( {D_{{{\text{Fiber}}}} } \right) - 14.6528 \times \left( {N_{{{\text{FT}}}} } \right) + 635.7143 \times \left( {D_{{{\text{Fiber}}}} } \right)^{2} + 51.7009 \times \left( {D_{{{\text{Fiber}}}} \times N_{{{\text{FT}}}} } \right) + 2.0186 \times \left( {N_{{{\text{FT}}}} } \right)^{2} {\text{with}} R^{2} = 0.966,$$
$${\text{ITS}}_{{\text{Prd } - \text{ CF}}} = 54.6133 - 94.1277 \times \left( {D_{{{\text{Fiber}}}} } \right) - 7.8318 \times \left( {N_{{{\text{FT}}}} } \right) + 533.333 \times \left( {D_{{{\text{Fiber}}}} } \right)^{2} + 26.6449 \times \left( {D_{{{\text{Fiber}}}} \times N_{{{\text{FT}}}} } \right) + 0.8693 \times \left( {N_{{{\text{FT}}}} } \right)^{2} {\text{with}} R^{2} = 0.938,$$
$${\text{ITS}}_{{\text{Prd } - \text{ HF}}} = 56.5676 - 33.1720 \times \left( {D_{{{\text{Fiber}}}} } \right) - 8.3773 \times \left( {N_{{{\text{FT}}}} } \right) + 477.3810 \times \left( {D_{{{\text{Fiber}}}} } \right)^{2} + 32.2430 \times \left( {D_{{{\text{Fiber}}}} \times N_{{{\text{FT}}}} } \right) + 0.8994 \times \left( {N_{{{\text{FT}}}} } \right)^{2} {\text{with}} R^{2} = 0.945 .$$

The above proposed nonlinear regression equations for both fibre-reinforced AAB-treated BCS revealed that the observation points for UCS and ITS tests were nearly close to the experimental results, which were confirmed through the coefficient of determination correlation (R2). Moreover, Tables 6 and 7 predicted the error percentages between the experimental and model-predicted results for UCS and ITS at different freeze–thaw cycles. Figures 11 and 12 showed the graphical representation between tested data and predicted data. Figures 11a and 12a indicated the UCS and ITS results for both experiment and predicted results of CF–AAB-treated BCS. Figures 11b and 12b represented the HF–AAB-reinforced BCS for tested, predicted UCS and ITS results, respectively.

Table 6 Error percentage between tested and predicted data of UCS for fiber–AAB–BCS at different freeze–thaw cycles
Table 7 Error percentage between tested and predicted data of ITS for fiber–AAB–BCS at different freeze–thaw cycles
Fig. 11

Comparison of tested and predicted results of UCS for a TCF-; b THF-reinforced AAB-treated BCS

Fig. 12

Comparison of tested and predicted results of ITS for a TCF-; b THF-reinforced AAB-treated BCS


  • The present paper proposed a novel technique of geopolymerisation of expansive soil with fly ash-based alkaline binders and at the same time improved the tensile properties of soil with the addition of naturally available coir and hemp fibres. The process, if successfully implemented, was expected to lead to an eco-friendly and sustainable solution for ground improvement.

  • Usage of chemically treated natural fibres with envirosafe AAB helped to improve the soil stiffness property and overcame the brittleness nature effectively during dry conditions. The effect of fibre and AAB inclusion in BCS through shear strength, tensile, and durability characteristics of both coir and hemp fibres reinforced soil was compared at different cyclic temperatures.

  • Microstructural studies on fibre-reinforced AAB-treated BCS confirmed the development of a new surface morphology and molecular bonding in both fibre AAB-reinforced BCS. The TGA results indicated a drop in the mass fraction of chemically treated fibres in comparison with untreated fibres.

  • Geomechanical studies indicated that THF–AAB–BCS achieved higher interlocking density, stiffness, compressive shear and tensile cracking resistance than GF–AAB–BCS by creating a spatial thread groove network. In addition, it improved soil–fibre durability after the fifth freeze–thaw cycle. Strength-bearing ratio in terms of CBR performance was greatly enhanced by the formation of active cementitious bonding and interfacial friction along the surface of fibre and clay particles.

  • An empirical model for prediction of fibre-reinforced AAB-treated BCS was proposed and validated, and it can be used to estimate the UCS, ITS of stabilized clay by considering the effects of fibre content and numbers of freeze–thaw cycles.


  1. 1.

    Das B (2003) Chemical and mechanical stabilization. Transportation Research Board

  2. 2.

    Al-Rawas AA, Hago AW, Al-Sarmi H (2005) Effect of lime, cement and Sarooj (artificial pozzolan) on the swelling potential of an expansive soil from Oman. Build Environ 40:681–687. https://doi.org/10.1016/j.buildenv.2004.08.028

    Article  Google Scholar 

  3. 3.

    Nelson D, Miller DJ (1993) Book review expansive soils-problems and practice in foundation and pavement engineering

  4. 4.

    Sivapullaiah PV, Prasad BG, Allam MM (2009) Effect of sulfuric acid on swelling behavior of an expansive soil. Soil Sedim Contam 18:121–135. https://doi.org/10.1080/15320380802660289

    Article  Google Scholar 

  5. 5.

    Saride S, Puppala AJ, Chikyala SR (2013) Swell-shrink and strength behaviors of lime and cement stabilized expansive organic clays. Appl Clay Sci 85:39–45. https://doi.org/10.1016/j.clay.2013.09.008

    Article  Google Scholar 

  6. 6.

    Ural N (2016) Effects of additives on the microstructure of clay. Road Mater Pave Des 17:104–119. https://doi.org/10.1080/14680629.2015.1064011

    Article  Google Scholar 

  7. 7.

    Gartner E (2004) Industrially interesting approaches to “low-CO2” cements. Cem Concr Res 34:1489–1498. https://doi.org/10.1016/j.cemconres.2004.01.021

    Article  Google Scholar 

  8. 8.

    Kar A, Ray I, Halabe UB et al (2014) Characterisations and quantitative estimation of alkali-activated binder paste from microstructures. Int J Concret Struct Mater 8:213–228. https://doi.org/10.1007/s40069-014-0069-0

    Article  Google Scholar 

  9. 9.

    Sharma NK, Swain SK, Sahoo UC (2012) Stabilization of a clayey soil with fly ash and lime: a micro level investigation. Geotech Geol Eng 30:1197–1205. https://doi.org/10.1007/s10706-012-9532-3

    Article  Google Scholar 

  10. 10.

    Lin B, Cerato AB, Andrew SM, Madden MEE (2013) Effect of fly ash on the behavior of expansive soils: microscopic analysis. Environ Eng Geosci 19:85–94. https://doi.org/10.2113/gseegeosci.19.1.85

    Article  Google Scholar 

  11. 11.

    Miller GA, Azad S (2000) Influence of soil type on stabilization with cement kiln dust. Constr Build Mater 14:89–97. https://doi.org/10.1016/S0950-0618(00)00007-6

    Article  Google Scholar 

  12. 12.

    Miao S, Wei C, Huang X et al (2017) Stabilization of highly expansive black cotton soils by means of geopolymerization. J Mater Civ Eng 29:04017170. https://doi.org/10.1061/(asce)mt.1943-5533.0002023

    Article  Google Scholar 

  13. 13.

    Kumar A, Walia BS, Bajaj A (2007) Influence of fly ash, lime, and polyester fibres on compaction and strength properties of expansive soil. J Mater Civ Eng 19:242–248. https://doi.org/10.1061/(asce)0899-1561(2007)19:3(242)

    Article  Google Scholar 

  14. 14.

    Tang C, Shi B, Gao W et al (2007) Strength and mechanical behavior of short polypropylene fibre reinforced and cement stabilized clayey soil. Geotext Geomembr 25:194–202. https://doi.org/10.1016/j.geotexmem.2006.11.002

    Article  Google Scholar 

  15. 15.

    Gupta S, GuhaRay A, Kar A, Komaravolu VP (2018) Performance of alkali-activated binder-treated jute geotextile as reinforcement for subgrade stabilization. Int J Geotech Eng. https://doi.org/10.1080/19386362.2018.1464272

    Article  Google Scholar 

  16. 16.

    Mazhar S, Guharay A, Kar A (2019) Stabilization of expansive clays by combined effects of geopolymerization and fibre reinforcement. J Inst Eng India Ser A. https://doi.org/10.1007/s40030-019-00418-3

    Article  Google Scholar 

  17. 17.

    Davidovits J (1994) Properties of geopolymer cements. Alkaline Cements Concret Kiev Ukraine. https://doi.org/10.1073/pnas.0811322106

    Article  Google Scholar 

  18. 18.

    Provis LJ, van Deventer SJ (2014) Alkali activated materials

  19. 19.

    Rios S, Cristelo N, Viana da Fonseca A, Ferreira C (2015) Structural Performance of Alkali-Activated Soil ash versus soil cement. J Mater Civ Eng 28:04015125. https://doi.org/10.1061/(asce)mt.1943-5533.0001398

    Article  Google Scholar 

  20. 20.

    Asadi A, Huat BBK, Cristelo N et al (2016) Application of alkali-activated agro-waste reinforced with wollastonite fibres in soil stabilization. J Mater Civ Eng 29:04016206. https://doi.org/10.1061/(asce)mt.1943-5533.0001735

    Article  Google Scholar 

  21. 21.

    Adhikari B, Khattak MJ, Adhikari S (2019) Mechanical and durability characteristics of flyash-based soil–geopolymer mixtures for pavement base and subbase layers. Int J Pavement Eng. https://doi.org/10.1080/10298436.2019.1668562

    Article  Google Scholar 

  22. 22.

    Sukmak P, Horpibulsuk S, Shen SL (2013) Strength development in clay-fly ash geopolymer. Constr Build Mater 40:566–574. https://doi.org/10.1016/j.conbuildmat.2012.11.015

    Article  Google Scholar 

  23. 23.

    Chinkulkijniwat A, Arulrajah A, Horpibulsuk S et al (2016) Stabilisation of marginal lateritic soil using high calcium fly ash-based geopolymer. Road Mater Pavement Des 17:877–891. https://doi.org/10.1080/14680629.2015.1132632

    Article  Google Scholar 

  24. 24.

    Mishra B, Kumar Gupta M (2018) Use of randomly oriented polyethylene terephthalate (PET) fibre in combination with fly ash in subgrade of flexible pavement. Constr Build Mater 190:95–107. https://doi.org/10.1016/j.conbuildmat.2018.09.074

    Article  Google Scholar 

  25. 25.

    Shukla SK (2017) Fundamentals of fibre-reinforced soil engineering. Springer, Berlin

    Google Scholar 

  26. 26.

    Anggraini V, Asadi A, Farzadnia N et al (2016) Reinforcement benefits of nanomodified coir fibre in lime-treated marine clay. J Mater Civ Eng 28:1–8. https://doi.org/10.1061/(ASCE)MT.1943-5533.0001516

    Article  Google Scholar 

  27. 27.

    Wang Y-X, Guo P-P, Ren W-X et al (2017) Laboratory investigation on strength characteristics of expansive soil treated with jute fibre reinforcement. Int J Geomech 17:04017101. https://doi.org/10.1061/(asce)gm.1943-5622.0000998

    Article  Google Scholar 

  28. 28.

    Verma D, Gope PC (2105) The use of coir/coconut fibres in composites

  29. 29.

    Bordoloi S, Hussain R, Garg A et al (2017) Infiltration characteristics of natural fibre reinforced soil. Transport Geotech 12:37–44. https://doi.org/10.1016/j.trgeo.2017.08.007

    Article  Google Scholar 

  30. 30.

    Al-Akhras NM, Attom MF, Al-Akhras KM, Malkawi AIH (2008) Influence of fibres on swelling properties of clayey soil. Geosynth Int 15:304–309. https://doi.org/10.1680/gein.2008.15.4.304

    Article  Google Scholar 

  31. 31.

    Graham J, Au VCS (1985) Effects of freeze-thaw and softening on a natural clay at low stresses. Can Geotech J 22:69–78. https://doi.org/10.1139/t85-007

    Article  Google Scholar 

  32. 32.

    Syed M, GuhaRay A, Agrawal S, Kar A (2019) Stabilisation of expansive clays by combined effects of geopolymerisation and fibre reinforcement. J Inst Eng (India) Ser A Springer 101(1):163–178. https://doi.org/10.1007/s40030-019-00418-3

    Article  Google Scholar 

  33. 33.

    Sivakumar Babu GL, Vasudevan AK (2008) Strength and stiffness response of coir fibre-reinforced tropical soil. J Mater Civ Eng 20:571–577. https://doi.org/10.1061/(ASCE)0899-1561(2008)20:9(571)

    Article  Google Scholar 

  34. 34.

    Moghal AAB, Chittoori BCS, Basha BM, Al-Mahbashi AM (2018) Effect of polypropylene fibre reinforcement on the consolidation, swell and shrinkage behaviour of lime-blended expansive soil. Int J Geotech Eng 12:462–471. https://doi.org/10.1080/19386362.2017.1297002

    Article  Google Scholar 

  35. 35.

    Latifi N, Meehan CL, Majid MZA, Horpibulsuk S (2016) Strengthening montmorillonitic and kaolinitic clays using a calcium-based non-traditional additive: a micro-level study. Appl Clay Sci 132–133:182–193. https://doi.org/10.1016/j.clay.2016.06.004

    Article  Google Scholar 

  36. 36.

    Guo X, Shi H, Chen L, Dick WA (2010) Alkali-activated complex binders from class C fly ash and Ca-containing admixtures. J Hazard Mater 173:480–486. https://doi.org/10.1016/j.jhazmat.2009.08.110

    Article  Google Scholar 

  37. 37.

    Aslani F (2015) Thermal performance modeling of geopolymer concrete. J Mater Civ Eng ASCE. https://doi.org/10.1061/(ASCE)MT.1943-5533

    Article  Google Scholar 

  38. 38.

    Ogundalu AO, Oyekan GL (2014) Mineralogical and geotechnical characterisation of maiduguri black cotton soil by X-ray diffraction (XRD), X-ray photoelectron (XPS) and scanning electron spectroscopy (SEM). Int J Eng Technol 4:345–353

    Google Scholar 

  39. 39.

    Sudhakaran SP, Sharma AK, Kolathayar S (2018) Soil stabilization using bottom ash and areca fibre: experimental investigations and reliability analysis. J Mater Civ Eng 30:1–10. https://doi.org/10.1061/(ASCE)MT.1943-5533.0002326

    Article  Google Scholar 

  40. 40.

    Moghal AAB, Chittoori BCS, Basha BM, Al-Shamrani MA (2017) Target reliability approach to study the effect of fibre reinforcement on UCS behavior of lime treated semiarid soil. J Mater Civ Eng 29:04017014. https://doi.org/10.1061/(asce)mt.1943-5533.0001835

    Article  Google Scholar 

  41. 41.

    Güllü H, Khudir A (2014) Effect of freeze-thaw cycles on unconfined compressive strength of fine-grained soil treated with jute fibre, steel fibre and lime. Cold Reg Sci Technol 106–107:55–65. https://doi.org/10.1016/j.coldregions.2014.06.008

    Article  Google Scholar 

  42. 42.

    Anggraini V, Asadi A, Huat BBK, Nahazanan H (2015) Effects of coir fibres on tensile and compressive strength of lime treated soft soil. Meas J Int Meas Confeder. https://doi.org/10.1016/j.measurement.2014.09.059

    Article  Google Scholar 

  43. 43.

    Bordoloi S, Garg A, Sekharan S (2017) A review of physio-biochemical properties of natural fibres and their application in soil reinforcement. Adv Civ Eng Mater 6:20160076. https://doi.org/10.1520/acem20160076

    Article  Google Scholar 

  44. 44.

    Ding M, Zhang F, Ling X, Lin B (2018) Effects of freeze-thaw cycles on mechanical properties of polypropylene fibre and cement stabilized clay. Cold Reg Sci Technol 154:155–165. https://doi.org/10.1016/j.coldregions.2018.07.004

    Article  Google Scholar 

  45. 45.

    Gullu H, Hazirbaba K (2010) Unconfined compressive strength and post-freeze-thaw behavior of fine-grained soils treated with geofibre and synthetic fluid. Cold Reg Sci Technol 62:142–150. https://doi.org/10.1016/j.coldregions.2010.04.001

    Article  Google Scholar 

  46. 46.

    Wang D, Ma W, Niu Y et al (2007) Effects of cyclic freezing and thawing on mechanical properties of Qinghai-Tibet clay. Cold Reg Sci Technol 48:34–43. https://doi.org/10.1016/j.coldregions.2006.09.008

    Article  Google Scholar 

  47. 47.

    Dayioglu M, Cetin B, Nam S (2017) Stabilization of expansive Belle Fourche shale clay with different chemical additives. Appl Clay Sci. https://doi.org/10.1016/j.clay.2017.05.033

    Article  Google Scholar 

  48. 48.

    Bordoloi S, Leung AK, Gadi VK et al (2019) Water retention and desiccation potential of lignocellulose-based fibre-reinforced soil. J Geotech Geoenviron Eng 145:1–9. https://doi.org/10.1061/(ASCE)GT.1943-5606.0002153

    Article  Google Scholar 

  49. 49.

    Shi B, Gu K, Cui Y-J et al (2012) Desiccation cracking behavior of polypropylene fibre–reinforced clayey soil. Can Geotech J. https://doi.org/10.1139/t2012-067

    Article  Google Scholar 

  50. 50.

    Cai Y, Shi B, Ng CWW, Sheng TC (2006) Effect of polypropylene fibre and lime admixture on engineering properties of clayey soil. Eng Geol 87:230–240. https://doi.org/10.1016/j.enggeo.2006.07.007

    Article  Google Scholar 

  51. 51.

    Tang C-S, Shi B, Li J et al (2016) Tensile strength of fibre-reinforced soil. J Mater Civ Eng 28:04016031. https://doi.org/10.1061/(asce)mt.1943-5533.0001546

    Article  Google Scholar 

Download references


The authors would like to express their sincere gratitude to the Central Analytical Laboratory Facilities at BITS-Pilani, Hyderabad Campus, for providing the setup for the XPS, FTIR and TGA analyses.

Author information



Corresponding author

Correspondence to Anasua GuhaRay.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Syed, M., GuhaRay, A., Goel, D. et al. Effect of Freeze–Thaw Cycles on Black Cotton Soil Reinforced with Coir and Hemp Fibres in Alkali-Activated Binder. Int. J. of Geosynth. and Ground Eng. 6, 19 (2020). https://doi.org/10.1007/s40891-020-00200-7

Download citation


  • Expansive black cotton soil
  • Freeze–thaw cycle
  • Alkali-activated binder
  • Regression analysis
  • Coir and hemp fibres