Introduction

Application of cement as binder in soil stabilization is a widely used method for ground improvement [1,2,3,4]. However the high quantity of CO2 released into the atmosphere is the main drawback of using cement [5]. The cement industry produces 5% of global man-made CO2 emissions [6, 7]. Beside the emission of CO2, another by-product of cement production is NO x . Most of these nitrogen oxides are produced in cement kilns, which can contribute to the greenhouse effect and acid rain [8]. Several attempts involving the use of alternative methods or by-products as partial or full replacements of cement as stabilizers have been made to alleviate this. In this respect, binders based on alkali-activated materials have received a significant interest due to their sustainability advantages [9,10,11,12,13,14,15].

The alkali activation is a process in which, the aluminosilicate materials (industrial wastes and by-products) were dissolved through an alkaline activator such as sodium hydroxide (NaOH) or potassium hydroxide (KOH) [16]. Alkali-activated materials may have low (e.g. fly ash class F and metakaolin) or high (e.g. ground granulated blastfurnace slag (GGBS) and fly ash class C) calcium contents [17,18,19,20]. The formed geopolymeric gel can be viewed as an alkaline aluminosilicate base on amorphous Na+ or K+ aluminosilicate structure [21].

The literature already has an abundance of papers on alkaline-activation technology. In this regard, a number of studies were conducted to investigate the environmental benefits of alkaline-activated binder [22,23,24]; notably that of Weil et al. [25], who assessed the CO2 emissions (kg CO2−eq/m3) and financial cost of alkali-activated binders. Those researchers reported that for an alkali-activated binder comprising 15.9% source binder dosage in the presence of high-alkali solutes, a reduction of 131 kg CO2−eq/m3 was achieved for the combined feedstock and transport of the binders compared with that observed for cement. The results obtained so far show the effectiveness of alkali-activated binders for soil stabilization. In this regard, the effectiveness of alkali-activated fly ash (FA) as a source of silica and alumina for soil stabilization has been studied [26,27,28,29,30]. The feasibility of using metakaolin as an alkali-activated soil stabilizer at shallow depths was also studied [31]. These studies were conducted by mixing the above-mentioned binders with soft soils in the presence of NaOH and a silica-rich source such as sodium silicate as the alkaline activator. A recent studies by Pourakbar et al. [32] investigated the feasibility of using palm oil fuel ash (POFA) in the presence of high alkaline solutes (NaOH and KOH) for the purpose of soil stabilization. According to this investigation, a significant unconfined compressive strength (UCS) increase occurred at long curing times (90 days and 180 days of curing) when 10 M KOH was used as the best concentration for strength improvement of the soil. This was caused by the fact that K+ has a larger size than the other alkaline metal cation, and therefore allows more dense and intimate polycondensation reactions, which substantially increase the overall long-term strength of the treated soil. In another attempt, Pourakbar et al. [33], have concluded that the inclusion of fiber reinforcement in alkali-activated POFA, regardless of their particular activator-precursor combinations, improved the postpeak behavior, by modifying the original brittle response into a more ductile behaviour.

Despite such positive findings, the above mentioned precursors (FA, POFA and metakaolin) are subjected to pre-treatment such as calcination and grinding to increase the reactivity of Al and Si present in binders [34,35,36]. The high temperatures utilized during the calcination process result in high energy usage and CO2 emissions.

This paper focuses on the use of olivine (Mg2SiO4) to provide a more sustainable approach for the preparation of alkali activated binders to be used in soil stabilization. Olivine is a magnesium silicate, whose deposits are globally located [37]. It contains 45–49% magnesium oxide (MgO) and 40% SiO2. Its weak nesosilicate structure and the absence of strong Si–O–Si bonds leaves olivine susceptible to dissolution and subsequent chemical reaction [38, 39]. Its high SiO2 and alkaline metal content makes this natural resource an ideal candidate for alkali activation.

Due to its role as an effective source of MgO and SiO2 with weak chemical bonds, olivine can be a good candidate for soil stabilization after being subjected to alkali activation. The aim of this study is to explore the use of olivine in the presence of KOH for the development of high strengths during soil stabilization. The behaviour of olivine treated soil was examined through UCS measurements. The composition and microstructure development of activated olivine treated soils were examined using scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDX), X-ray diffraction (XRD).

Materials and Methods

Materials

Soil

According to the Unified Soil Classification System [40], the original soil is classified as high-plasticity clay (CH). Table 1 shows the physical and engineering properties of soil as-received and soil treated with different percentages of olivine. Table 2 lists the chemical composition of soil determined by X-ray fluorescence (XRF) analysis. Some of the main constituents are 28.5% SiO2, 15.8% CaO, 8.3% Al2O3 and 0.25% MgO. The particle size distribution of the soil sample is shown in Fig. 1. The strength of this type of soil is often insufficient to enable its use in earth works or foundation layers, and thus constitutes an ideal challenge.

Table 1 Physical and engineering properties of soil
Full size table
Table 2 Chemical composition of silty clay soil determined by XRF
Full size table
Fig. 1
figure 1

Particle size distribution of soil

Full size image

Olivine

The olivine was obtained from MAHA chemical company in Malaysia. The chemical composition of olivine obtained using XRF is listed in Table 3. This olivine sample contains 48.3% MgO and 40.3% SiO2. The D50 and SSA of olivine was 2.24 μm and 6.07 m2/g, respectively.

Table 3 Chemical composition of olivine determined by XRF
Full size table

Potassium Hydroxide

Potassium hydroxide (KOH), containing alkaline cation, was selected as activator because of its well-known efficiency. Although the price of NaOH is less than KOH, the authors used KOH as an alkali activator because K+ has a larger size than the other alkaline metal cation, and therefore allows more dense and intimate polycondensation reactions, which substantially increase the overall long-term strength of the treated soil. These explanation is consistent with a study by Phair and Deventer [41]. According to those researchers’ findings, matrices containing K+ exhibited higher compressive strength and specific surface area and a lower degree of crystallinity in long curing regimes. This reagent was supplied in pellet form by the company R&M chemical and were previously diluted in distilled water to achieve a predesigned concentration. Note that the reaction with water of KOH is strongly exothermic, as with all strong bases. Thus, the alkaline solute was left to cool to ambient temperature before use.

Methods

Mix Composition and Sample Preparation

The compositions of the various mixes of soil, olivine, KOH and water prepared under this study are given in Table 4. Mix labelled “KS” represents an alkali treated soil. Mixes labelled KOBS represent as alkali treated soils containing B: 5, 10, 15 and 20% of olivine respectively at different curing time of 7 and 90 days. The concentration of KOH was fixed at 10 M for all samples according to the findings of previous studies [27, 29, 42].

Table 4 Mix compositions prepared under this study
Full size table

The alkali activator was dissolved in distilled water at a predetermined concentration. The solution prepared was left to cool down for 24 h before being added to the olivine and soil mixture. The soil was air-dried before mixing to remove any remaining moisture. All specimens were cured in plastic bags for 7 and 90 days to prevent any changes in the initial water content due to evaporation. Note that the optimum water content (OWC) corresponding to maximum dry density (MDD) was determined for each mix composition (Table 4).

Unconfined Compressive Strength

To prepare the samples, alkali activators were dissolved in distilled water at predetermined concentrations to avoid the effect of unknown contaminants in the mixing water. The dry soil was initially mixed with the olivine and activator solution was then added to the solids accoriding to OWC and MDD of each percanatge of olivin in soil and thoroughly mixed until a uniform blend was achieved. After de-moulding, all specimens were cured in plastic bags for 7 and 90 days to prevent changes in water content due to evaporation. Finally, the UCS measurements were conducted on the prepared samples with a diameter of 50 mm and a height of 100 mm according to BS 1924: Part 7 [43] in three different specimens, and the results were accepted only if deviated <5% from the average. The equipment used for this purpose was an Instron 3300, operated at a loading rate of 100 kN/min.

Microstructural Analysis

Samples were examined in a presence of KOH by using a JSM 5700 SEM coupled with an EDX spectrometer. Samples were sputter coated with gold using a Emitech K550X Sputter Coater before analysis to increase the electrical conductivity of the surface and reduce charging. Crystalline phases were investigated using a Philips X-Ray diffractometer equipped with X’Pert software over a 2-theta range of 3°–50° Cu K alpha.

Results and Discussion

Strength Development

Figure 2 shows the stress–strain behaviour of soil (S), alkali-activated soil (KS) and alkali-activated soil (KOBS) with different percentages of olivine (5, 10, 15, and 20) after 7 days of curing. Untreated soil specimens (S) showed a ductile behaviour with strength of 103.4 kPa at a failure strain of around 1.8%. Alkali activation improved the strength of the soil with strengths reported as 292 kPa after 7 days of curing. Furthermore, the stress–strain behaviour of alkali activated olivine treated soil increased by increasing the olivine percentages of 5, 10, 15 and 20%. According to Fig. 2 the strength of alkali activated olivine treated soil after 7 days of curing was 389, 550, 675 and 953 kPa for KO5S, KO10S, KO15S and KO20S samples, respectively.

Fig. 2
figure 2

UCS of 0, 5, 10, 15 and 20 of olivine treated soil in the presence of 10 M KOH after 7 day curing times

Full size image

Figure 3 shows the stress–strain behaviour of alkali-activated soil (KS) and alkali-activated olivine treated soil after a curing time of 90 days. This figure shows the strength of the soil increased to 5.05, 5.27, 6.71 and 7.4 MPa for the KO5S, KO10S, KO15S and KO20S samples respectively. A rapid increase in the strength of soil specimens in the presence of KOH was observed over time, which was attributed to the role of KOH in promoting the dissolution of Si and Al present in soil [29]. The increase in the strength of KS after 7 and 90 days curing was not significant due to the low reactivity of Si and Al. A progressive increase in strength was observed when different amounts of olivine were added to the mixture. The strength of samples increased with olivine content (0–20%) and curing time (7 and 90 days). In this respect, KO20S achieved the highest strength within all alkali treated soil samples after 7 and 90 days of curing. The important role of olivine in the strength development of soil samples was obvious as significant increases in strength were observed with an increase in the olivine content. Figure 3 shows that alkali-activated specimens achieved higher strengths after 90 days of curing. The rate of increase was almost constant at all ages although it augmented as the olivine content increased to 20%.

Fig. 3
figure 3

UCS of 0, 5, 10, 15 and 20 of olivine treated soil in the presence of 10 M KOH after 90 day curing times

Full size image

The underlying mechanisms responsible for the strength development of the soils after alkali treatment can be explained in terms of the reactions that take place during alkali activation. When olivine is mixed with KOH within a soil mix, the KOH solution leaches the silicon from amorphous phases. The formation of Mg(OH)2 is an expansive process, which fills the available pores in the soil, thereby increasing the density and final strength. The presence of KOH can result in the leaching of the Si and Al in the amorphous phases of soil, producing an alumina-silica-hydrate (A-S-H) gel.

Previous studies have shown that the addition of calcium (Ca) generally has a positive effect on the mechanical properties of alkali activated binders [33, 44]. Calcium ions are capable of acting as a charge-balancing cation within the alkali-activated binder structure. It was also proposed that the simultaneous formation of A-S-H and C-S-H gels may contribute to an increase in the compressive strength of the matrix [45]. Therefore, besides Si and Al, Mg ions may play a significant role in crystal growth. In the presence of an alkali activator, Mg ions could provide additional nucleation sites for the precipitation of dissolved species and contribute to the formation of magnesium silicate hydrate (M-S-H) gel [21, 46, 47]. The strength development of the prepared samples could be attributed to these reactions.

Figure 4 shows the variation in UCS after 7 and 90 days curing time with the KOH and olivine (KOH/olivine) mass ratio ranging between 0.39 and 1.7. Strength generally increased as the KOH/olivine weight ratio decreased. The maximum UCS level was obtained at the lowest KOH/olivine ratio of 0.39 for all curing periods. These findings are consistent with the existing literature, which report an increase in strength with a reduction in activator/source binder ratios [27, 29, 36, 48]. This might be explained by using the dry unit weight-liquid content relationship theory for soil stabilization. According to Table 4, the highest MDD and the lowest OWC were obtained when the KOH/Olivine ratio was at a minimum.

Fig. 4
figure 4

Influence of KOH/Olivine weight ratio on strength development at 7and 90 day

Full size image

SEM Characterization of Olivine Treated Soil in the Presence of KOH

As can be seen in Fig. 5a, b, the SEM analysis was performed on untreated soil and selected treated specimens that obtained the highest strength (KO20S) after a curing time of 90 days, based on pervious study by Fasihniukotalab et al. [15], the 20% of olivine has a highest strength among other dosages. As can be clearly seen in this Fig. 5b, the microstructure of treated soil consists of soil particles and the irregular shape of olivine particles in the presence of 10 M KOH after 90 day curing time. SEM images demonstrate a compact morphology without any major discontinuities, which is consistent with the mechanical properties observed (Fig. 3). This is because the addition of KOH to the olivine treated soil and reaction between MgO and SiO2 in the olivine which led to the production of Mg(OH)2. These results match those observed in earlier studies showing that using Mg(OH)2 in soil stabilization produces a dense surface structure for soil and the formation of Mg(OH)2 fills in the available pore space [5, 49, 50]. Accordingly the previous studies have shown that the produced gels were leached clay and Fly ash particles in soil and made strong bond among clay particles as a result of forming relatively dense alumino-silicate gels [51].

Fig. 5
figure 5

SEM images of a untreated soil (S) and b KO20S -sample after 90 day curing

Full size image

EDX Characterization of Olivine Treated Soil in the Presence of KOH

Figure 6a–d shows the EDX analysis of—KO5S, KO10S, KO15S and KO20S after 90 days of curing respectively. As shown in this figure, the most common phases observed contain the elements such as K, Si and Al, suggesting the formation of a silicate-activated gel as a result of the activation process. Table 5 lists the molar ratios of Si/Al, K/Al, Mg/Al and SiO2/K2O of the samples analysed under EDX. An increase in the strength of soils was observed with an increase in the Si/Al, K/Al ad Mg/Al ratios, which was in line with previous studies that showed higher strengths at Si/Al ratios ranging between 1.15 and 2.15 [28, 52,53,54]. This corresponded to a higher degree of geopolymerization because of the higher K/Al ratio. Moreover, K+ cations also played a role in balancing the net negative charge of Al3+ resulting from the Al:O coordination [9, 27, 55, 56].

Fig. 6
figure 6

EDX analysis of samples: a KO5S, b KO10S, c KO15S, d KO20S and e KS after 90 days curing

Full size image
Table 5 The EDX and UCS results of olivine treated soil in the presence of 10 M of KOH after 90 days curing
Full size table

XRD Characterization of Olivine Treated Soil in the Presence of KOH

Figures 7 and 8 show the XRD of untreated soil and 20% olivine treated soil after 90 days of ambient curing. The roughly peaks of 11.5 and 24.5 for kaolinite and of 21, 26, 36, 40 and 44.5 for quartz are identified at untreated soil. Furthermore, the roughly peaks of (18, 23 and 39) shows the brucite, (21, 27, 51 and 68) shows the quartz, serpentine identified as peak of 13, (26, 37 and 60) shows the mullite, and hematite (55) are identified as a result of 20% olivine dissolution through reaction with the KOH present in the soil. One important conclusion that may be drawn from the observation of the diffractogram in Fig. 8 is that when KOH is added to the olivine treated soil, the formation of Mg(OH)2, SiO2, and mullite can be observed. As can be seen in Fig. 7, peaks at 2-theta values of 20, 26, 34, 50 and 67 indicate the presence of quartz (SiO2), and those at 18, 21, 36 correspond to magnesium hydroxide Mg(OH)2 Moreover, peaks of 12 and 54 indicate serpentine and hematite respectively in the XRD [27, 49, 57].

Fig. 7
figure 7

The XRD of untreated soil (K kaolinite, Q quartz)

Full size image
Fig. 8
figure 8

The XRD of KO20S after 90 day curing time (B brucite, H hematite, M mullite, S serpentine, Q quartz)

Full size image

Conclusions

This study has investigated the role of alkali activated olivine treated soil. The results indicate that olivine can be used as a sustainable precursor in alkali activated treated soil. The following conclusions can be drawn:

  • The strength of olivine treated soil in the presence of KOH increased with olivine content and curing time, reaching a value up to 7.4 MPa after 90 day curing. In this respect, the duration of curing has a direct effect on the amount of activated reactants transformed into the binding products. This high strength development implies the dissolution of olivine through the addition of KOH. XRD diffractograms and SEM images confirmed the formation of Mg(OH)2 as a result of the hydration of magnesium in olivine treated soil.

  • The dissolution mechanism of olivine and the influence of Si/Al, K/Al and SiO2/K2O ratios of alkali activated systems led to an increase in strength as confirmed by EDX results.

  • This study has identified the successful use of olivine for soil stabilization in the presence of strong alkaline activators like KOH.