A Review of Physicochemical Stabilization for Improved Engineering Properties of Clays

Abstract

Severe climatic and environmental conditions warrant the use of stabilization agents in aid of compaction for sustainable improvement in engineering properties of clays. Physicochemical agents are a viable option because they are cost effective, environmentally friendly, and offer improved long-term performance of treated soils. This research developed a fundamental understanding of the clay–water–electrolyte admixtures relations. Based on a comprehensive literature review, the effect of nanomaterials, biopolymers, and geopolymers on the behavior of compacted clays was investigated. It was found that all of these admixtures facilitate the development of an aggregated soil microstructure through unique mechanisms. Biopolymers have the highest water adsorption capacity followed by geopolymers and then by nanomaterials. The effect of admixtures on optimum compaction properties follows a decreasing trend similar to untreated clays (S = 80% ± 20%). The variation of hydraulic conductivity, compression index, and compressive strength are largely within the family of curves identified by typical relationships for compacted clays. These preliminary findings indicate that not all engineering properties are improved to the same level by the different types of physicochemical admixtures. The specific nature of geotechnical engineering (soil type and site conditions) as well as the wide range of admixture types and potential biodegradation of some of the reagents are the major shortcoming of using this class of materials.

1. Introduction

The geotechnical engineering properties (such as flow through, volumetric changes, and shear strength) of clayey soils are marginal and may not be able to ensure the integrity of civil infrastructure systems. Municipal facilities, such as buried pipelines, road networks, and sanitary landfills, are particularly vulnerable because of urban sprawl and societal demands. Compaction can improve soil behavior although field performance may change over the lifetime of earthen structures. Such facilities fail to perform their intended purpose, especially in Canada, where clays are the primary construction material, and the constructed facilities are exposed to harsh climatic conditions (extreme seasonal weather and vulnerability to the impacts of global warming) as well as environmental conditions (annual cycles of de-icing salts and frequent oil spills) [1,2]. For example, leachate percolated from an unlined municipal landfill and resulted in contamination of the Condie aquifer, Regina, Saskatchewan, over the course of 18 years [3,4]; about 900 mm differential settlement of the clay core in Saint-Marguerite–3 dam at Côte-Nord, Quebec, due to consolidation over 10 years post construction [5]; and a 27° tilt of the Transcona grain elevator in Winnipeg, Manitoba, due to low compressive strength one month after construction [6]. Similar distress and damage issues in urban infrastructure systems are routinely reported from across the globe raising questions about the capability of using conventional methods of soil improvement. To develop superior materials for improved geotechnical performance, there is an exigent need to investigate emerging stabilization admixtures in aid of compaction.

Table 1 provides a summary of the various classes of stabilization admixtures commonly used in compaction of clayey soils. Physical admixtures (inert materials) have limited success because these materials rely on mechanical improvement requiring large dosages and are affected by seasonal climatic variations [7,8]. In contrast, chemical admixtures are mostly Ca²⁺-based, which improve soil properties through chemical interactions with the various clay mineral species present in the soil. These types of admixtures are used in moderate dosages but can still harm the environment and may not be suitable for sulfate-rich soils [9,10,11,12].

The table indicates that physicochemical admixtures are a viable alternative for stabilization of clays because they are cost effective, owing to small dosages of up to 2% [11] at ambient field temperatures [13,14], and environmentally friendly because the admixtures are mostly inert, recycled from waste residues, and expected to be durable [11,15]. For this class of admixtures, complex physicochemical interactions occur because the clays are net negatively charged and have large specific surface areas whereas the admixtures have high hydrophilicity, gelation ability, variable charge density, and long polymeric chains [16,17,18]. The reported data is usually developed for a selected admixture at the optimum dosages for modifying a specific clay at a given site. Therefore, there is a lack of understanding related to the underlying micro-level mechanisms of improvement in soil behavior.

Table 1. Summary of stabilization admixtures commonly used in the compaction of clays.

Admixture and Reference	Mechanism and Dosage (Dry Mass Basis)	Limitations
Physical
Sand [7,8]	Addition of inert and coarse particles (10–90%)	- Susceptible to harsh climate and environment - May require high dosages for effective improvement
Shredded tire [19,20]	Addition of inert and fine-to-coarse lightweight particles (≤10%)	- Reduced density due to flexibility of rubber - High compressibility (soil re-configuration) due to large grain size range
Chemical
Cement [11,21,22,23,24]	Three stages comprising cation exchange (Na+ and Mg2+ in clay replaced by Ca2+ of cement); cementitious hydration (Ca2+-based compounds in cement react with water to form silicates, aluminates, and hydrated lime); and pozzolanic reaction (Ca2+ on a clay surface react with dissolved silicates and aluminates to form gels) (≤20%)	- Increased brittleness for higher dosages - Incompatible with sulphate-rich soils due to formation of expansive ettringite - Incompatible with organic soils (low pH) that hinders pozzolanic reactions
Lime [24,25]	Four stages with the first three similar to the above (albeit Ca2+ derived from lime) followed by carbonation cementation where CaO reacts with atmospheric CO2 to precipitate as CaCO3 (≤8%)	- Same issues as with cement - Decreased durability due to leaching of hydrated lime
Fly ash [22,23,26,27]	Accelerates cement or lime stabilization (≤20%)	- Same issues as with cement - Low self-cementation and may require primary admixture/activator - Leaching of toxic trace elements into groundwater
Cement kiln dust [28,29]	Same as lime due to abundance of free lime (≤8%)	- Same issues as with lime
Steel slag [30,31,32,33,34,35]	Cementitious hydration (albeit Ca2+ derived from lime) similar to cement (≤25%)	- Prone to swelling due to long-term hydration of free MgO - Leaching of toxic trace elements into groundwater
Silica fume [36,37,38]	Silica accelerates the pozzolanic reaction (≤50%)	- Same issues as with cement - Cannot be used as a primary admixture
Physicochemical
Nanomaterials [39,40,41,42]	Development of viscous gel due to water adsorption through H-bonding that coats the clay surfaces and reduces the diffuse double-layer thickness (≤3%)	- Negligible adsorption onto clay surfaces for neutral nanomaterials only - May leach toxic trace elements into the groundwater
Biopolymers [14,16,18]	Adsorption through electrostatic attraction and development of hydrogel (≤2%)	- Prone to biodegradation
Geopolymers [11,15,43,44,45,46]	Activation of aluminosilicate source by aqueous alkaline solution via geopolymerisation reaction to form geopolymeric gel (source: 10–20%) (activator/source: 0.4–0.8)	- May lead to poor workability due to high alkalinity

Table 1. Summary of stabilization admixtures commonly used in the compaction of clays.

Admixture and Reference	Mechanism and Dosage (Dry Mass Basis)	Limitations
Physical
Sand [7,8]	Addition of inert and coarse particles (10–90%)	- Susceptible to harsh climate and environment - May require high dosages for effective improvement
Shredded tire [19,20]	Addition of inert and fine-to-coarse lightweight particles (≤10%)	- Reduced density due to flexibility of rubber - High compressibility (soil re-configuration) due to large grain size range
Chemical
Cement [11,21,22,23,24]	Three stages comprising cation exchange (Na+ and Mg2+ in clay replaced by Ca2+ of cement); cementitious hydration (Ca2+-based compounds in cement react with water to form silicates, aluminates, and hydrated lime); and pozzolanic reaction (Ca2+ on a clay surface react with dissolved silicates and aluminates to form gels) (≤20%)	- Increased brittleness for higher dosages - Incompatible with sulphate-rich soils due to formation of expansive ettringite - Incompatible with organic soils (low pH) that hinders pozzolanic reactions
Lime [24,25]	Four stages with the first three similar to the above (albeit Ca2+ derived from lime) followed by carbonation cementation where CaO reacts with atmospheric CO2 to precipitate as CaCO3 (≤8%)	- Same issues as with cement - Decreased durability due to leaching of hydrated lime
Fly ash [22,23,26,27]	Accelerates cement or lime stabilization (≤20%)	- Same issues as with cement - Low self-cementation and may require primary admixture/activator - Leaching of toxic trace elements into groundwater
Cement kiln dust [28,29]	Same as lime due to abundance of free lime (≤8%)	- Same issues as with lime
Steel slag [30,31,32,33,34,35]	Cementitious hydration (albeit Ca2+ derived from lime) similar to cement (≤25%)	- Prone to swelling due to long-term hydration of free MgO - Leaching of toxic trace elements into groundwater
Silica fume [36,37,38]	Silica accelerates the pozzolanic reaction (≤50%)	- Same issues as with cement - Cannot be used as a primary admixture
Physicochemical
Nanomaterials [39,40,41,42]	Development of viscous gel due to water adsorption through H-bonding that coats the clay surfaces and reduces the diffuse double-layer thickness (≤3%)	- Negligible adsorption onto clay surfaces for neutral nanomaterials only - May leach toxic trace elements into the groundwater
Biopolymers [14,16,18]	Adsorption through electrostatic attraction and development of hydrogel (≤2%)	- Prone to biodegradation
Geopolymers [11,15,43,44,45,46]	Activation of aluminosilicate source by aqueous alkaline solution via geopolymerisation reaction to form geopolymeric gel (source: 10–20%) (activator/source: 0.4–0.8)	- May lead to poor workability due to high alkalinity

The objective of this paper was to investigate physicochemical stabilization for improved engineering behavior of clays. Initially, the chemical composition of admixtures (nanomaterials, biopolymers, and geopolymers) and colloid–water interactions are separately presented. Next, colloid–water admixture interactions and the resulting microstructure development are discussed. Finally, published data were compiled to compare the improvement in engineering properties of clays for each admixture.

2. Background Review

2.1. Composition of Physicochemical Admixtures

Figure 1 shows schematic diagrams of the composition of physicochemical admixtures. Nanomaterials (Figure 1a) are inert metal oxides, such as silica dioxide (SiO₂), copper oxide (CuO), and aluminum oxide (γ-Al₂O₃). The crystal structure is usually face-centered cubic (FCC) comprising closely packed arrays of O²⁻ and metal cations occupying the interstitial sites [47,48]. The nanoparticles are 1–100 nm, which result in high surface area to volume ratio that facilitates coating of the soil colloids [49]. When water is added, these materials exhibit viscous gelling due to water adsorption, thereby infilling the soil voids [50,51].

Biopolymers (Figure 1b), with polysaccharides being the most abundant, are long-chain organic molecules with the general formula (C₆H₁₀O₅)_n where 40 ≤ n ≤ 3000. Cellulose is derived from plants and has glucose units linked together by β-(1–4)-glycosidic bonds. In contrast, chitin is obtained from animals and comprises β-(1–4)-2-acetamido-2-deoxy-D-glucose bonds. Chitosan is an N-deacetylated derivative of chitin with β-(1–4)-2-amino-2-deoxy-D-glucose bonds that form when the fraction of acetamido groups is less than 50% [52]. The chemical composition (charge density, molecular weight, degree of branching) governs adsorption of biopolymers onto soil colloids [16,17]. When water is added, biopolymers promote a conglomeration of colloids and infilling of voids with hydrogel [53].

Geopolymers (Figure 1c) are inorganic molecules comprising of ${\mathrm{S}\mathrm{i}\mathrm{O}}_4^{-}$ and ${\mathrm{A}\mathrm{l}\mathrm{O}}_4^{-}$ tetrahedra bonded together through the oxygen atoms to form an amorphous or a semi-crystalline structure through the geopolymerisation reaction. This chemical reaction mainly depends on the Si:Al ratio, such that an increase of up to 2:1 increases the number of relatively strong Si-O-Si bonds (siloxo) [54,55]. However, further increase in the Si:Al ratio may be redundant for the chemical process [56]. When water is added, a series of chemical reactions occur culminating in the formation of a geopolymeric gel. As before, the gel fills the void spaces within the soil matrix and enhances cementation that, in turn, improves with time.

Figure 1. Schematic of chemical composition of physicochemical admixtures: (a) nanomaterials (after [48,57,58]); (b) biopolymers (after [52]); and (c) geopolymers (after [54]).

Figure 1. Schematic of chemical composition of physicochemical admixtures: (a) nanomaterials (after [48,57,58]); (b) biopolymers (after [52]); and (c) geopolymers (after [54]).

2.2. Interaction of Clays with Water

Figure 2 gives the interactions in a clay–water system. A diffuse double layer (DDL) of ions is formed around negatively charged colloidal surface (Figure 2a). The cations in water are strongly bound to the colloid by electrostatic attraction to form a Stern layer whereas the outer mixed-movable ions form a diffuse layer. Due to ionic movement, a shear surface is created between the Stern layer and the diffuse layer. The thickness of the DDL (1/K) is calculated according to the following equation [59]:

$$\frac{1}{K}={\left(\frac{{\epsilon }_{0}kDT}{2n_0{e}^2{v}^2}\right)}^{0.5}$$

(1)

In the above equation, ${\epsilon }_0$ is the permittivity of vacuum (8.85 × 10⁻¹² C²/J m) (k is Boltzmann’s constant (1.38 × 10⁻²³ J/K), D is the dielectric constant of the medium, T is temperature (K), $n_0$ is electrolyte concentration (mol/l), $e$ is an electronic charge (1.6 × 10⁻¹⁹ C), and $v$ is the cation valence. This equation means that the DDL thickness (1/K) decreases with an increase in electrolyte concentration in the solution. Likewise, an increase in cation valence decreases (1/K): Trivalent cations (Al³⁺ and Fe³⁺) result in the thinnest, followed by divalent cations (Ca²⁺ and Mg²⁺), and then by monovalent cations (Na⁺ and K⁺). The combined effect of DT on (1/K) is insignificant [59]. The electrical potential decreases with the distance away from the colloidal surface where it is maximum, to a lower non-measurable value at the Stern layer surface and then to a minimum measurable value (zeta potential (ζ)) at the shear surface between the Stern layer and the diffuse layer, and reaches a zero value at the free water surface [60].

Figure 2b gives potential energies of colloidal interactions, as described by the Derjaguin, Landau, Verwey, and Overbeek (DLVO) theory [61,62]. This theory uses Van der Waals attractive forces and electrostatic repulsive forces to explain the behavior of two adjacent colloids. The net energy curve depends on the distance between the two colloids; attraction is represented when the curve is plotted below the line of zero charge and repulsion in the opposite case. These colloidal interactions govern particle assemblages, as identified by the following distinct terms [63,64]: (i) Flocculation pertains to large colloidal clusters formed when the attraction forces (as affected by D and T) are dominant to develop thin DDL; and (ii) dispersion is when colloids are detached from one another when repulsion forces (as affected by $n_0$ and $v$) are prevailing to develop thick DDL. Such interactions can be detected by ζ because its value decreases as the DDL thickness decreases [65].

2.3. Interaction of Admixtures in Clay–Water System

Figure 3 shows the development of microstructure in clay–water admixture system. Nanomaterials (Figure 3a) adsorb water through hydrogen bonding to develop a viscous gel that, in turn, adheres to and partly coats the clay surfaces [49], thereby coexisting with the Stern layer [66]. The adhesive effect of the viscous gel gradually increases with an increase in nanomaterial concentration to reduce DDL thickness [67]. This results in the following mechanisms: (i) isolating the negative charges of the clay to decrease the repulsive forces between adjacent colloids; and (ii) increasing the contact area of the clay to increase van der Waals attractive forces between adjacent colloids. At the bulk level, an aggregated microstructure with reduced pore sizes is formed within the clay clods. This mechanism pertains to inert nanoparticles, which negligibly adsorb onto clay surfaces [41,42]. When the nanoparticles possess ionic charges, electrostatic interactions occur within the system to enhance adsorption onto clay surfaces [40].

Biopolymers (Figure 3b) interact with clay surfaces through electrostatic attraction thereby facilitating adsorption of polymer chains onto colloidal surfaces [16,69,70]. The adsorption process occurs between charges on the biopolymer and the negatively charged clay surfaces, that is, through the positive amine group for cationic biopolymers, via counter-ions in solution to engage the carboxyl group (COOH⁻) for anionic biopolymers, and by the dipole attraction of the hydroxyl group (OH) for nonionic biopolymers [18,71]. Furthermore, adsorption occurs through the following mechanisms: (i) bridging of the loops and tails of the adsorbed polymer chains to adjacent colloids; and (ii) charge patching of free segments of polymer chains with exposed sites on partially covered clay surfaces. Bridging results in strong particle assemblages and is dominant when the biopolymer has high molecular weight (long-chain polymers that can stretch out to attach to several colloids), low charge density (low number of functional groups on the chains, thereby facilitating stretching out and attaching to several colloids), and low dosage (low repulsion between chains facilitates adsorption onto several colloids). In contrast, charge patching results in weak particles’ assemblage and is favored by the opposite polymer parameters [16,72,73]. In addition to the above, a hydrogel (a network of cross-linked polymer chains retaining part of the pore water) is formed and infills the space within the newly formed clods [74,75]. As before, an aggregated clay fabric is formed with reduced pore sizes [76,77,78].

Geopolymers (Figure 3c) are products of the chemical reaction between an aluminosilicate source and an aqueous alkaline solution, such as NaOH and KOH. This reaction occurs in five stages: dissolution, speciation equilibrium, gelation, reorganisation, and polymerisation. Upon mixing the source with the alkaline solution, a high pH solution is formed consuming water and causing dissolution of the source into aluminate (${\mathrm{A}\mathrm{l}\left(\mathrm{O}\mathrm{H}\right)}_4^{3-}$) and silicate (${\mathrm{S}\mathrm{i}\left(\mathrm{O}\mathrm{H}\right)}_4^{2-}$). This hydrolysis reaction continues until the concentration of the two species are equilibrated. When the solution becomes saturated with the ions, gelation (sharing of corner oxygen between aluminate and silicate tetrahedra to form aluminosilicate) occurs and results in the formation of large polymer networks. This condensation reaction releases water and allows the gel system to undergo rearrangement due to increased connectivity and eventually form a complex 3-D aluminosilicate molecular structure. Further hardening (polymerization) of the gel system results in the formation of geopolymers (sodium/potassium aluminosilicate hydrate). Again, the geopolymers provide time-dependent cementation leading to aggregated clods with reduced pore sizes [77,79,80].

3. Soil Properties

Figure 4 gives the reported values of soil consistency for treated clays on the plasticity chart [81] with typical mineral ranges given by [82]. Nanomaterial-treated soils fall in the regions designated as low plastic clay (CL) and low plastic silt (ML) because of their minimal adsorption onto colloidal surfaces [41,42]. In contrast, biopolymer-treated soils fall mostly in the regions of clay with high plasticity (CH) and silt with high plasticity (MH). This is attributed to the high water-adsorption capacity of biopolymers [75]. Geopolymer-treated soils are on both sides of the 50% liquid limit and in the regions of ML and MH. This is due to the released water from geopolymerisation that is adsorbed by the clay [79,80]. Generally, the behavior of treated soils can be viewed to be similar to clay minerals with low water adsorption (kaolinite and chlorite) to moderate water adsorption (illite and halloysite).

Figure 5 shows reported maximum dry unit weight (γ_dmax) and optimum water content (w_opt) of treated clays with a theoretical degree of saturation (S) lines based on specific gravity (G_s) = 2.70. No discernible trends are observed to distinguish between soils treated with the various admixtures. Nonetheless, the data fitted linearly (R² = 0.75) within one standard deviation on each side of the fit. The decreasing γ_dmax trend with increasing w_opt fell around S = 80% ± 20% and closely matched with the line of optimum (S = 75%) for compacted clays [82]. This means that treated clays are comparable to untreated soils because compaction and/or loading can suppress the effect of an initial microstructure derived from a physicochemical interaction [83]. Nonetheless, this figure should be understood in light of the entire compaction curve.

Figure 4. Plasticity chart showing physicochemically treated clay with typical mineral ranges from [82] (data from [84,85,86,87,88,89,90,91,92,93,94]).

Figure 4. Plasticity chart showing physicochemically treated clay with typical mineral ranges from [82] (data from [84,85,86,87,88,89,90,91,92,93,94]).

Compaction pertains to the densification of soil by means of mechanical energy that expels air from the pore spaces. With increasing water content, the compaction curve shows an initial increase in the dry unit weight, reaches a maximum value, and is followed by a decrease [82,101]. The γ_dmax and w_op vary in response to soil type, admixture type, and compaction energy [102]. The dry of optimum is characterized by an aggregated morphology with dual porosity comprising large voids (radii = 0.4 µm) between clay clods and small pores (radii = 0.3 µm) within clay clods. In contrast, the wet of optimum is defined by a dispersed microstructure with homogenous porosity and uniform size pores (radii = 0.5 µm) in the clay matrix. These changes in soil fabric govern the variation in the engineering properties of the compacted materials. Generally, compacted soils on dry of optimum are characterized by high hydraulic conductivity, high swell–shrink potential, high compressibility (at high stress levels), and high shear strength whereas the opposite is true on wet of optimum [82]. This means that the optimum conditions represent a transition in soil fabric that is related to the type of physicochemical additive used for stabilization [83,103,104].

Figure 6 shows geotechnical properties of treated clays with respect to the initial dry unit weight (γ_d) and water content (w). The reported data showed extensive scatter because of the large variation in soil types, admixture types, admixture dosages, and sample preparation. Likewise, the data do not necessarily pertain to the same treated soil because of not being readily available. Nonetheless, these data plotted within the family of curves identified by the typical relationships for clayey soils (Table 2).

Theoretically, hydraulic conductivity decreases with increasing the dry unit weight because of the reduction in pore sizes that, in turn, results in increased tortuosity (ratio of the convoluted pore paths to the straight path) and dead ends [105]. Likewise, the compression index decreases because a reduced void ratio does not facilitate any further reduction in the soil volume [101]. Finally, the compressive strength increases because the denser soil offers more shear resistance due to increased interlocking [82]. In contrast, hydraulic conductivity mostly decreases with increasing water content owing to a more homogeneous/oriented soil fabric with larger water films around particles, thereby reducing pore sizes [106]. Conversely, the compression index increases because particle swelling increases the initial soil volume that can be compressed to a greater degree [107]. Finally, the compressive strength decreases due to the reduction in shear resistance as a result of reduced particles interlocking [82].

Figure 6. Geotechnical properties of physicochemical admixtures-treated clays: (a) hydraulic conductivity; (b) compression index; and (c) compressive strength (data from [80,85,86,88,90,91,92,94,97,100,108,109,110,111,112,113,114,115,116,117,118]).

Figure 6. Geotechnical properties of physicochemical admixtures-treated clays: (a) hydraulic conductivity; (b) compression index; and (c) compressive strength (data from [80,85,86,88,90,91,92,94,97,100,108,109,110,111,112,113,114,115,116,117,118]).

Table 2. Typical soil property relationships with respect to dry unit weight and water content.

Soil Property and Reference	Relationships
Relationship with dry unit weight	-
Hydraulic conductivity [119]	- ln ks = −16.91${\gamma }_d$+15.16
[120]	Best-fit for the reported data
Compression index [121]	- $C_c=0.5{\left({\gamma }_{w/{\gamma }_d}\right)}^{2.4}$
[122]	$C_c=0.41{\gamma }_d+0.85$
Compressive strength [123]	- $q_u=18.282{{\gamma }_d}^{5.099}$
[7]	Best-fit for the reported data
Relationship with water content	-
Hydraulic conductivity [124]	- Best-fit for the reported data
[82]	Best-fit for the reported data
Compression index [125]	- $C_c=0.007w_L-7$
[126]	$C_c=1.35I_P$ (Gs = 2.70)
Compressive strength [7]	- Best-fit for the reported data
[127]	Best-fit for the reported data

Table 2. Typical soil property relationships with respect to dry unit weight and water content.

Soil Property and Reference	Relationships
Relationship with dry unit weight	-
Hydraulic conductivity [119]	- ln ks = −16.91${\gamma }_d$+15.16
[120]	Best-fit for the reported data
Compression index [121]	- $C_c=0.5{\left({\gamma }_{w/{\gamma }_d}\right)}^{2.4}$
[122]	$C_c=0.41{\gamma }_d+0.85$
Compressive strength [123]	- $q_u=18.282{{\gamma }_d}^{5.099}$
[7]	Best-fit for the reported data
Relationship with water content	-
Hydraulic conductivity [124]	- Best-fit for the reported data
[82]	Best-fit for the reported data
Compression index [125]	- $C_c=0.007w_L-7$
[126]	$C_c=1.35I_P$ (Gs = 2.70)
Compressive strength [7]	- Best-fit for the reported data
[127]	Best-fit for the reported data

Given the empirical nature of geotechnical engineering, the published empirical equations or best-fits based on reported data (given in Figure 6 and Table 2) should also be considered in the context of variability in soil type (clay content and mineralogy) and liquid composition (use of distilled or deionized water) as well as test settings (initial and boundary conditions, stress paths, temperature, and humidity). Furthermore, these behavioral curves represent the most common correlations (based on regression analysis of experimental data using linear, exponential, and power functions) used by geotechnical practitioners for the prediction of soil properties; not all empirical equations are given in this paper. In essence, these general trends provide a preliminary assessment of typical changes in untreated soil behavior in response to changes in dry unit weight and water content. Consequently, the relationships are useful in understanding potential improvements in the engineering behavior of compacted clays and in selecting appropriate admixtures for modifying a desired soil property.

4. Summary and Conclusions

The long-term maintenance of aging civil infrastructure networks across the globe necessitates the use of sustainable admixtures during soil compaction. By virtue of being cost-effective and environmentally viable, physicochemical agents offer a viable alternative to the traditional physical and chemical additives. However, the scarcity of published work related to this relatively new class of geomaterials is a major hurdle in this emerging research area and, as such, to the development of innovative methods for stabilization of clayey soils.

The main contribution of this research is to develop a fundamental understanding of the clay–water–electrolyte admixtures relations at phase boundaries with specific focus on physicochemical admixtures. Based on a comprehensive literature review, the effect of nanomaterials, biopolymers, and geopolymers on the engineering behavior of compacted clay was investigated. The findings of this research are summarized as follows: (i) All of the physicochemical admixtures facilitate the development of an aggregated soil microstructure, albeit through unique mechanisms; (ii) biopolymers have the highest water adsorption capacity, followed by geopolymers, and then by nanomaterials; (iii) the effect of admixtures on optimum compaction properties follows a decreasing trend similar to untreated clays (S = 80% ± 20%); and (iv) bulk geotechnical properties of hydraulic conductivity, compression index, and compressive strength are within the family of curves identified by the typical relationships for clay soils. These preliminary findings demonstrate that not all engineering properties are improved to the same level by the different types of physicochemical admixtures

Physicochemical admixtures are a new class of admixtures that can offer a sustainable solution to improve soil properties during compaction. However, this class of materials faces shortcomings that need to be addressed before these can be recognized by the geotechnical engineering community. There is a general lack of standardization with respect to nomenclature and product specification in the industry. The hesitance by the manufacturers of these products in releasing proprietary information is the major drawback with respect to understanding their interaction with clay particles. Furthermore, generalization of soil behavior is difficult owing to the large range of materials within each of the sub-groups. Clearly, research and development are in the infancy stage and quite specific to local soil and site conditions. Over time, it is important that a standard code of practice is developed for the use of this class of admixtures during compaction. In addition, biodegradability of some of the admixtures may affect the durability of a stabilized surface and buried structures. Such a code should include the various factors affecting the rate of biodegradation, such as polymer type, molecular weight, microorganisms’ type, and incubation conditions.