Fibers, Geopolymers, Nano and Alkali-Activated Materials for Deep Soil Mix Binders

Ordinary Portland Cement (OPC) and Lime (CaO) have traditionally been used as binder materials for Deep Soil Mix (DSM) ground improvement. Research has been conducted into possible alternatives such as pozzolans to reduce reliance on either cement or lime. However, pozzolans still undergo similar calcium-based reactions in the strengthening process. In this review, further alternative binder materials for soil strength development are explored. These recent developments include fiber reinforcement materials, alkali activation methods, nanomaterials and geopolymers, which can potentially achieve equal or improved performance. Research to date has shown that alkali-activated materials and geopolymers can be equivalent or superior alternatives to pozzolanic supplemented cement binders. The case is made for GP cements which potentially produces 80% less CO2 than conventional portland cement during manufacture. One-part AAM and GP cements are a promising substitute for portland cement in DSM. A combined approach which incorporates both Ca and alkali activated/geopolymer types of materials and hence reactions is proposed.


Introduction
The Deep Soil Mix (DSM) method applies soil stabilization principles, which comprises inserting binder materials with other fillers and mixing together with the soil to form strengthened columns of treated soil below ground. The manufacture of the predominant binder materials, cement and lime, impose significant CO 2 emission and high energy demands. Although studied as a potential supplement and/or partial replacement to reduce the usage of OPC and lime, pozzolanic materials still rely primarily on similar calcium (Ca) reaction processes to produce the same calcium silicate hydrate (C-S-H) and calcium aluminate hydrate (C-A-H) gel products for treated soil compressive strength improvement. In addition, pozzolanic binders react incompletely and require a longer time to realize improvements in the treated soil. This paper reviews several alternatives to Ca type reactions to improve soil properties for DSM, which offer different pathways for strength / soil properties improvement. The primary soil properties of interest for improvement are 1) shear strength (SR u R); 2) Compressive strength (qR u R); 3) Stiffness (Young's Modulus, E); 4) Dynamic properties (G, D) and 5) Permeability (k). 832 compressive strength was determined by the primary chemical stabilizer. Dispersion of fibers influenced UCS as it may have introduced pockets of weakness/failure planes [8]. Arasan et al. (2015) studied the feasibility of utilizing polymers as a binder for rapid stabilization of sandy soils with DSM by conducting UCS tests on different ratios of polyester-soil mix. UCS was observed to increase with increasing polyester ratio, effective diameter, relative density and curing period. The highest UCS achieved was with 0.6-1.18 mm grain size, 30% polyester ratio and 40% relative density respectively [9].
Laboratory studies testing on fiber reinforced soil cement (FRSC) samples of clay soils in Bangkok, Thailand were conducted to investigate effect on flexural performance (bending strength @ L/150 deflection, 150 ) improvement [10]. It was observed that PP fiber performed better than steel fibersthe fully crimped shape of PP fibers achieved greater bond strength than straighter steel fibers. Hence, brittleness in cement-based binders in deep soil mix columns could be improved with fiber reinforcement which increases toughness. Some recent research is listed below:

Reaction Mechanism
Nanomaterials are defined as physical substances with at least one dimension from 1 to 150 nm (1 nm = 1 × 10 −9 m) [16]. Nanomaterials can be produced through the following approaches:  "Top Down" approach whereby larger particles are scaled down in size to nano range of dimensions whilst maintaining original properties without resorting to "atomic scale level of control or manipulation (e.g. electronics industry miniaturization of computer chips). Nano scale versions of pozzolans can be produced mechanically with pulverizing techniques using high energy ball milling [17];  "Bottom Up" approach (known also as molecular nanotechnology) whereby materials are assembled from singular atoms or molecules.
Nanomaterials have been proven to enhance the performance of concrete [18,19] e.g. with the addition of nanosilica (nS), concrete compressive strength can increase by up to 30% [20]. Improvements in the flexural strength and elasticity (Young's modulus, E) can be achieved by introducing Carbon nanotubes (CNT) [21].
The application of nano materials has been considered as a potential binder material in soil improvement [2,22]. The primary characteristics from nanomaterials that induce soil improvement can be described below [23]:  Particle size -As the particle size approaches or decreases to less than light wave wavelength (De Broglie wave), the periodic boundary condition of the particle is destroyed, and surface atomic density decreases, leading to different physical properties from that at the micro or macro scale;  Microstructurewhich can be categorized into 2D structured nano tubes and 3D structured nano particles. The two categories induce different improvement effects on the soil;  Particle surface arealikewise, the specific surface area of the particle increases, which leads to greater ion exchange capacity and increase interaction with other particles.
Furthermore, since the nanomaterial will need to be mixed in some form of solution prior to injection into the soil, the rheological suspension properties in solution will have also an influence on dispersion and enhancement effect into the soil. Nanomaterials applications and understanding of it is still in its infancyit has only been recently researched as a potential additive for concrete and many developments would still be under commercial consideration [21]. Some examples of recent research are compiled in Tables 2 to 4.

Carbon Nanotubes (CNT)
Carbon nanotubes (CNT) are tube-shaped carbon materials where the diameter of the tube is measured in the nanometer scale [16]. CNTs can be up to several millimeters in length and can be arranged in one-layer walls (known as Single-wall carbon nanotubes -SWCNT) or multi-wall structures (known as Multi wall carbon nanotubes -MWCNT). In general, CNTs have 5 x Young's modulus (E) and 8 x the strength of steel at 1/6 the density of steel. This strength is due to the covalent sp 2 bonds formed between the individual carbon atoms. Thus, there can be potential gains in increased flexural strength and improved control of crack propagation with CNT enhanced concrete and by extension, to soil treated with a combination of a primary binder (e.g. cement) and CNT. Due to the size and fineness of CNTs, it is possible to distribute/disperse on a finer scale within concrete than for micro reinforcement fibers, thus CNT infused composites can achieve greater compressive and flexural strengths. A comparison on flexural and compressive strengths achieved from the research by Kahidan and Shirmohammadian [25] showed up to 19% compressive and 25% flexural improvement. This is supported by Sáez de Ibarra et al. [26] who reported 80-90% increase in E concrete with the addition of only 0.10-0.20% MWCNT/SWCNT by cement weight and Cwirzen et al. [27] reporting 50% improvement in compressive strength with 0.045 -0.15% MWCNT by cement weight.
Direct addition of nanocarbon filaments to sandy soil mixed with bentonite in proportions ranging from 0.05 to 0.2% of total dry weight of soil resulted in dry density increase and subsequent decrease in soil shrinkage / expansive strains [28,29]. This is compared with fiber reinforcement addition to soil which also reduced soil cracks but increased hydraulic conductivity. It is suggested by Taha et al. [30] that the better dispersion of the CNT additives would provide the potential to overcome these desiccation cracks in soil associated when using fiber reinforcement and binder.
The potential for utilizing CNT together with binder material as deep soil mixing supplement has not been explored at depth. The present high expense of carbon nanotubes may have a deterring effect from the application in this field [21]. In addition, challenges faced with consistent dispersion of the CNT needs to be overcome when applied to soil stabilization [30]. Correia and Rasteiro [31] reported adding a surfactant to the blended solution of MWCNT/ OPC in improving the dispersion and achieving higher UCS and Youngs Modulus (E).

Nano Titania (nT iOR2R) / Nano Fly Ash (nFA)
Nano Fly Ash enhances the activation of pozzolanic reaction and further densifies concrete like its non-nano counterpart [32]. Babu and Joseph [33] conducted research to study the effect of Nano Titanium Dioxide (nTiO2) and Nano Fly Ash (nFA) on properties of soft soil (Silty clay of USCS classification CH-MH). With increasing nano material additive (from 0% to 2%), the following improvements were observed:  Atterberg limitsdecreased LL;  Compactionincreased MDD and decreased OMC;  Strengthincreased shear strength (and consequently, UCS);  CBR and Settlementimproved CBR values and reduced settlement.

Nano Clay (nC)
Nano Clays (nC) can be defined as nanoparticles of layered mineral silicates also commonly known as phyllosilicates. Some Nano Clays are:  Montmorillonite Nano Claysthis is the most common Nano Clay, consisting of ~ 1 nm thick alumino silicate layers with lateral dimensions ranging from few hundred nM to ~ 10 μm, combined into large stacks;  Halloysite Nano Claysnaturally occurring aluminosilicate nanotube with a wall thickness of 10-15 atomic aluminosilicate sheets, an outer diameter averaging 50 nM, an inner diameter of 12-15 nM, and length of 0.5-1 μm. Ali et al. [35] studied the potential use of nC (montmorillonite) to increase the strength of contaminated soil. Soil contaminated with kerosene experienced reduction in UCS down to 61% of original UCS depending on kerosene %. However, the addition of nC up to 2% for contaminated soil and 1% for uncontaminated soil increased the bearing capacity of the soil. Increase in LL, PL and shear strength of silty soils was observed by Bahari et al. [36] with the addition of up to 2% by weight nC (montmorillonite) to silt. A linear correlation for shear strength was observed from direct shear tests conducted.

Nano Silica (nSi)
Nano Silica (nSi) is silicon dioxide in nano particle size form. It can directly be produced from bio waste e.g. rice husk ash (RHA) aerosol-gel method, electric arc method, precipitation method etc. [39].
Chemically, nSi particles would react with calcium hydroxide (Ca (OH) 2 ) to develop more of the strength contributing C-S-H material in concrete. In addition, concrete workability, increased resistance to water penetration and reduced leaching of calcium has also been encountered with nSi. The effective surface area plays a large part in influencing the improvement and nSi was reported to be more effective than micron-sized silica e.g. Silica Fume due to the exposed surface area to the pozzolanic reaction [18]. Improvement from nSi addition is affected by water/binder ratios [40].

Nano Alumina (nA)
Nano Alumina (nA) may come in the form of nearly spherical nanoparticles or as oriented / undirected fibers. As nano fibers, manufacture can be by [49]: When blended with concrete, it results in higher tensile and flexural strength (up to 2% replacement of cement with typical particle sizes of 15 nM) [50]. León et al. (2014) achieved increased abrasion and fracture resistance in cement mortars blended with nA [51].
Pozzolanic binders (CaO and CaO-MK binder mix) when combined with of nSi and nA, have been reported with enhanced properties such as increased compressive strength in conjunction with reduced porosity values [52]. This is due to denser microstructure, decreased carbonation and water absorption. nA did not contribute to the development of C-S-H compounds which are the main contributors to strength. Decreased carbonation was encountered, due to nA which had a negative effect on compressive strength, porosity and water absorption. However, this is on a smaller scale compared to the nSi reaction.
Naval and Chandan (2017) researched the effect of nano MgO (nMg) and nA when added to kaolinite clay and demonstrated improvements in soil properties such as Atterberg limits (e.g. reduced LL, PL and PI of soil), swelling potential decrease and maximum dry density increase [53].
Both Luo et al. (2012) and Lin et al. (2016) investigated using nA from sewage sludge ash/cement mix to stabilize clay soil. The optimum 1% nA achieved UCS 4.2 times higher than untreated soil after 7 days. Improved CBR values and reduced volumetric swelling was also observed [48,54].

Nano Lime (nCaO)
Nano-sized lime particles (nCaO) have recently been considered to overcome some limitations of traditional lime binders, e.g. difficulties in achieving complete carbonation. This may be addressed by the much smaller particle size and greater surface area for reaction of nCaO. Improvements were reported in Atterberg limit properties (e.g. decreases in liquid limit and plasticity index), maximum dry density decreases/increase in OMC as well as compressive strength increase by using nCaO / nSi when compared to conventional particle size lime and silica binder in silty clay [55]. Permeability was also recorded to be reduced by a factor of 10 in comparison to non-nano sized additives of silica and lime. In addition to greater reactivity because of surface area, it is proposed that nano particles can effectively fill in the pores of soil particles due to greater fineness.
The strength contribution of nCaO becomes more apparent with increasing curing time [56]. The effect of nCaO exhibits both positive (aggregation and pozzolanic reaction increasing C-S-H content and creating alkaline environment with more OH-/Si ratio) and negative effects (cement hydration is impeded when Ca (OH) 2 concentration becomes too high and excessive Ca (OH) 2 crystallization leads to strength decrease). Wang et al. (2016) therefore concluded that there was insufficient justification to utilize nCaO in lieu of common CaO. Comparison is made with other nanomaterials whereby nSi is considered more effective, nCaO being comparable to nA but better than nTiO2 which exhibited negative effects [56].

Alkali-Activated Materials (AAM)
The development of alkali-activated binder materials has been claimed to have commenced from ancient binders used for the pyramids to recent applications utilizing Palm Fuel Oil ash (POFA) precursors with NaOH / KOH alkali [5, 58, 59 and 60]. AAM has also been utilized to produce alkali-activated cements with lower CO 2 / energy requirements compared to conventional cement types (OPC) [61].
Alkali-activated material (AAM) binders react with any amorphous mineral aluminosilicates source either in the soil or introduced with alkali (Na or K base) or alkali earth metals (typically Ca). The process mechanism requires a source material for the Si-Al raw material, namely the prime material or precursor together with corresponding alkali activators (which can be a liquid or solid, but which requires water to dissolvee.g. NaOH, KOH etc.) [58]. The AAM binder is formed from the Si-Al raw material that has dissolved in a solution of alkali activators to form a mixture of gels and crystalline compounds which then hardens to a new, strong matrix amorphous condensed structure. It involves [58,62]:   Where Ca is insignificant or absent, Si + Al precursors (MK, FA etc.) may react with a medium to highly alkaline solution (either NaOH or KOH). Activation produces a polymeric structured material. It is noted that the requirement for calcium in any part of the alkali-activated structure is bypassed in the second model (in the case of A-S-H) [4].
Where Al is prevalent, the reaction is as follows: (1) Where Si is prevalent, the following reactions are included: In the case of alkali activation using Si + Al rich precursors ; for metakaolin (MK), a suggested optimum ratio to achieve strength was reported at Si/Al and Na/Al in the range of 3.5-3.8 for the former and 1.2 for the latter respectively [66]0 T. For Fly Ash (FA) precursors, Si/Al at 3.9 and Na/Al at 1 are the observed optimum ratios [67].
Studies also indicated that by increasing the amount of silicon, it results in more Si-O-Si bonds, which are stronger than Al-O-Al and Al-O-Si bonds [62,63] . Hence, the strength of the alkali-activated binder would increase with the Si/Al ratio since Si-O-Si bonds density increases as the Si/Al ratio increases.
The stabilization of organic peat soil with sodium silicate system grout with a combination of sodium silicate with calcium chloride/aluminium sulphate acting as a reactor/accelerator was studied by Moayedi et al. (2012) whereby UCS of stabilized soil increased to 270% untreated organic soil [68].
The effect of alkaline activation using Fly Ash with Portland cement as a binder was researched using sodium silicate / sodium hydroxide as the alkaline activator solution [4,69]. Fly Ash as a binder was utilized to achieve longterm strength gain in stabilized soil. Optimum levels of sodium hydroxide concentration were established together with the influence of solution/ash ratio. Fly ash was selected as more cost effective than metakaolin (MK).
Work done similarly on fly ash with sodium silicate + sodium hydroxide on silty sand achieved up to 2.8 MPa @ 28 days and 5.2 MPa @ 90 days UCS [70,71]. Sargent et al. (2013) experimented with sodium silicate / sodium hydroxide precursors and FA / GGBS / red gypsum binder on soft alluvial soil. The stabilized soil exhibited higher UCS but higher brittleness over untreated soil. Highest strength gains were gained, using alkali activated GGBS binders [72].
Pourakbar et al. [60] experimented with sodium hydroxide and potassium hydroxide together with Palm Oil Fuel Ash binder material and revealed the main factors for determining the strength of stabilized soil being 1) quantity of source binder 2) type of alkali activator 3) water content of soil 4) curing conditions.
Alkali-activation reactivity depends on the amorphous content of silica and aluminium [58,59]. The reactivity is linked to the material structure, being higher for higher amorphous content. Provis (2018) has recently discussed the utilization potential of AAM as a replacement for OPC in construction practices [73]. The disadvantage and hazards associated with difficult to handle concentrated alkali activator solutions may be resolved with recently developed one-part alkali-activated materials. One-part AAM which involves a dry mix of the solid aluminosilicate precursor, solid alkali source and other admixtures would only require water added. Similar end-products (N-A-S-H and K-A-S-H gels) are produced from one-part AAM reactions. These include instantaneous dissolution of the solid alkali activators and slower reactions involving aluminosilicates like two-part AAM [74]. Some examples of recent research (mainly focused on strength) are as follows:

Geopolymers (GP)
Geopolymers are a class of synthetic inorganic alkali aluminosilicate materials. They are be produced by the reaction of a solid alumino silicate with a highly concentrated alkali hydroxide/alkali silicate solution. The resultant geopolymer product from reaction is a generally amorphous, polymeric Si-O-Al framework binder material with potential in soil stabilization [76]. Feng et al. (2004) defined a geopolymer as basically a 3D aluminosilicate mineral polymer formed by several amorphous to semi-crystalline phases [77]. Geopolymers are formed by dissolution of aluminosilicate solids in a solution of alkali or alkali salts producing a mixture of aluminosilicates, aluminates and silicates in solution. With sufficient concentration, the solution solidifies through several gel phases and undergoes polymerization that hardens into a 3D aluminosilicate framework.
Like AAM, geopolymer materials have advantages over traditional cement binders in that the production process imposes less demands on energy consumption and produces less greenhouse gasses (CO 2 ) [3]. Raw materials for polymers can be sourced from a wide range of industrial waste materials which contain silicate and/or aluminae.g. natural pozzolans such as fly ash, ground granulated blast furnace slag, agricultural / construction waste materials with high silica/alumina content such as RHA, palm oil fuel ash (POFA), red clay brick waste and metakaolin [78].
Improved sulphate resistance properties of geopolymer concrete (prepared from blended Waste Fuels Ash precursor and sodium silicate alkali activator) was observed [79]. Du et al. (2017) investigated the physical, hydraulic and mechanical properties of clayey soil stabilized by geopolymer composed of a GGBS precursor and sodium silicate/ calcium carbide residue alkali activator [80]. The lightweight geopolymer stabilized soil (LGSS) developing greater water absorption, permeability ( of LGSS being 10 x k of LCSS) and material strength ( LGSS = 2-3.5 LCSS) when compared to benchmark lightweight cement stabilized soil (LCSS). C-S-H content in LGSS was ~ 2 times found in LCSS.
The effectiveness of soil-geopolymer stabilization and comparison between Fly Ash and GGBS based geopolymer types were made by Singhi et al. (2016) [81]. Alkali activator solution used selected was sodium hydroxide and sodium silicate. The unconfined Compressive strength from slag based geopolymer stabilized soil (~11 MPa @ 20%) was found to be much higher than that with Fly Ash-based (~0.3Mpa @ 20%). The difference in UCS between the two geopolymer base types starts to increase beyond 8% content of source geopolymer stabilization material in the soil.
Study by Zhang et al. (2013) on the effectiveness of metakaolin based geopolymer (MKG) soil stabilization on clay soils showed UCS values of MKG stabilized soils being much higher than for original soil as well as 5% cement stabilized soil at MKG contents > 11%. There was also an improvement in shrinkage strains at MKG concentrations > 8%. UCS values of MKG is not significant between 7 and 28-day strengths indicating predominantly fast reactions leading to early strength gainthis may be due to the precursor completing geopolymerization and strength development within 7 days [82]. Improvements in soil Young's Modulus were also recorded, but still, less than 5% cement stabilized soil. However, Davidovits (1994) [84] asserts that geopolymers (GP) should be separated from AAM in definition, as both belong to different chemistry systems. Although conceptually following a similar reaction mechanism for creation, Davidovits [3,84] differentiates GP from AAM by restricting geopolymer definition to only those obtained by pure metakaolin precursors and with end products derived, namely, through polycondensation to a 3D Kpoly(sialate-siloxo) polymer excluding N-A-S-H / K-A-S-H previously included by earlier research [84]. The interchangeable interpretation of AAM and geopolymers are due to similarities between alkali-activation and first step of geopolymerization. For geopolymers, the first step should instead be termed alkalination instead of alkaliactivation.
AAM type cements (e.g. Alkali-activated Slag cement) have a disadvantage to geopolymers due to the generation of leachates leading to potential long-term stability problems although AAM type cements generally achieve higher initial strengths over geopolymer cements. Geopolymer (GP) cements are now being developed in the form of 1) slag-based; 2) rock-based 3) fly ash-based and 4) ferro-sialate-based types. The manufacture of GP type cement can generate up to 80% less CO 2 and requires far less energy than portland cement [86].

Discussion
Being a mechanical process of improvement to the soil, reinforcement fiber materials have an immediate effect without curing in comparison to other binders which require curing time due to the hydraulic / chemical reaction process. Review of laboratory research results on reinforcement fiber materials demonstrated improvements in treated soil shear strength and axial strain to failure. However, if acting by itself, the level of compressive / shear strength improvement was still inferior to other binder materials. A higher tensile strength which defines the brittle to ductile post-peak behaviour transition in the improved soil due to the embedded fibers was clearly observed [5]. This can be important in the case of soils subjected to cyclic / dynamic loading.
The mixing of reinforcement fibers in soil also encountered difficulties in compaction and subsequent maximum dry density reduction with increasing dosage. Consistent and effective mixing of the fibers (possibly affecting viscosity in the slurry and hence limited to WDSM approach), application and distribution in soil may be difficult to achieve for DSM in the field. It may be more practical to pair with chemical based bindere.g. cementitious, pozzolanic, alkali-activated types etc. This could lead to a synergistic combination of improved compressive and tensile strength.
Due to smaller dimensions, nanoparticles increase rate of improvement compared to their micron / macro sized counterparts. With a different order of magnitude on the specific surface, reaction of the same materials, albeit on a nano vs micro scale, is more rapid and effective. Moreover, lesser amounts of nanomaterials can produce significant enhancements in soil improvement. Different nanomaterials lead to different effectse.g. CNT potentially increases flexural strength and control crack propagation whereas nano-versions of lime and silica increased compressive strength. The use of nanomaterials for ground improvement is presently hindered by high cost and the requirement to install via a slurry media (hence by the wet method) to ensure effective dispersion into the soil. The challenges of effective field mixing and dispersion into soil again needs to be resolved in DSM ground improvement.
Whilst cementitious and pozzolanic binders rely on Ca based reactions, AAM and GP binders undergo a different reaction pathway (Si/Al-based) without reliance on calcium, to improve soil properties. Research to date has shown the potential of alkali-activated binders to match or surpass the compressive strength gain compared to traditional cementitious binders. However, the strength gain comes at the cost of increased curing period (beyond 28 to 90 days) compared to short curing period of soil-cement mixtures (7 to 28 days). This is due to the faster dissolution rate of the calcium-type glassy material, forming the C-S-H gel that can be found in cement hydration [70]. However, this is compensated by greater benefits in environment reduction in CO 2 emissions (due to reduced usage of cementitious binder materials) and better resistance to aggressive soil environments, due to sulphate, chloride and acid exposure.
The difficulties and hazards associated to on-site handling of highly alkaline aqueous solutions can be largely avoided with development of one-part alkali activated materials which already comes pre-mixed in dry powder form like cement, requiring only addition of water. This allows the possibility of dry method DSM in high moisture content soils.
A recent development is the potential application of geopolymers to ground improvement. Research conducted on geopolymers produced from fly ash and metakaolin source materials have demonstrated the clear advantages of geopolymers over other binders in toughness and durability whilst further improving mechanical strength. Geopolymers are also more stable than AAM as they are not subject to potential leachate generation in AAM binders in the long-term.
High early strength for type 1 alkali -activated Fly Ash geopolymers (Si: Al ratio of 1 to 2) can be achieved with curing at optimum temperature range of 40 o C to 90 o C [85,87]. It may be possible to achieve this from exothermic reaction with sufficient dosage of lime. Another possibility would be a combined approach, when both Ca and alkali activated types of reactions are allowed to take placecalcium-based reactions leading to C-S-H / C-A-H phase and through activation of suitable silica/ alumina rich materials by strong alkali agents to produce alumino-silicate geopolymers (UCS potentially reach up to 160 MPa [64]). In the combined approach, it is noted that cementitious calcium-based reactions are also exothermic. A combined mix incorporating primary calcium-based reactions supplemented by pozzolanic secondary and alkali-activated/geopolymer tertiary reactions is also recommended for further research. In addition, type 2 slag / fly ash-based GP cements (Si: Al ratio of 2) which may harden at room temperatures and which do not require toxic solvents can also be applied.

Conclusions
The background and reaction mechanism behind alternatives to cementitious and pozzolanic binders and its application to DSM ground improvement have been covered in this review. Of these alternatives, both alkali-activated materials (AAM) and Geopolymer materials (GP) present an effective alternative chemical process pathway to soil improvement. Hazards associated with handling of alkali activator chemicals on site -e.g. Na 2 SiO 3 , NaOH, KOH may be largely avoided by using one-part pre-blended AAM or GP cements. This leads to research opportunities on the applicability of this hybrid implementation in real geotechnical solutions such as:  A systematic investigation using various combinations of calcium-based (which may include both cementitious and pozzolanic materials) and alkali activated alumino-silicate based binders and the effectiveness and specific improved properties for different soil conditions. The optimum proportions for combined binder types which can work synergistically for application to DSM can be determined;  Methods in mixing of these combined binder materials and effective dispersion into the soil;  Deriving a predictive constitutive improved soil model of 1) calcium based 2) alkali-activated / geopolymer as well as 3) hybrid combination in deep soil mix methods design for various soil types;  Application geopolymer cement as a binder;  As strength of the improved soil in DSM columns increase to higher compressive strengths approaching conventional concrete, the geotechnical model now transitions from improved ground to that of ground with rigid inclusions like unreinforced piles. Research is needed to establish the crossover point.