One-Part Geopolymer for Stabilising Crushed Rock Road Base Material

Abstract

Geopolymers have attracted wide attention as effective soil stabilisers, presenting significant potential for several geotechnical engineering applications. These binders offer environmental benefits by utilising abandoned aluminosilicate industrial by-products, such as fly ash and slag, through mixing with an alkaline solution. In addition, they also decrease dependency on conventional Ordinary Portland Cement (OPC), which is identified with substantial artificial greenhouse gas emissions and high energy consumption during manufacture. However, the practical utilisation of geopolymers for the stabilisation of road materials is hindered by the intricate preparation process, which necessitates precise control over the proportions of the ingredients to achieve the required mechanical properties. This complexity becomes more pronounced when compared to the relatively simple method of using conventional cement, which requires fewer safety precautions while mixing with soil. This study investigates the development of a One-Part Geopolymer (OPG) powder, specifically formulated for the stabilisation of a Crushed Rock Base (CRB) material used for road construction. The optimal blend of OPG powder, comprising fly ash, slag and sodium metasilicate, is identified by assessing the monotonic and dynamic mechanical performances of the treated CRB compacted at the optimum moisture content using Unconfined Compressive Strength (UCS) and Repeated Load Triaxial (RLT) tests. The results of the study indicate that enhancing the strength performance of the OPG-treated CRB requires the calibration of the sodium oxide (Na₂O) content in the alkaline activator with the total binder. It was also found that increasing the OPG content from 1% to 3% significantly enhances both the uniaxial strength and resilient modulus of the treated CRB, while simultaneously reducing the permanent deformation. Notably, the CRB specimens stabilised with 2% OPG exhibit mechanical properties comparable to those of bound Portland cemented materials.

1. Introduction

Chemical stabilisation is a widely utilised technique that aims to improve the mechanical properties of soils that fail to satisfy the designated design criteria. This technique is also deemed necessary to stabilise the Crushed Rock Base (CRB) commonly employed as a base material for flexible pavement in road construction. Ordinary Portland Cement (OPC) is typically utilised for the stabilisation of the CRB to improve its uniaxial strength and stiffness [1]. The extent of the enhancement in the stabilised CRB progressively varies with increasing OPC content, resulting in the formation of three distinct categories of road base: modified, lightly bound and fully bound. According to Austroads [2], the modified road base should achieve a maximum Unconfined Compressive Strength (UCS) of 1.0 MPa, which is obtained following 28 days of curing. On the other hand, the lightly bound road base should exhibit 1 to 2 MPa of UCS with an OPC content of less than 3%, whereas the bound road base should possess a UCS exceeding 2.0 MPa with a minimum OPC content of 3%. A different perspective emerges when considering a reliable OPC content to comply with the requirements of the resilient modulus (RM) as a design parameter that reflects the ability of a granular material to recover its initial axial dimension when subjected to specific loading conditions [3,4]. Main Roads Western Australia [5] established a range between 1000 MPa and 1500 MPa of RM to define the required OPC content for the stabilisation of CRB material. However, OPC poses significant environmental challenges, as it is recognised as one of the most energy-intensive materials within the construction sector and significantly contributes to carbon dioxide artificial emissions during its production process [6]. Specifically, the manufacturing of OPC results in the emission of approximately one ton of carbon dioxide for every ton of cement produced, culminating in an estimated annual total of 1.35 billion tons. Consequently, OPC production accounts for 5–8% of global carbon dioxide emissions [7]. Considering these detrimental environmental impacts, there has been a concerted effort in recent years to develop more sustainable alternative binders for soil stabilisation without compromising its effectiveness, including the stabilisation of road base materials in road construction.

Recently, considerable focus has been placed on the efficient recycling of abundant aluminosilicate waste materials, specifically fly ash and slag, through the process of alkaline activation. These materials can be transformed into alkali-activated binders, specifically in the forms of Sodium Aluminium Silicate Hydrate (N-A-S-H) and Calcium Aluminate Silicate Hydrate (C-A-S-H), commonly referred to as geopolymers [8]. Within the context of road construction, geopolymers exhibit potential to stabilise the base layer and amend its dynamic mechanical properties, specifically the strength and resilient modulus. According to Abdullah et al. [9], a design mix incorporating 6% geopolymer has demonstrated an average uniaxial compressive strength (UCS) of 1.2 MPa within the Crushed Rock Base (CRB) commonly used in Western Australia, aligning with the standard strength range for cemented lightly bound base course materials, as specified by Austroads. Nevertheless, while the treated mixtures satisfied the criteria for permanent deformation, none of the tested samples achieved the prescribed range for the resilient modulus. This underscored the necessity for comprehensive research to develop a dependable mixture that meets all design requirements. Moreover, the implementation of the geopolymer in its current form as a two-part mixture, comprising the precursor and the liquid activator, poses a challenge to the widespread adoption and handling of the binder for field implementation in soil stabilisation. The process of the two-part geopolymer mixture necessitates precise control over the mixture proportions, particularly the SiO₂/Na₂O ratio within the activator (ideally between 1 and 1.5) [10] and the sodium oxide (Na₂O) weight ratio in the activator relative to the total weight of solids in the binder (ideally between 6% and 8%). Such strict control appears impractical for large-scale in situ applications, including the stabilisation of CRB for road construction. Furthermore, the field mixing process requires safety measures that incur additional costs compared to straightforward stabilisation using OPC mixed with water and treated materials [11,12]. To enhance the performance and applicability of the use of geoploymers in field operations by streamlining the handling procedures and improving the safety protocols, a dry form of geopolymer binder is developed and presented in this paper. This approach involves utilising a One-Part Geopolymer (OPG), which requires only the addition of water when mixed with the treated material (in particular, in the CRB herein).

Recent studies have attempted to synthesise a One-Part Geopolymer (OPG) powder for soil stabilisation by incorporating a dry alkaline activator, such as sodium silicate (Na₂SiO₃·nH₂O), sodium hydroxide (NaOH) and sodium metasilicate (Na₂SiO₃), with a precursor (fly ash or slag) in a dry format [13,14,15]. Sodium metasilicate, synthesised through the reaction of silica sand (SiO₂) with sodium carbonate (Na₂CO₃) at elevated temperatures ranging from 1000 to 1400 °C followed by the removal or modification of water, has demonstrated significant efficacy in the development of an OPG binder characterised by superior mechanical properties [16]. This directly contributes silicate ions to the system when mixed with the precursor and activated by water. The produced ions form robust silicate-based binding products, specifically C-A-S-H or N-A-S-H, which are effective for soil stabilisation. In contrast, alternative activators, such as sodium silicate in its dry form, frequently require supplementation or combination with additives (e.g., sodium hydroxide) to optimise the SiO₂/Na₂O ratio and thereby enhance their effectiveness. The SiO₂/Na₂O ratio offered by sodium metasilicate ranges between 0.9 and 1.5, aligning closely with the optimal range required to maximise the geopolymerisation reaction [17]. However, to advance the application of the OPG synthesised from sodium metasilicate for the stabilisation of CRB in road construction, an efficient and practical mixture is needed that can surpass traditional binders in all aspects, particularly in terms of the strength and durability performance under dynamic loading. This research thus aims to develop a One-Part Geopolymer (OPG) powder exclusively for CRB stabilisation that can be activated simply by adding water during the mixing process. A specific aim is to optimise the OPG components by assessing the influence of different geopolymer ingredients (i.e., fly ash, slag and sodium metasilicate) on the mechanical properties of the stabilised CRB. The research also aims to advance the comprehension of the durability behaviour, quantified by the resilient modulus and permanent deformation of the treated CRB. It is anticipated that this research can advance the use of the OPG in the road construction sector as a more sustainable binder than conventional OPC binders.

2. Materials and Methods

2.1. Crushed Rock Base (CRB)

The Crushed Rock Base (CRB) material utilised in this study was sourced from Western Australia. The CRB was evaluated for its suitability following the parameters outlined in the specifications of Main Roads Western Australia (MRWA) 501.11 [5]. Initially, the material was subjected to particle size distribution testing following MRWA 115.1:2019 [18], modified compaction as per MRWA 133.1:2022 [19] and Atterberg limits as specified by AS 1289.3.9.1:2015 [20] and AS 1289.3.2.1:2009 [21]. The results of the sieve analysis, presented in Figure 1, show compliance with the particle size distribution requirements outlined in the specifications of MRWA 501.11 [5]. The material exhibited a moisture content of 20.8% at the liquid limit, which is found within the maximum allowable value of 25% specified in [5]. Results could not be obtained from the plastic limit test, as the material exhibited negligible plasticity. Regarding the results from the modified compaction test, an optimum moisture content (OMC) of 6.4% was determined, corresponding to a maximum dry unit weight (MDU) of 23.1 kN/m³.

2.2. One-Part Geopolymer (OPG)

The One-Part Geopolymer (OPG) employed in this research comprised three components: Ground Granulated Blast-Furnace Slag (GGBFS), fly ash and sodium metasilicate pentahydrate. The GGBFS was supplied by Cement Australia Pty Limited following AS 3582.2:2016 [22]. The fly ash, categorised as Class F and supplied by Cement Australia Pty Limited, was sourced from Gladstone Power Station in compliance with AS/NZS 3582.1:2016 [23]. The solid alkali activator was sodium metasilicate pentahydrate (Na₂SiO₃.5H₂O), supplied by Chem-Supply Pty Ltd. The activator was characterised by a silicon dioxide to sodium oxide (SiO₂/Na₂O) ratio of 0.97, which falls within the range of 1 to 1.5 necessary for achieving superior compressive strength in geopolymer materials [24]. The three components of the OPG applied in this study are illustrated in Figure 2, and a breakdown of their chemical composition is presented in

Table 1. Chemical composition of fly ash, slag and sodium metasilicate used.

Material	Composition (%)
	SiO2	Al2O3	Fe2O3	CaO	Na2O	MgO	K2O	SO3	H2O	LOI *
GGBFS	31.9	13.6	1.35	45.3	0.26	6.2	0.31	2.04	–	0.06
Fly ash	56	26.8	6.6	9.79	0.28	1.22	0.78	0.26	–	1.59
Solid activator	28.3	–	–	–	29.2	–	–	–	42.5	–

* LOI: loss of ignition.

.

The fly ash and slag, which are industrial by-products, were subjected to dry mixing before use, maintaining a ratio of fly ash-to-slag of 4:6 in the OPG binder. This ratio was chosen due to the role of slag in enhancing initial strength and the contribution of fly ash to long-term strength development [17]. Subsequently, the blend was combined with varying quantities of sodium metasilicate to identify its optimal content. The concentration of Na₂O in the total dry weight of the geopolymer powder was identified as a variable exerting a significant influence on the compressive strength of the geopolymer cement, as noted in previous studies [25]. Consequently, this research quantifies the amount of sodium metasilicate to be incorporated into OPG mixtures based on the concentration of Na₂O, as will be seen later in

Table 2. Variations of mixtures and testing.

Test	Na2O (%)	OPG Content (%)
UCS	3	2
	6	1
		2
		3
	9	2
RLT	6	1
		2
		3

.

2.3. Specimen Preparation and Testing

In the process of the sample preparation, precise quantities of the developed One-Part Geopolymer (OPG) were utilised as partial replacements in three varied proportions of 1%, 2% and 3% by weight of the Crushed Rock Base (CRB). These were incorporated into the CRB and thoroughly mixed using a Hobart mixer bowl. Subsequently, free water was introduced to achieve the optimal modified compaction, followed by an additional three minutes of mixing. The compaction parameters, specifically the maximum dry unit weight and optimum moisture content, for each blend of the OPG-treated CRB were determined through a series of modified Proctor tests. These tests were performed following the MRWA test method 133.1 [19]. However, the results indicated that the small additions of 1%, 2% and 3% OPG at a typical value of 6% of the Na₂O content within the binder did not significantly impact the compaction parameters of the CRB, with the optimum moisture content aligning with that of the untreated CRB at 6%. It is important to highlight that the water chemically bound to the solid activator is included in the total water content; hence, the binder content and Na₂O concentration calculations are based solely on the solid components. The 6% moisture content encompasses both the chemically bound water in sodium metasilicate and the free water added to the mixture, as recommended by previous studies [17].

Following the mixing process, the OPG-treated CRB was immediately utilised to cast cylindrical specimens with different dimensions to comply with various testing regimes. For the Unconfined Compressive Strength (UCS) test, the diameter of the mould was similar to that of the modified compaction test mould, and the soil was compacted using a modified compaction hammer. The mould had a diameter of 105 mm and a height of 115 mm. For the Repeated Load Triaxial (RLT) test, the dimensions of the mould were 100 mm in diameter and 200 mm in height. The OPG-treated CRB was compacted in layers of controlled weight and thickness to replicate the optimum compaction parameters. Following compaction, the specimens were promptly unmoulded, wrapped in plastic film and cured for 28 days in a controlled environment, with the temperature maintained between 20 and 25 °C and the humidity at 60%. Post-curing, the specimens were unwrapped, and their top surfaces were levelled with a thin plaster layer.

For the UCS testing, the designated specimens were submerged in water for four hours to simulate the potential weakening effects of water infiltration on the treated CRB [26]. At least two UCS tests were conducted per design mix, ensuring the results remained within a 20% variance. The mean values from these tests were utilised in the analysis for precision. The UCS tests adhered to the MRWA 143.1 guidelines [26], using a strain rate of 1% per minute, and were conducted at 28 days of curing to assess the long-term strength performance.

The Repeated Load Triaxial (RLT) tests followed the Austroads Repeated Load Triaxial Test Method, AG: PT/T053 [3], and were conducted to evaluate the resilient modulus and permanent deformation of the OPG-treated CRB. The primary objective of assessing the resilient modulus is to determine the capacity of the base material to recover its original shape and resist deformation following triaxial testing, where the specimen is subjected to various vertical and lateral cyclic stresses. This test is essential for understanding the stiffness and resilience of the material. Quantitatively, the resilient modulus is defined as the ratio of the repeated deviator stress to the recoverable or resilient axial strain [3]. Within the context of base materials, a lower resilient modulus indicates reduced stiffness, thereby diminishing the material’s capacity to withstand repeated loading. This property is correlated with an increased propensity for deformation, and ultimately results in a decreased lifespan of the road. Consequently, a low resilient modulus of a road base material is considered disadvantageous in the context of road construction. In this research, the RLT tests employed the UTM-14P digital servo control testing machine under unsaturated drained conditions on specimens cured for 28 days. At the designated testing time, the cured specimens were placed in the RLT apparatus, covered with a rubber membrane and sealed with O-rings. Each sample underwent 65 stress stages for the resilient modulus test, each characterised by a different deviator stress (σ_d) and confining pressure (σ₃), as outlined in Figure 3. Two hundred loading cycles were applied per stress stage, with each cycle consisting of 1 s of deviator stress application followed by 2 s of rest, resulting in a total cycle period of 3 s. The resilient modulus was then calculated as the average of the last six cycles at each stress stage, yielding a total of 66 resilient modulus values per specimen tested. Note that before the testing stage itself, the specimens were subjected to 1000 loading cycles for pre-conditioning, at a confining stress of 50 kPa and a deviator stress of 100 kPa, to allow the end caps to bed into the specimen and for the stabilisation of the resilient strains. Following the resilient modulus test, the same specimen was tested for permanent deformation. In compliance with the test method of AGPT-T053-07 [3], the test consisted of three stress stages of 10,000 loading–rest cycles each, adding up to 30,000 cycles of 3 s each. During these cycles, the confining stress was kept constant at 50 kPa, while the deviator stress varied: 350 kPa in the first stage, 450 kPa in the second stage and 550 kPa in the last stress stage.

3. Results and Discussion

3.1. Unconfined Compressive Strength Tests

Impact of Na₂O: Figure 4 presents the effects of altering the solid alkali activator within the One-Part Geopolymer (OPG) powder by modifying the sodium oxide (Na₂O) content on the strength performance of the treated Crushed Rock Base (CRB) cured for 28 days. The OPG content within the treated CRB was consistently maintained at 2%, while three distinct Na₂O levels of 3%, 6% and 9% were examined. As demonstrated in Figure 4a, the incorporation of the OPG generally enhances the strength of the CRB beyond that of the untreated CRB, which could not be tested due to the loose nature of the sample. With a 3% Na₂O content in the OPG, the predominant stress–strain response exhibited semi-brittle yielding, where the stress reached a peak of 0.66 MPa before gradual failure. Increasing the Na₂O content to 6% resulted in a higher peak strength of 3.5 MPa and demonstrated more brittle behaviour, as indicated by a reduction in the strain corresponding to the peak stress. At 9% Na₂O, the behaviour remained brittle, with a slightly higher peak stress of 3.7 MPa than that of the 6% Na₂O content. The variation in the responses to geopolymer treatment across the Na₂O levels is further evidenced by changes in stiffness, as shown by E₅₀, the secant modulus of the elasticity at 50% of the maximum stress, depicted in Figure 4b.

Both the qualitative and quantitative changes in the stress–strain behaviour of the OPG-treated CRB, captured in Figure 4, can be attributed to the formation of artificial cementitious products and bonding between the CRB particles.

To elucidate the bond mechanism within the OPG-treated CRB, a limited microstructural investigation using SEM was conducted on the treated specimen with 6% Na₂O, as illustrated in Figure 5. The SEM analysis indicates the presence of a specific substance on the smooth surfaces of the partially reacted plate-like slag and spherical fly ash particles. This indicates the leaching of silica and alumina from such particles and the formation of cementitious products (i.e., C-A-S-H and N-A-S-H) between the CRB particles. The partially reacted slag and fly ash particles and the formed cementitious products seem to act as nucleation sites that bond the CRB particles into clusters, thereby modifying the CRB structure and enhancing the mechanical response [27]. Consequently, the additional increase in strength and stiffness with a higher Na₂O content in the OPG can be explained by the increased formation of cementitious products and clay clusters within the treated soil mass [28]. Based on the findings in Figure 4 and Figure 5, it is likely that the specimens with 3% Na₂O lacked sufficient sodium metasilicate to fully react with the precursors to produce sufficient cementitious products. Conversely, in the specimens with 9% Na₂O, while a more solid activator is available, the excess sodium metasilicate may create larger pores that partially compromise the strength increase. This suggests an optimal Na₂O content of around 6% within the OPG, as previously reported by other studies on concrete as stated in Dong, et al. [17].

Impact of OPG Content: To examine the impact of the OPG on the compressive strength of the treated CRB, three levels of OPG content, specifically 1%, 2% and 3%, were evaluated for the UCS performance at 28 days of curing. During this phase of testing, the sodium metasilicate pentahydrate content was maintained at a constant level, regulated by its optimal Na₂O content by dry weight of the OPG, which was established to be 6%. The UCS results are depicted by the stress–strain curves illustrated in Figure 6a. The results demonstrate that an increased OPG content enhances the compressive strength of the treated CRB. Notably, the compressive strength peak value more than doubled, from 2.0 MPa to 4.6 MPa, when the OPG content increased from 1% to 2%, achieving a peak value of 10.8 MPa for the specimens treated with 3% OPG. Qualitatively, the specimens with 3% OPG exhibited a more brittle failure compared to those with 1% and 2% OPG. The differential response to CRB treatment at varying OPG contents is further substantiated by alterations in stiffness. This is indicated by E₅₀, the secant modulus of the elasticity measured at 50% of the maximum stress, as illustrated in Figure 6b. This enhancement in the strength performance with an increased geopolymer content concurs with findings reported for conventional geopolymers by [9]. The presence of higher OPG contents resulted in the formation of larger quantities of N-A-S-H and C-A-S-H gels, which consequently improved the compressive strength of OPG-treated CRB specimens.

Compared to conventional binder materials, the specific treated material in this study outperformed CRB treated with Ordinary Portland Cement (OPC) with similar binder contents, as documented by other research [1]. Moreover, the results exceeded the compressive strength achieved in a previous study [9] involving CRB from the same source, stabilised with two-part geopolymer cement (precursor and liquid activator). The results underscore the potential application of CRB treated with a 1% OPG content as a lightly bound base material, while CRB treated with a 2% OPG content can serve as a bound cemented material in road pavements, consistent with the standards specified in [5].

3.2. Repeated Load Triaxial (RLT) Tests

Resilient Modulus: The resilient modulus of the One-Part Geopolymer (OPG)-treated Crushed Rock Base (CRB) was evaluated for durability through Repeated Load Triaxial (RLT) testing on specimens that were cured for 28 days. This testing was conducted at three distinct levels of OPG content, specifically, 1%, 2% and 3%, as well as untreated CRB (0% OPG). The sodium metasilicate content was maintained at a constant level, corresponding to the optimal content of 6% Na₂O by dry weight of the OPG. Figure 7 illustrates the relationship between the resilient modulus (M_r) and the 65 stress stages, which represent varying combinations of the deviator and confining stresses applied during the RLT test, as previously shown in Figure 3. The results indicate an increase in the resilient modulus with the increase in OPG content across all stress stages. Notably, the samples withstood the final stages of the resilient modulus testing, where failure becomes more probable due to a higher ratio of deviator stress to confining pressure. This finding underscores the efficacy of OPG addition, even at a relatively low concentration of 1%.

Table 3. Values of resilient modulus of OPG-treated CRB with 1, 2 and 3% OPG.

OPG Content (%)	Resilient Modulus (MPa)
	Minimum	Maximum	Average
0	170	587	483 ± 63
1	893	1471	1151 ± 146
2	1117	2028	1563 ± 240
3	1757	2614	2165 ± 174

presents the maximum and minimum values of the resilient modulus of the Cemented Residual Base (CRB) for varying OPG contents. The final column of the table details the average resilient modulus along with the corresponding standard deviation. It is apparent that the average resilient modulus consistently increases as the OPG content rises, ranging from 1151 MPa at 1% OPG to 2165 MPa at 3% OPG. Furthermore, the standard deviation findings indicate that the CRB specimens exhibit low variability in resilient modulus; the coefficients of variation are 12.7%, 15.4% and 8.0% for the specimens treated with 1%, 2% and 3% OPG, respectively. In this study, the resilient moduli for the CRB sourced from Western Australia treated with the OPG are higher than those documented by [29], who evaluated a hydrated OPC-treated CRB with 1%, 2% and 3% OPC contents. Moreover, the resilient modulus achieved in this study using the OPG surpasses the results obtained by [9], who examined CRB from the same source treated with a two-part geopolymer at 2%, 4% and 6%. In the context of substituting conventional cement for the stabilisation of road pavement bases in Western Australia, the analysis of the average resilient modulus indicates that CRB treated with 1% and 2% OPG offers a viable alternative. This conclusion is supported by the standards set forth by MRWA, which stipulate a resilient modulus range of 1000 MPa to 1500 MPa as the acceptable minimum and maximum thresholds [5].

The correlation between the resilient modulus and the bulk stress was also assessed in this section. Given that the resilient modulus serves as a critical input parameter in the structural design of pavements, understanding this correlation is essential for predicting the values encountered in the field when materials are exposed to specific confining stresses within the pavement layers. Figure 8 illustrates the results, where the resilient modulus is plotted against the bulk stress, denoted as (σ = σ₁ + σ₂ + σ₃), (σ₂ = σ₃), σ₁ represents the deviator axial stress and σ₃ signifies the confining stress. The correlation was examined using the K-θ model as proposed by [9]. This model calculates the resilient modulus (M_r) using the formula (M_r = K₁ × θ^K₂), where (θ) represents the bulk stress and K₁ and K₂ are the regression coefficients. In Figure 8, this model is depicted through exponential trend lines.

Table 4. Regression coefficients of k-θ model of OPG-treated CRB.

OPG Content (%)	K1	K2	R2
1	214.18	0.2816	0.8009
2	185.4	0.3554	0.8446
3	1136.9	0.1074	0.2775

presents the regression coefficients corresponding to each OPG-treated CRB mixture. The coefficients of determination (R²) displayed in the final column signify a strong correlation between the bulk stress and the resilient modulus for the CRB mixtures treated with 1% and 2% OPG. Conversely, the coefficient of determination reveals a weak correlation for the CRB treated with 3% OPG. The observed weak correlation may be ascribed to the pronounced sensitivity of the resilient modulus in cemented materials, particularly at elevated levels of stabilisation, to varying stress levels, as highlighted by previous studies such Ref. [4], given that the specimens treated with this binder content exhibited a compressive strength and stiffness within the range typical of bound cemented materials. Nevertheless, the average resilient modulus obtained for the CRB treated with 3% OPG demonstrated a notably low coefficient of variation (8%), suggesting it is both accurate and representative for design applications.

Permanent Deformation: In the context of assessing the resilient modulus, the CRB specimens treated with 1%, 2% and 3% OPG were subjected to permanent deformation tests to evaluate the performance associated with varying OPG contents. The testing protocol involved exposing each specimen to three loading stages, each comprising 10,000 cycles. The stress conditions during these cycles were defined by dynamic deviator stresses of 350, 450 and 550 kPa, coupled with a static confining stress of 50 kPa. In terms of the permanent strain, the data illustrated in Figure 9 reveal that the CRB treated with 2% and 3% OPG exhibited minimal permanent deformation (below 0.05%) after the third stress stage, following exposure to 30,000 load cycles. In contrast, the specimens treated with 1% OPG underwent significantly greater deformation, approximately 0.4%, under the same loading conditions. This indicates a substantial increase in stiffness with higher levels of OPG incorporation. Notably, a sudden increase in the cumulative permanent strain curves was observed between stress stages, underscoring the relationship between the magnitude of the deviator stress (350, 450 and 550 kPa) and permanent deformation.

A comprehensive qualitative analysis of the material performance can be conducted using the methodology developed by [30], which forecasts material behaviour in the field based on the characteristics of the permanent strain and resilient deformation curves.

Table 5. Material behaviour for granular bases [30].

Performance	Permanent Strain Rate	Resilient Deformation
Stable	Decreasing	Decreasing to constant
Unstable	Decreasing to constant	Constant to increasing
Failure	Constant to increasing	Increasing

delineates the expected performance categories (stable, unstable and failure) as suggested by the authors. According to their methodology, the CRB treated with 1% OPG, despite showing a decreasing resilient deformation (Figure 10) within each stress stage, exhibited a constant permanent strain rate, leading to an anticipated unstable performance. Conversely, the specimens treated with 2% OPG demonstrated a marked improvement in the predicted performance. As the permanent strain rate decreased and the resilient deformation diminished (Stage 2) and remained constant (Stage 3), stable performance in the field is likely. Finally, the CRB with 3% OPG is anticipated to function as a stable pavement material, as both the permanent deformation rate and resilient deformation decreased across all stress stages.

4. Conclusions

This paper investigated developing and employing a One-Part Geopolymer (OPG) to stabilise a Crushed Rock Base (CRB) material for road construction. The primary objectives were to ascertain the optimal components and examine the impact of the developed powder content on the compressive strength, resilient modulus and permanent deformation of the OPG-treated CRB. The results of the study indicate that an increased content of OPG significantly enhances the performance of the material when used as the base layer in pavements. Specifically, the OPG contributes to the artificial cementation of the CRB particles, resulting in a stiffer material that exhibits a reduced susceptibility to permanent deformation. The following specific findings can be drawn from this study:

• Uniaxial Compressive Strength (UCS) tests revealed that the optimal Na₂O content, a component of the solid activator, is 6% by dry weight of the OPG. This concentration resulted in CRB specimens with markedly superior compressive strength compared to those with a 3% Na₂O content, while achieving comparable strength to specimens with a 9% Na₂O content.
• CRB specimens treated with 1%, 2% and 3% OPG contents displayed compressive strength values exceeding the minimum requirements for use as a bound cemented base material, as stipulated by Austroads. Furthermore, the compressive strength consistently increased with higher OPG contents, from 2 MPa at 1% OPG to approximately 11 MPa at 3% OPG, suggesting that the OPG can effectively enhance the artificial cementation of the CRB particles, resulting in a stiffer material.
• The results of the resilient modulus demonstrated that the OPG-treated CRB specimens stabilised with 1% OPG achieved values meeting the MRWA requirements for CRB treated with 2% Ordinary Portland Cement (OPC). The specimens with 2% and 3% OPG exhibited superior resilient strength, attaining a maximum average value of 2165 MPa. There is a clear enhancement in the resilient modulus with an increasing OPG content. Moreover, the developed K-θ models indicated a strong correlation between the resilient modulus and bulk stress for the CRB treated with 1% and 2% OPG.
• The CRB specimens treated with 2% and 3% OPG content demonstrated minimal cumulative permanent strain, less than 0.05%, after 30,000 loading cycles. Qualitatively, these specimens are anticipated to remain stable as pavement base materials, whereas those containing 1% OPG are likely to exhibit unstable behaviour.

Overall, this study highlights the benefit of using the developed One-Part Geopolymer as a promising binder material, with the potential to replace and possibly surpass Ordinary Portland Cement in CRB stabilisation for road applications concerning compressive strength, resilient modulus and permanent deformation. However, further research is necessary to optimise OPG-treated CRB mixtures and explore other pertinent mechanical properties, such as tensile and flexural strength, taking into account the effects of the moisture content and gradation in CRB.