Calcined Shale and Palm Oil Fuel Ash as Partial Replacements for Calcium-Based Binders to Enhance Compaction Properties of Marine Clay

ABSTRACT Marine clay is a problematic soil; therefore, its properties have been improved by mixing calcium-based binders, with reinforcing components. However, these materials exhibit many shortcomings, and better options are sought. Calcined shale (CS) is frequently used in concrete as a supplementary cementitious material, but it has not yet been used in soil stabilization. This study investigates the partial replacement of lime in marine clay treatment with CS and palm oil fuel ash (POFA). This study comprises two parts. In Part (1), POFA – lime-treated marine clay was assessed to determine the optimum proportion of POFA. In Part (2), CS – POFA – lime-treated marine clay was evaluated to determine the optimum ternary blending mixtures through many tests. Results of Part (1) indicate that the optimum proportion of POFA is 5%. In Part (2), CS exhibits effective enhancement in the properties of marine clay, and the optimum mixture was 10% CS, 5% POFA, and 3% lime. When lime content was reduced from 5% to 1% through replacement with CS, maximum dry density and optimum water content increased by 1.7% and 5.2%, respectively. Hence, CS can be effectively used as a partial replacement for lime in marine clay treatment.


Introduction
The use of marine clay can be problematic (Das et al. 2021;Kamaruddin et al. 2020), and thus, its properties are occasionally enhanced by adding various substances, including cement and lime (Alrubaye, Hasan, and Fattah 2016;Lu et al. 2020).Stabilization by using lime and cement increases the strength and reduces the compressibility of soil (Lu et al. 2020).Alrubaye, Hasan, and Fattah (2016) specified 5% lime as the optimal percentage for enhancing the compressibility of soft clay.
Although cement and lime provide many benefits, they can also compromise the environment and structures (Emmanuel et al. 2020;Gartner 2004).Rae et al. (2019) determined that 100% of the carbon in CaCO 3 is released into the atmosphere in the form of CO 2 .Meanwhile Gartner (2004) found that cement accounts for up to 7% of all CO 2 emissions from artificial sources.Cement can also present the problem of sulfate attack (Emmanuel et al. 2020).Therefore, the partial or full replacement of calciumbased binders with sustainable materials is necessary for soil stabilization.Many attempts have been made to address this issue.Scholars have suggested the use of various natural pozzolanas, such as metakaolin (Hasan 2012); however, its high cost has restricted this practice (Khatib, Baalbaki, and ElKordi 2018).Other natural pozzolanas, such as calcined shale (CS), have been proven beneficial for concrete construction (Seraj et al. 2015;Yuan et al. 2015) and less expensive.Yuan et al. (2015) reported that CS improves the strength and reduces the corrosion resistance coefficient of cement; moreover, the performance of CS is comparable with that of metakaolin in cementitious mixtures (Cheng et al. 2013;Seraj et al. 2015).However, CS has not yet been used in soil stabilization.
This study investigates the effects of CS on marine clay properties and aims to clarify its potential to replace, and consequently, reduce the use of potentially harmful lime and cement in marine clay stabilization.Research has indicated that the inclusion of this cementitious, pozzolanic material, coupled with calcium-based binders, can improve the properties of concrete (Seraj et al. 2015), and that the addition of another reinforcing material, such as palm oil fuel ash (POFA), can increase soil strength (Tan et al. 2021).POFA is abundant in Malaysia because approximately 70% of raw materials used in palm oil production generate solid waste (Vijayaraghavan, Ahmad, and Ezani 2007).Thus, the current study focuses on the development of a new soil matrix, which can be achieved by combining POFA and CS with lime, to optimize the improvement of marine clay.
This study comprises two parts.In Part (1), the properties of POFA-5% lime-treated marine clay were evaluated and quantified.In Part (2), the properties of the composite were assessed to determine the applicability of CS as a new supplementary cementitious material (SCM).This assessment was achieved through multiple tests, including the Atterberg limits.This test includes the liquid limit (LL), plastic limit (PL), and shrinkage limit (SL).Specific gravity, compaction, and consolidation tests were also conducted.In addition, field-emission scanning electron microscopy (FESEM), energy-dispersive X-ray spectroscopy (EDX), X-ray fluorescence (XRF), and X-ray diffraction (XRD) analysis were performed on marine clay before and after treatment to observe the textural and chemical changes between the treatment materials and the soil.

Materials
Figure 1 shows the collection of the marine clay sample used in this study.The sample was obtained from Selangor, Malaysia at a depth of 1.5 m below the ground surface.The soil was initially air-dried and then kept in a sealed bucket.Particle size distribution was determined in accordance with the Unified Soil Classification System by using a laser diffraction particle size analyzer (Figure 2).The soil sample was inorganic, high-plasticity clay soil (CH) composed of clay (35%), silt (62%), and sand (3%).The gradation and uniformity coefficients were 0.853 and 4.47, respectively, indicating well-graded soil.Tables 1 and 2 respectively provide the physical characteristics and chemical compositions of marine clay.
The POFA used in this study was prepared by drying it in an oven at a temperature of 105°C for 24 h.Thereafter, it was passed through a 300 µm sieve to remove any incompletely combusted fibers and nutshells.The results of the laser diffraction particle size analysis are presented below in Figure 2. Approximately 90% of the particles were less than 50 µm, and the mean particle size (D 50 ) was 16 µm.The POFA used in the current study exhibits properties that are similar to, but finer than those of the POFA samples used by Pourakbar et al. (2015) and Khasib et al. (2021).Hence, no further grinding was required.The loss on ignition (LOI) value of the collected Class F POFA was 1.8%, which is within the range of the maximum value permitted for the use of POFA as SCM (i.e., 6% in accordance with ASTM C618).Table 2 also provides the chemical compositions of the POFA and CS used in the current study.The hydrated lime [Ca(OH) 2 ] used in this experiment was stored in an airtight bin to avoid any changes in its properties over time.Its physical and chemical compositions are listed below in Table 3.
CS is a soft, layered sedimentary rock formed by the sustained, long-term deposition of clay-sized particles in relatively calm, muddy waters (Seraj et al. 2015).In the present study, laser diffraction particle size analyzer test (Figure 2) was performed to determine the particle size distribution of CS.Around 90% of the particles were less than 100 µm, and D 50 was 20 µm.Approximately 69% of the particles were below 45 µm.ASTM C618 specifies that 2/3 of the particles must be smaller than 45 µm.Table 2 indicates that the CS used is classified as N pozzolana with SiO 2 +AL 2 O 3 +Fe 2 O 3 above 70% and SO 3 below 4% (Seraj et al. 2015).LOI was 2.694%, which is below 6% of the maximum value allowed by ASTM C618.

Microstructural analyses
Analyses via XRF, XRD, FESEM, and EDX were used to determine the chemistry, crystalline phases, and microstructural changes of marine clay, CS, POFA, lime, and the optimally stabilized soil samples from Part (2) after a curing period of 7 days.

Atterberg limits and SL
The Atterberg limits, such as the LL, PL, and SL of the samples, were determined in accordance with ASTM D4318.

Specific gravity
The specific gravity of the samples was determined in accordance with ASTM D854-00.

Compaction test
Compaction testing was conducted in accordance with ASTM D698-70.

Consolidation test
Consolidation testing was performed in accordance with ASTM 2435-70, following a curing period of 7 days.The samples were prepared at the optimum water content (OWC) and maximum dry density (MDD) determined from the compaction test.

Structural and morphological analyses of marine clay, lime, POFA, and CS
FESEM tests were conducted on marine clay and the treatment materials to gain a deeper understanding of the stabilization mechanism after treatment (Figure 3).The marine clay used in this study exhibited a discontinuous and fractured structural composition, with voids and without hydrates, as shown in Figure 3a.The high porosity originated from the inter-agglomerate pores between clay particles.Figure 3b shows that the isolated soil particles of marine clay exhibited a flaky and fragile structure with a certain degree of organization.Similar findings for marine clay structures were reported by Kamaruddin et al. (2020) and Khasib et al. (2021) who observed that marine clay structures are fractured with voids and macropores.
Figure 4 shows the image of the FESEM tests conducted on POFA.The microstructural characterization of POFA is irregular and round, as depicted in Figure 4a.Other particles of different sizes and shapes were also found, demonstrating a well-graded distribution wherein particle sizes ranged from extremely small to large (Figures 4b,c).Kamaruddin et al. (2020) and Khasib et al. (2021) reported similar results for the structure of POFA.Figures 5 and 6 present the FESEM results for lime and CS, respectively.The particle shape of CS (natural pozzolana) is generally irregular or angular, which agrees with other studies (Ramezanianpour 2014) that investigated a type of natural pozzolan known as pumicite.Pumicite is similar to CS in its effect as pozzolanic material, because both materials contain high precentages of silica and alumina and are composed of angular and porous particles.
Table 2 provides the chemical compositions of marine clay, POFA, and CS, as determined via XRF.For marine clay, the predominant chemical components were SiO 2 , Al 2 O 3 , and Fe 2 O 3 .A similar proportion was reported by Tan et al. (2021), who found that SiO 2 , Al 2 O 3 , and Fe 2 O 3 in marine clay are 40.62%,34.27%, and 10.49%, respectively.For POFA, three major oxides (i.e., Fe 2 O 3 , Al 2 O 3 , and SiO 2 ) combined to account for 70.35%, exceeding 70%, which is the minimum amount required by ASTM C 618.This finding also indicates that the POFA used is suitable as a pozzolanic additive in  geotechnical construction projects.The proportions of most chemicals are within the range specified in Abdeldjouad et al. (2019), who found that the SiO 2 , Al 2 O 3 , and Fe 2 O 3 contents of POFA were 46.04%, 19.39%, and 6.1%, respectively.
For CS, the dominant components were silica and alumina, which are important minerals that react with lime in a pozzolanic reaction.The total content of the three major oxides (i.e., Fe 2 O 3 , Al 2 O 3 , and SiO 2 ) was 93.195%, which is higher than the minimum requirement set by ASTM C 618, i.e., 70%.This finding indicates the good suitability of CS as a pozzolanic additive in geotechnical construction projects.Furthermore, the chemical composition of CS was compared with those of the natural pozzolanas used by Al-Swaidani, Hammoud, and Meziab (2016), which revealed that the SiO 2 , Al 2 O 3 , and Fe 2 O 3 contents were 46.5%, 19.28%, and 11.2%, respectively.The natural pozzolana used by Al-Swaidani, Hammoud, and Meziab (2016) and the CS used in this work exhibit similar effects as pozzolanic materials, considerably depending on their high silica and alumina contents to react and produce the cementitious component.Notably, the SiO 2 and Al 2 O 3 contents used in the present research (CS) are higher.Thus, a conclusion can be drawn that CS is better for pozzolanic activity.

Atterberg limits and SL
Figure 8 illustrates the effects of 5% lime combined with various percentages of POFA on the consistency properties of marine clay.PL initially increased with the addition of lime to soil (from 36.4% to 52.4%).Then, it decreased (up to 46.8%) with the addition of 20% POFA, as illustrated in Figure 8b.A similar trend was reported by Prasad, Prasad, and Shivanarayana (2018), who found that LL values were reduced while PL values were increased with the addition of lime.Meanwhile, Figure 8a,c depict the effects of lime and POFA on the LL and plasticity index (PI) of marine clay, respectively.In the present study, LL and PI decreased from 80.5% and 44.1% to 78.6% and 26.2%, respectively, with the addition of 5% lime to marine clay.As lime is added to soil, Ca + ions are released into the pore fluid, generating an increase in electrolyte concentration that reduces the thickness of the double layer retained around soil particles, which, in turn, decreases LL (Samantasinghar 2014).Similarly, the LL and PI of marine clay decreased with the addition of POFA to lime-treated marine clay.LL and PI fell from 80.5% and 44.1% to 67.4% and 20.6%, respectively, with the addition of 5% lime and 20% POFA, as shown in Figure 8a,c.The reduction observed reached up to 16% and 53% for LL and PI, respectively.
The inclusion of lime and POFA reduced PI by about 50%, corresponding to a change from high to medium plasticity soil.When 5% POFA was used to treat the samples, the amount of reduction was significant at 45%.This change in plasticity can be attributed to the increasingly granular nature of lime-treated marine clay induced by short-term reactions (cation exchange and flocculation -agglomeration) (Bahmyari et al. 2021).In addition, cation exchange and the reaction between CaO in limestone and SiO 2 from natural pozzolana, which is considered a good source for the major components of secondary pozzolanic reactions (SiO 2 and Al 2 O 3 ) and is comparable with POFA used in recent work, have caused clay particles to come closer to one another and form a coarsegrained texture.Moreover, flocculation and agglomeration have changed the orientation of soil particles, reducing plasticity and improving the consistency of marine clay (Bahmyari et al. 2021).Nabil, Mustapha, and Rios (2020) reported a similar behavior.Figure 8d presents the results for SL with the addition of lime and POFA.The addition of lime alone and a combination of lime and POFA to marine clay reduced shrinkage with increasing lime and POFA contents.The SL of marine clay dropped from 11.4% to 9.3% and 2.5% with the addition of 5% lime alone and 5% lime with 20% POFA, respectively.A similar trend was reported by Al-Hokabi et al. (2021), who found that linear shrinkage was generally reduced with the addition of POFA and lime.When lime is added to soil, chemical reactions occur, including the replacement of naturally carried cations on the clay surface by Ca+ ions, an increase in pH value, and a decrease in the thickness of the double layer retained around soil particles.These reactions help in the flocculation and aggregation of clay particles, making them less plastic (Samantasinghar 2014).Such reduction improve soil behavior, because soil with low SL will not undergo desiccation cracking, and thus, is considered nonproblematic.

Specific gravity
Figure 9 shows that the specific gravity of the treated marine clay with added lime and POFA decreased as POFA and lime contents increased.This result is attributed to the lighter weight of the mixture (Al-Hokabi et al. 2021), and similar findings were reported by Yusoff (2016), Al-Hokabi et al. (2021), andAtahu, Saathoff, andGebissa (2019).The last used rice husk ash.Rice husk ash exhibits similar pozzolanic properties as POFA.Both materials depend on the presence and percentage of high silica and alumina contents.The specific gravity of POFA (2.23) and lime (2.41) was lower than that of marine clay.

Compaction tests
When marine clay is compacted at or above optimal moisture content, extremely significant shrinkage occurs with oven drying.MDD decreases with lime and POFA stabilization, while compaction effort remains constant.However, the ensuing strength gains normally exceed the compensation for the reduction of MDD.Figures 10 and 11a,b show that the MDD of marine clay decreased with the addition of 5% lime and decreased further with the addition of POFA.By contrast, OWC increased.The increase in OWC with lime and POFA is related to the high tendency of lime to absorb water (Abdullah et al. 2009), which occupies the pore space of the compacted specimen and make the particles buoyant (Yusof, Brown, and Umar 2011).The increase in OWC can also be explained in terms of calcium ions, which are released from lime or POFA and crowded out during the ionic dissociation of hydrolyzed calcium oxide, during the pozzolanic reaction between calcium ions and SiO 2 from POFA.Both factors have been identified as important in the chemical stabilization of soft clay (Alhokabi et al. 2022).Lime is known to reduce PI and MDD and to increase OWC (Firoozi et al. 2017).This information concurs with the findings of Al-Swaidani, Hammoud, and Meziab (2016) and Harichane et al. (2011), who determined that adding lime increased OWC and reduced MDD.
Furthermore, this reduction in MDD and increase in OWC may occur because lime causes particles to stick together and occupy a larger space (Alhokabi et al. 2022).In addition, the flocculation of soil  particles (Nabil, Mustapha, and Rios 2020) due to the addition of lime and the introduction of more fine particles in the form of POFA and lime, which have lower specific gravity (2.23 and 2.41, respectively), are counterweighted by marine clay, and thus, can occupy voids in soil (Alhokabi et al. 2022).The MDD and OWC results in the current study are in line with those of Al-Hokabi et al. (2021) and Piew and Shariff (2016), who found that POFA and lime replacement in soft soil reduced MDD but increased OWC.
Al-Hokabi et al. ( 2021) observed that the high volume of CaO in lime and the large surface area provided by POFA increased the water demands of the soil -binder mixture, generating an increased capacity to hold water within the flocs from soil flocculation.A similar behavior was reported by Renjith et al. (2021), who used fly ash.This fly ash and the POFA used in the current work have high silica and alumina contents.Firoozi et al. (2017) also demonstrated that lime decreases the PI and MDD of soil and increases its OWC.

Consolidation tests
Figure 12 shows the change in the void ratio (e) versus axial stress (kPa) of the untreated marine clay, lime-treated marine clay, and lime-treated marine clay mixed with different percentages of POFA.The change in void ratio was greatest in the untreated marine clay sample and decreased up to 42% with the addition of 5% lime.The changes in void ratio decreased further with the addition of different percentages of POFA (1%, 5%, 10%, 15%, and 20%) to the mixture of 5% lime-treated marine clay, where the decrements observed were up to 66.4% and 53% for S + 5% L + 1% POFA and S + 5% L + 5% POFA, respectively.This result may be attributed to the pozzolanic reaction that occurs between CaO from lime and SiO 2 or Al 2 O 3 from POFA, and the formation of cementation products that close voids in the mixture.A similar behavior was noted by Atahu, Saathoff, and Gebissa (2019), who used rice husk ash as a source of SiO 2 and Al 2 O 3 for the pozzolanic reaction.Rice husk ash has similar pozzolanic properties as POFA.Thus, the number of contact points between soil particles increases, leading to an increase in frictional forces (e.g., an increase in strength).A similar behavior was demonstrated by Jamsawang et al. (2017), who used bagasse ash waste, which can be considered a good source of SiO 2 for reacting with CaO from cement and exhibits similar pozzolanic properties as POFA.Bagasse ash waste and POFA depend on the presence and percentage of silica and alumina.The decrease may also be attributed to the addition of lime and POFA, which fill the inter-aggregate pores, reducing changes in void ratio.Changes in void ratio became unstable when POFA content reached 10% and above, probably because POFA content increased and CaO from lime was consumed.This CaO from lime is important for the reaction to continue.The addition of more POFA will not contribute to improvement, because POFA covers soil particles, and thus, only acts as a diffusion barrier for secondary pozzolanic reactions.Changes in void ratio increase with increasing load applied (axial stress, kPa) in the same mixture.
At 800 kPa, the changes in the void ratio of the untreated marine clay (S), S + 5% L, and S + 5% L + 1% POFA were 0.135, 0.086, and 0.05, respectively.In addition, the initial void ratio (e o ) of the mixture decreased as lime and POFA proportions increased under the same saturation conditions (Table 6).Table 6 provides the consolidation properties of untreated and treated marine clay.The current study observed reductions in e o of up to 3% and 17% for S + 5% L and S + 5% L + 5% POFA, respectively, and a continuous reduction in the void ratio for all the samples as load increased.The reduction in void ratio for the treated marine clay samples may be attributed to the closure of voids because they are filled with the formed cementitious products (e.g., CSH and CAH gel).According to Atahu, Saathoff, and Gebissa (2019), the high contents of silica and alumina from the pozzolanic material (rice husk ash or POFA) is responsible for producing these cementitious products.This phenomenon occurs when CaO from lime reacts with silica from soil or POFA through a secondary pozzolanic reaction or ion exchange capacity and produces a cementitious gel that fills voids and increases the number of contact points between soil particles.This finding is similar to the results of Jamsawang et al. (2017), who used bagasse ash waste as a pozzolanic material.Bagasse ash waste is similar to POFA in its effect as a good source of silica and alumina for reacting with CaO in secondary pozzolanic reactions.
Table 6 indicates that additives generate improvements in marine clay's compressibility properties.The compression index (C c ) and swelling index (C s ) of marine clay decreased with an addition of 5% lime alone and 5% lime with 1%, 5%, and 10% POFA.Thereafter, the behavior changed, and C c and C s started to increase as POFA content increased.Ultimately, the C c of marine clay was reduced by 61%, 68%, and 73% with the addition of 5% lime alone, 5% lime with 1% POFA, and 5% lime with 5% POFA, respectively.Meanwhile, C s was reduced by 66%, 70%, and 74%.This reduction in consolidation parameters (C c and C s ) might have arisen from the formation of cementitious products with the addition of lime and POFA, which filled inter-aggregate pores.This phenomenon, in turn, reduced compressibility characteristics and switched away from decrementing behavior.It might also arise from the critical role of lime's CaO in pozzolanic reactions.As POFA content increases, it consumes CaO, and adding more POFA will not contribute to improvement.Additional POFA will only work as a diffusion barrier for ion exchange or secondary pozzolanic reactions by covering soil particles.Consequently, the C c and C s indexes increase as POFA is added by up to 10%.The coefficients of compressibility (a v ) and volume compressibility (m v ) are also given in Table 6.The m v and a v of marine clay decreased as the proportions of POFA and lime increased, with up to 5% POFA.This result may be due to the reduced volume of voids owing to the rearrangement of soil particles, and a reduction in plasticity behavior as POFA content increased, making the soil more compact, and therefore, less compressible.The coefficients of consolidation (c v ), permeability (k), free swelling, and swelling pressure of marine clay also decreased with increasing proportions of POFA and lime, as indicated in Table 6.With the addition of lime and POFA, the free swelling and swelling pressure of the soil were reduced to nearly zero.Lime caused the production of agglomerates (flocculation and aggregation of clay), followed by the precipitation of hydrated phases that cemented the agglomerates into a more rigid mass, minimizing swelling (Benyahia et al. 2020).The addition of lime and POFA to marine clay reduces the sensitivity of soil to water and generates strong bonds between particles, which counteracts swelling.Prasad, Prasad, and Shivanarayana (2018) and Hozatlıoglu and Yılmaz (2021) reported similar findings.The fly ash used by Hozatlıoglu and Yılmaz (2021) can be considered similar to POFA in terms of its effect on soil behavior as a pozzolanic material because it also depends considerably on the presence of silica and alumina to produce cementitious products from pozzolanic reactions.Concluding that the addition of POFA and lime positively affects the compressibility of marine clay is reasonable.The standard deviation for the results in Part ( 1) is provided in Table 7.

Identification of a suitable amount of POFA for lime-treated marine clay
This study used compressibility criteria from various samples and numerous tests to determine a suitable percentage of POFA content to treat marine clay.After identifying that percentage, further tests were conducted in Part (2), with CS added to optimize the POFA -lime-treated marine clay and facilitate the study of CS's suitability as a replacement for lime in soil stabilization.The results presented in Table 6 show that increasing POFA content considerably enhanced the compressibility behavior of soil.With the addition of 5% POFA, the free swelling, a v , C s , C c , C v , k, and m v of the untreated marine clay or lime-treated marine clay were reduced from 1.15 mm to 0.019 mm, 0.00025 to 0.0000727, 0.0847 to 0.0215, 0.199 to 0.0531, 0.00253 to 0.001, 3.17 × 10 −8 to 4.69810 −9 , and 0.000125 to 0.000039, respectively.Therefore, a conclusion can be drawn that 5% is the optimum proportion of POFA that should be used to optimize the improvement of lime-treated soil.

Atterberg limits and SL
Figure 13 shows the effects of adding various percentages of CS as a ternary blending system with various percentages of lime and 5% POFA on the consistency properties of treated and untreated marine clay.The results showed that as CS increased, the plasticity properties of marine clay (LL, PI, and SL) decreased, as illustrated in Figures 13a,c,d.LL was reduced by 23%, 27%, and 28%; PI was reduced by 72%, 75%, and 86%; and SL was reduced by 41%, 46%, and 54% with the addition of 5%, 10%, and 15% CS, respectively, in the S + 5% L + 5% POFA mixture.Figure 13b shows the effect of adding CS, POFA, and lime on marine clay.The result may be attributed to CS's properties as a cementitious binder, operating as the pozzolan material and as a source of SiO 2 and Al 2 O 3 , which are necessary for a secondary pozzolanic reaction with calcium from lime.The reaction between CaO in lime and SiO 2 and Al 2 O 3 in CS and POFA leads to the formation of cementitious products (CSH and CAH).Consequently, voids in the soil are filled with cementitious gel, forcing clay particles to move closer to one another and form a coarse-grained texture, which has been further indicated via microstructural testing that compares the treated and untreated marine clay.Similar findings were reported by Harichane et al. (2011) andAl-Swaidani, Hammoud, andMeziab (2016), who utilized natural pozzolanas.Both studies used materials that exhibit similar pozzolanic properties to CS in the current study, i.e., a high percentage of silica and alumina, which caused the secondary pozzolanic reaction between CaO and SiO 2 and Al 2 O 3 .Reduced PI also indicates improved workability of the soil as plasticity is reduced (Harichane et al. 2011).The reduction in lime content from 5% to 1% and its replacement with CS caused PI to increase slightly (Figure 13).For example, the PI of S + 5% L + 5% POFA + 10% CS, S + 3% L + 5% POFA + 10% CS, and S + 1% L + 5% POFA + 10% CS was 59.5, 63.4, and 64.4,respectively.Meanwhile, PL decreased with a reduction in lime content.The higher the PI, the greater the quantity of water that the soil can absorb, and thus, its swelling potential is also increased.This increase in PI and LL can be attributed to a reduction in the chemical reactions among POFA, CS, and soil, which include ion exchange, secondary pozzolanic reactions, and associated flocculation reactions that resulted from a reduced lime content.The addition of lime to soft soil increases pH value and diminishes doublelayer water.Lime also plays an important role in these reactions as a source of CaO.In addition, lime helps in the flocculation and aggregation of clay particles, making them less plastic (Samantasinghar 2014).Harichane et al. (2011) noted a similar behavior, reporting that PI and LL increased while PL decreased with decreasing lime content.

Specific gravity
Figure 15 shows the effects of different percentages of CS on the specific gravity of POFA -limemarine clay mixtures.The addition of CS caused the specific gravity of these mixtures to decrease, i.e., the mixture became lighter than its original condition (untreated marine clay).Figure 16 compares the results of Parts ( 1) and ( 2).Notably, a reduction in lime content from 5% to 3% or 1% caused a slight increase in specific gravity.For example, the specific gravity of (S + 5% L + 5% POFA + 10% CS) was 2.55.When lime content was reduced to 3% and 1%, it rose to 2.61 and 2.62, respectively, due to the low specific gravity of lime.However, specific gravity remained less than that of the POFA -limetreated marine clay.

Compaction tests
Figures 17 and 18 present the compaction test results for Part (2).Evidently, the treatment that incorporated CS and POFA into lime-treated marine clay was highly effective in terms of enhancing the compaction characteristics of soil (Figures 17 and 18). Figure 17a,b illustrate the effect of adding CS and POFA into lime-treated marine clay on the compaction properties of marine clay, respectively.Meanwhile, Figure 18a-c, depict the effect of decreasing lime content from 5% to 3% and 1% on the compaction properties of marine clay.The addition of CS to the mixture of S + 5% L + 5% POFA increased MDD and decreased OWC.MDD increased by 6%, while OWC was reduced by 11% with the addition of 10% CS to the mixture S + 5% L + 5% POFA.Natural pozzolana has lower affinity to water, causing a drop in OWC.Al-Swaidani, Hammoud, and Meziab (2016) observed a similar behavior when using natural pozzolana, which exerts an effect similar to that of CS as a pozzolanic material.Both materials depend on high contents of silica and alumina.Calcination leads to the  formation of calcium precipitation, which affects cementation.This phenomenon results in closer bonding between particles, reducing compressibility properties and increasing MDD.Dry density increases because of decreased void ratio, which, in turn, is due to the closure of voids when cementitious products are generated from reactions between CaO from lime and silica from CS and POFA, through ion exchange capacity and pozzolanic reaction (Harichane et al. 2011).These constituents join soil particles, decrease voids, and increase the density of the mixture (Harichane et al. 2011). Al-Swaidani, Hammoud, andMeziab (2016), who made similar observations, also found that OWC decreased and MDD increased as natural pozzolana content increased.However, the behavior of the mixture in the present study changed: MDD decreased and OWC increased when 15% to 20% CS was added.The increase in MDD improved soil properties.In the current study, MDD began to fall when more than 10% CS was added to the mixture.The addition of CS to the mixture that is more than the amount required to react with calcium released from lime is believed to generate an effective diffusion barrier for pozzolanic reaction, because CS covers the particles of soil and stops water from reaching the reaction point.

Optimization of the mixtures and additive content of CS
Figure 19 shows 10% CS as the optimum proportion to be used as a partial replacement for lime in soil stabilization, enabling the reduction of lime from 5% to 3%.The mixture of 10% CS + 5% POFA + 3% L-treated marine clay provided better MDD and OWC than S, S + 5% L, and S + 5% L + 5% POFA.For example, MDD was 14.3, 13.81, 13.76, and 14.8 kN/m 3 , and OWC was 26%, 32%, 33%, and 30.2% for S, S + 5% L, S + 5% L + 5% POFA, and S + 3% L + 5% POFA + 10% CS, respectively.This soil improvement is due to the formation of cementitious products (CSH and CAH), as shown in Figure 20.The S + 3% L + 5% POFA + 10% CS mixture generated a smoother structure, with smaller voids than those in the untreated marine clay.
Figure 20b,c show the formation of CSH and the needle-shaped cementation material for S + 3% L + 5% POFA + 10% CS.Therefore, the optimum ternary blending mixture of the binder can be specified as a combination of 10% CS, 5% POFA, and 3% lime, because this combination was determined to be the most effective in stabilizing marine clay.Similar findings were reported by Cheng and Huang (2018), who used volcanic ash as a natural pozzolana and reported that the combination of 3% lime and 15% pozzolana was the best combination for meeting the performance specifications associated with roadbed construction.This result was attributed to the similar pozzolanic properties between the natural pozzolan used by Cheng and Huang (2018) and the CS used in the   current study, which both considerably depend on the percentage of silica and alumina to cause a secondary pozzolanic reaction between CaO limestone and SiO 2 and Al 2 O 3 from natural pozzolana.A smooth structure with smaller voids than those in the untreated marine clay is shown in Figure 20a.Cementing compounds, including CSH and CAH, were formed, as shown in Figures 20b,c.The formation of cementation gels and the filling of voids cause soil particles to bond, consequently increasing their strength and reducing their compressibility.These products are expected to be formed continuously as curing time progresses.These findings confirmed the efficiency and applicability of CS in soil property improvement and its potential as a partial replacement for lime in soil stabilization.

Microstructural analysis after a curing period of 7 days Structural and morphological analyses of S + 3% L + 5% POFA + 10% CS.
The findings are also supported by the EDX results, wherein chemical composition analysis indicated the presence of silicon and aluminum in the samples (Table 9).This result implies that lime, POFA, and CS have taken their place in the treated marine clay.Table 9 provides the weight percentages and major element ratios in the EDX samples at selected points 1, 2, and 3 (Figure 20c).EDX detected major elements (silicon, aluminum, and oxygen) and minor elements (calcium, sodium, and iron), all of which are basic components of natural soil.The presence and changes in the percentages of silicon and aluminum confirm the formation of a cementitious gel, indicating that large voids in the untreated marine clay (Figure 3) have diminished with the influx of the cementitious gel.

Mineralogical analysis.
In this study, XRD testing revealed some visible variations in mineralogy.The intensities of peaks linked to kaolin and montmorillonite decreased noticeably to the point of nearly disappearing with the S + 3% L + 5% POFA + 10% CS mixture, while those for quartz dropped moderately (Figure 21). Figure 21 presents the results of the XRD test on S + 3% L + 5% POFA + 10% CS.The declining intensity of kaolin and montmorillonite peaks can be attributed to their low intensity.By contrast, the pattern for quartz can be attributed to the formation of cementitious products.For the hydrate states, the amorphous structure of CSH gels makes them extremely difficult to identify with XRD.They will be reflected in hump forms around the 2θ values of 29.5°, 32°, and 50°, as reported by Kulasuriya et al. (2014).
Table 10 lists a series of microstructural studies conducted on two different samples: untreated marine clay and the optimal mixture of lime, POFA, and CS-treated marine clay from Part (2) (S + 3% L + 5% POFA + 10% CS) after 7 days of curing.XRF test was performed to determine the elemental composition of materials in this study.The presence and changes in the percentages of SiO 2 , Al 2 O 3 , and other elements in the samples treated with POFA, lime, and CS confirm the formation of the cementitious compound (CSH) caused by the pozzolanic reaction after a curing period of 7 days.This finding is in contrast with that of the untreated marine clay, and it is supported by the FESEM results shown in Figure 20.The percentage of Al 2 O 3 was considerably diminished following treatment with the additives because it was consumed in the pozzolanic reaction.By contrast, although SiO 2 was consumed in the pozzolanic reaction, its percentage increased, probably due to the addition of CS and POFA, which are important sources of SiO 2 elements (Table 2).Overall, the elements' percentages changed and declined with curing as pozzolanic reaction proceeded.

Conclusions
This study investigated the physical -mechanical and microstructural properties of marine clay stabilized with lime, POFA, and CS.The major conclusions drawn from this study are as follows.
• In Part (1), the PL of marine clay increased as lime was added but decreased with the addition of POFA.Increased POFA inclusion and lime content accompanied a reduction in LL, SL, Gs, PI, and MDD, while OWC increased.The e o of the mixtures decreased as lime and POFA proportions increased, while C c , C s, mv, av, Cv, and k decreased with the addition of lime and POFA.Therefore, the optimum proportion of POFA that should be used in Part (2) is 5%.• The addition of CS to POFA -lime-treated marine clay in Part (2) caused a reduction in LL, PI, SL, and OWC, but an increase in MDD.A similar behavior was observed when 5% lime was replaced with 3% or 1% lime.The increase in MDD indicated an improvement in soil properties, particularly a reduction in compressibility and an increase in strength due to the formation of CSH and CAH compounds.This finding was supported by microstructural tests.• This study shows that treatment that incorporates CS and POFA is highly effective for enhancing soil compaction and compressibility, indicating a potential use of CS in a mixture of POFAlime-marine clay in engineering practices, including road construction structures (subgrade) and buildings.This study identified the optimum mixture as 10% CS, 5% POFA, and 3% lime.

Highlights
• Marine clay has many problems, such as high compressibility and low shear strength.
• Calcined shale (CS) material has not been used for soil stabilization yet.
• Calcined shale (CS) decreases the production of CO2 emissions and sulfate attacks from the other supplementary cementitious materials (cement and lime) that cause environmental and structural issues.• Calcined shale (CS) enhance the strength and compressibility behavior of lime-treated marine clay.
• Calcined shale (CS) has the potential to be used as a partial replacement of lime in marine clay treatment.
Figure 1.a) location of the sampling point of the marine clay at Kuala Langat, Selangor, Malaysia, and b) marine clay at the site.

Figure 2 .
Figure 2. Particle size distribution for marine clay, lime, CS and POFA.

Figure 9 .
Figure 9.Effect of various percentages of POFA, lime and CS on specific gravity of Marine clay.

Figure 10 .
Figure 10.Dry unit weight-moisture content relationship of natural Marine clay and stabilized soil with 5% lime and different percentages of POFA.

Figure 11 .
Figure 11.Influence of 5% lime and various percentages of POFA on the a) maximum dry Density(MDD) and b) optimum water Content(OMC) of marine Clay.

Figure 12 .
Figure 12.Influence of 5% lime (L) and different percentages of POFA on the axial stress -changes in void ratio relationship of Marine clay.

Figure 14 .
Figure 14.Comparison between results of part (1) and the optimum proportion of part (2).

Figure 15 .
Figure 15.Effect of POFA, lime (L) and CS on the specific gravity of marine clay.

Figure 16 .
Figure 16.Results of specific gravity for part (1) and the optimum proportion of part (2).

Figure 17 .
Figure 17.Influence of blended lime (L), POFA and calcined shale (CS) on the unit weight and optimum water content of marine clay soil.

Figure 18 .
Figure 18.Compaction relationship of marine clay stabilized soil with lime, POFA and CS.
Figure 20 presents the results of FESEM testing.The images for S + 3% L + 5% POFA + 10% CS indicate that the microstructure of the untreated marine clay was significantly modified by adding CS, POFA, and lime.

Table 1 .
Physical properties of marine clay.

Table 2 .
The chemical composition of marine clay, CS and POFA.

Table 3 .
The physical and chemical composition of lime.

Table 6 .
Influence of 5% lime and different percentages of POFA on compression and consolidation parameters.

Table 7 .
Standard deviation for the results of part (1).
Table 8 provides the standard deviation of the results for Part (2).

Table 8 .
Standard deviation for the results of part (2).

Table 9 .
Energy dispersive X-Ray (EDX) element for different specimens at 7 days curing periods.

Table 10 .
X-Ray fluorescence test for different specimens at 7 days curing periods.