The Evolution of the Hydraulic Conductivity of Fly Ash-Treated Sand as a Liquefaction Countermeasure

Abstract

A soil improvement method that utilizes fly ash as a cementing material for liquefaction countermeasures has been examined. This study investigates the changes in the hydraulic conductivity of sand treated with fly ash, which has prospects for use as a liquefaction countermeasure. By tracking the development of hydraulic conductivity over time, the study seeks to determine how well fly ash mitigates liquefaction. The hydraulic conductivities were investigated through a series of constant head laboratory tests on sand specimens treated at varying proportions of fly ash, degree of saturation during curing, and curing time. The hydraulic conductivity results on treated sand were compared to that on untreated sand to assess the impact of fly ash on permeability reduction. The research findings indicate that adding fly ash to sand significantly reduces hydraulic conductivity compared to untreated sand. Furthermore, the study also shows that fly ash-treated sand’s hydraulic conductivity, k, decreases over time. This evolution is attributed to the pozzolanic reaction over time and the cementitious properties of the fly ash, leading to the formation of a cementitious gel, gradual densification, and a soil structure with reduced permeability. The results of this study provide valuable insights into the long-term behavior of sand treated with 5–10% fly ash as a liquefaction countermeasure. The evolution of hydraulic conductivity highlights the potential for sustained improvement in soil properties, making fly ash an effective solution for mitigating liquefaction hazards in geotechnical engineering and earthquake-prone regions.

Introduction

Fly ash, a by-product of coal combustion, is frequently used as a cementing material in ground improvement approaches [1, 2–3]. Others use it as a binder for liquefaction countermeasures being developed [4, 5]. One of the most effective methods for in situ soil improvement, particularly for the ground beneath existing buildings, is chemical grouting [6]. Large-scale treatments are believed to be unfeasible due to the injected fluid's high viscosity and quick rate of fluid hardening [7] since several injection wells are required to increase the enormous volume. As a coal combustion by-product, fly ash has been found to be a valuable option for chemically boosting soil stiffness and strength [8, 9–10]. In the presence of water, chemical processes separate the lime (CaO) from the fly ash and produce pozzolanic reaction and formation of cementitious gel. The voids are filled by the generated gels, consisting of calcium silicate hydrate (CSH) and calcium aluminate hydrate (CAH), which also lessen porosity. Reducing permeability and declining porosity will diminish the possibility of volumetric changes that could result in liquefaction.

The potential environmental benefits or concerns associated with the widespread use of fly ash as a liquefaction countermeasure include reducing environmental pollution, inexpensive material costs, having the ability to reduce CO₂ emissions, and controlling the greenhouse effect [2, 3, 11, 12]. In addition to reducing carbon emissions, the greener (low-carbon) options could offer a successful waste management plan. The greenest construction material with the most promise is geopolymer (cement-free) binders. Fly ash is an important component in sustainable construction, as it increases workability, lowers heat of hydration, increases resistance to sulfate attack, and stronger weather resistance [11]. As an industrial by-product, fly ash contains a substantial quantity of alumina (Al) and silica (Si), which are needed to create a robust and long-lasting geopolymer matrix and also used as source of material in creating of geopolymer binders.

The permeability of fly ash is influenced by the total concentration of calcium oxide (CaO) in the by-product mass. For ashes with the same particle size distribution, calcium fly ashes (class C fly ashes) exhibit much poorer hydraulic conductivity than do silicon ashes (class F fly ashes) [13]. The ability of water to travel through a substance is referred to as hydraulic conductivity, denoted by k. The permeability of fly ash with a higher CaO concentration decreases with time; whereas, the permeability of fly ash with a lower CaO content increases and remains steady. Different fly ash grain morphologies were used to explain this behavior. For example, fly ash grains with a greater CaO content had needle-shaped protuberances that have crystallized and are cause of the permeability to decrease with time [13].

The amount of fly ash that can be mixed with non-cohesive soils to act as a cementitious agent for liquefaction countermeasures can generally be determined by hydraulic conductivity [13]. The resilience of cemented soil to liquefaction can be increased and maintained by good drainage. The ease with which extra pore water pressure is discharged during an earthquake increases with hydraulic conductivity. It was concluded that the liquefaction resistance of treated sand is highly dependent on the bond strength and increases as the amount of binder increases [14]. However, bonds between grains can be harmed by preventing the extra pore water pressure that develops in the soil. The result is liquefied soil, that is, when the soil grains separate from one another, lose their strength and behave like a liquid. Keramatikerman et al. [4] conducted research on 2% fly ash-treated sand at various relative density DR of 20, 40, 60, and 80%. They concluded that liquefaction resistance increases with DR.

Class F fly ash with a fine and homogeneous grain size of less than 0.1 mm was assessed by Kaniraj and Gayathri [15]. In a consolidation cell with a falling head and a pressure of 5 kPa (field saturation), they evaluated the hydraulic conductivity of a gravitationally saturated material. Samples were crushed using the Modified Proctor procedure at their optimum water content (w_opt). For class F fly ash comprising sand silt grains, hydraulic conductivity k values between 10⁻⁷ and 4 × 10⁻¹⁰ m/s were obtained [16].

Palmer et al. [17] studied the hydraulic conductivity k for mixing bottom ash and fly ash (class F). The falling head method was used to perform the test in cells with a diameter of 10 cm under consolidation stress of 10–12 kPa. Samples had gravitationally saturated. The correlation between hydraulic conductivity and moisture content of fly ash during compaction was shown by the results of the test gained for fly ash of class F. When the moisture content (w) was higher than optimum, the highest results were attained. The lowest results were achieved at the same water content, which is more than 5% and 10% above the optimal moisture level, according to the Standard Proctor and Modified methods. It was shown to be impossible to lower hydraulic conductivity k below 10⁻⁹ m/s, the minimum permeability of soil as a barrier material, for any combination of water content and compaction [18]. However, after 7 days of curing time and adding 20–30% class C fly ash, the hydraulic conductivity began approaching the desired level. On the other hand, class C fly ash consolidated at lower moisture content had a hydraulic conductivity that was twice as high as that of samples compacted at the optimum moisture level.

Pal and Ghosh [19] researched the falling head hydraulic conductivity of the class F fly ash-montmorillonite clay mixtures. Conventional proctor compaction tests were used to evaluate the ideal moisture content and maximum dry density at which they should be compacted. The hydraulic conductivity values for fly ash decrease to around 10⁻¹⁰–10⁻¹¹ m/s from approximately 10⁻⁶–10⁻⁷ m/s for fly ash alone when fly ash is coupled with 50% montmorillonite clay. Zabielska-Adamska [20] examined the hydraulic conductivity of compacted fly ash samples using the Standard and Modified Proctor techniques at various moisture levels, including an optimum moisture content of 5%. Both compaction methods' lowest hydraulic conductivity value was attained at the highest moisture level, w = w_opt + 5%. The hydraulic conductivity dependency on fly ash water content was noticed by Palmer et al. [17]. They discovered that when the mixture's moisture content increased, the hydraulic conductivity k of the field mix decreased.

In fact, most of the scope of previous publications related to the prediction of hydraulic conductivity was limited to studies of materials evaluated at their optimum moisture content (w_opt). The other test relies entirely on the degree of compaction at optimum moisture content to get the highest density with the least hydraulic conductivity. Limited literature explains the findings of the permeability test of sand mixed with fly ash at a specified relative density DR, for example DR = 50%, with a certain saturation level that produces the greatest hydraulic conductivity that allows it to be used as a liquefaction countermeasure. This target relative density DR of 50% was chosen because medium density of sand is susceptible to liquefaction [21]. Although the mechanical properties of fly ash-treated sand are high at low degrees of saturation, the hydraulic conductivity is not known with certainty. Moreover, hydraulic conductivity of fly ash-treated sand at a specified relative density and at a certain saturation level probably quite different from that at the highest density and at the optimum moisture content.

On the other side, prediction of natural soil’s hydraulic conductivity using advanced modeling and simulation techniques, such as ANN (artificial neural networks) and MLR (multiple linear regression), has been carried out [22]. However, information regarding the application of this technique to predict the hydraulic conductivity of sand treated with fly ash is still limited, to be able to assist in the design and optimization of liquefaction mitigation strategies.

In light of the aforementioned problems, this study presents the findings from constant head laboratory tests on the hydraulic conductivity of sand treated with fly ash at various concentrations to determine how the saturation level during sample preparation affects the fly ash-treated sand's hydraulic conductivity. Numerous experimental variables were examined for their effects, such as the quantity of fly ash in the mixture, the saturation level utilized to create the samples, and the time the specimens were exposed to curing. The effect of evolution on the mixture's hydraulic conductivity was evaluated by applying fly ash as a liquefaction countermeasure. Investigations were also conducted on the mechanical characteristics, namely the unconfined compressive strength, or UCS, at the corresponding hydraulic conductivities. The amount of fly ash that can be used as a liquefaction countermeasure was finally verified based on the final hydraulic conductivity generated during curing. The use of fly ash as a liquefaction countermeasure based on the liquefaction resistance, that is the cyclic stress ratio (CSR) needed to produce 5% double amplitude axial strain in 20 cycles, generated from sand stabilized with fly ash needs to be clarified in further research.

Material and Methods

Materials

Sand, fly ash (FA), and water are the main components utilized in this study. The sand used in the experiment was extracted from the Pohara River in Southeast Sulawesi province, Konawe Regency, Indonesia, and passed through the number four sieve before being captured in sieve No. 200. Table 1 lists the specific gravities Gs of those sand-FA mixes. Before usage, the sand was heated in a drying oven. The grain size distribution of the sand itself is presented in Fig. 1. The sand has a specific gravity Gs of 2.66 and a mean grain size D50 of 0.465 mm was obtained from the graph presented. This value of D50 is within the limits for most liquefiable soil proposed by Tsuchida [23]. It means that the practical significance of mean grain size D50 of sand within the thresholds for most liquefiable soil indicates the susceptibility of sand resistance to liquefaction.

Table 1. Nii Tanasa fly ash's chemical compositions (in wt%) [25]

SiO2	19.88	MgO	8.72	K2O	2.23
CaO	24.04	TiO2	0.58	P2O5	0.18
Al2O3	12.00	MnO2	0.22	SO3	10.26
Fe2O3	12.63	Na2O	7.47	LOI	1.134

LOI loss on ignition

[See PDF for image]

Fig. 1

Grain size distribution of sand used

The second ingredient was the fly ash from the Nii Tanasa Steam Power Plant in Konawe Regency. The FA passed sieve No. 200 since its grain size was smaller than 0.075, and it has a specific gravity of 2.13. With a CaO content of more than 20%, the fly ash used in this study belongs to class C according to American Society for Testing and Materials (ASTM) C618 [24]. The Nii Tanasa fly ash's chemical compositions (in wt%) are presented in Table 1. Additionally, it behaves like cement and has pozzolanic qualities, which allows it to attach to the mixture. Tap water was the third ingredient utilized.

Methods

Testing of Materials

The specific gravity test (Gs) followed the Indonesian National Standard (SNI 03-1964-2008) [17], and the density test (to measure the pore-e ratio) was carried out for each percentage of mixed fly ash. The fly ash percentages were set at 5, 10, and 20% of the total dry mix weight to determine the properties of the sand treated with fly ash. Following the calculation of dry unit weights (${\gamma }_{\text{d}}$) using the data shown in Table 2, the void ratio (e) was determined following SNI 03-3637-1994 [26]. By assuming DR = 50% in advance, the void ratio (e) was calculated using Eq. (1).

1

$$\text{DR}=\frac{e_{max}-e}{e_{max}-e_{min}}$$

Table 2. Properties of fly ash-treated sand

Description	Values of the properties
	FA 5%	10%	20%
Gs	2.62	2.69	2.71
emax	1.01	1.01	0.97
emin	0.73	0.66	0.56
e at 50% DR	0.87	0.83	0.76
${\gamma }_{\text{d}}$ (kN/m3)	14.1	14.6	15.4

Sample Preparation

Five, ten, and twenty percent of the dry mixture's total weight of fly ash, respectively, were added to the sand mixture. After mixing the sand-fly ash mixes carefully, the mixture was placed into a plastic bag. A certain amount of water was added to the plastic bag to reach the predetermined degree of saturation of the specimen, either 30, 40, 50, 70, 80, or 90%. The fly ash, sand, and water mixture was then well-swirled in a plastic bag before being tightly sealed. This was done to prevent evaporation when preparing the sample. The next step was to pour the mixture that had been mixed evenly into the PVC mold having the dimensions of 11 cm in height and 5.5 cm in diameter. The mold was prepared by the author with the condition that the mold height was twice the diameter. Compaction was conducted in each layer up to five layers with kerfing on each surface. Then, the surface of the specimen was leveled and weighed up after the mold had been filled to ensure it reached the desired relative density of 50 ± 2%. The target relative density of 50% was impossible to accomplish precisely; hence, a tolerance of ± 2 is established. The specimens were then stored for 1–4 months in a controlled room with constant humidity and temperature at 25 °C. Each specimen was then ready to undergo subsequent tests for hydraulic conductivity (k) and unconfined compressive strength (UCS) on the same specimen.

Testing of Specimen

Following the curing time, the mold containing specimen was set up for permeability testing. Water was first allowed to circulate to inundate the specimen at a constant height of 5.5 cm above the specimen surface. It was presumed that the samples were saturated since air vents were left open long enough for air bubbles to be expelled from the specimen. The hydraulic conductivity was measured using the constant head permeability test after saturation. Water was constantly allowed to flow through the sample for a set period, after which it was collected in a measuring cup to evaluate hydraulic conductivity.

The hydraulic conductivity k at a constant hydraulic gradient was calculated using Darcy's law as follows:

2

$$k=\frac{Q}{A·i·t}$$where Q is the fluid’s volumetric discharge over the specified cross section area A, t is the flow's duration, and i is the hydraulic gradient, which is represented by the head loss per unit length:

3

$$i=\frac{\mathrm{\Delta }h}{L}$$

L represents the sample height, and h represents the variation in total pressure head (fluid levels).

Unconfined compressive strength (UCS), q_u, was tested on the same specimen immediately following the hydraulic conductivity test. Please refer to the author's paper presented on Simatupang et al. [25] for more information about UCS testing procedures.

Condition of the Testing

The test conditions were carefully planned to investigate the factors' effects such as: FA, CT, and Sr on the hydraulic conductivity k of fly ash-treated sand. Three different fly ash percentages, i.e., 5, 10, and 20%, were used for the hydraulic conductivity k tests. These ingredients were chosen based on the findings of earlier tests conducted by past researchers [25, 27, 28]. They observed that adding just 5% of fly ash may significantly boost the shear strength of the fly ash-treated sand, making it nearly ten times stronger than untreated sand. Additionally, the unconfined compressive strength of the sand significantly increased when it was treated with fly ash of at least 20% and allowed to cure for at least 1 month [25]. In the current investigation, the curing times lasted for 1–4 months.

Saturations Sr ranged from 30 to 90% with 10% intervals during sample preparation. The predetermined saturation was achieved by figuring out the volume of water (V_w) required at the specified Sr, as per:

4

$$\text{Sr}=\frac{1+e}{e}·\frac{V_{\text{w}}}{V};\left(V=\text{total volume}\right)$$

The volume of water in each saturation Sr was applied to each of the three predetermined fly ash percentages, and it was possible to observe which mixture produced the largest k in each curing time. All specimens had a targeted relative density, DR, of 50%. The quantity of materials utilized in this study was determined and is displayed in Table 3. Regardless of Sr, the combination's fly ash concentration (in percent of weight) determines the weight of dried sand. The concentration of fly ash and saturation Sr significantly impact the water volume needed.

Table 3. Quantity of material utilized

FA	Material	Weight of the material needed (gr)
(%)	of specimen	Sr 30%	40%	50%	70%	80%	90%
5	Total	403.81	415.91	428.04	452.29	464.42	476.62
	Sand	349.03	349.03	349.03	349.03	349.03	349.03
	FA	18.37	18.37	18.37	18.37	18.37	18.37
	Water	36.41	48.51	60.64	84.89	97.02	109.22
10	Total	418.32	430.2	442.09	465.85	477.74	489.62
	Sand	344.4	344.4	344.4	344.4	344.4	344.4
	FA	38.27	38.27	38.27	38.27	38.27	38.27
	Water	35.65	47.53	59.42	83.18	95.07	106.95
20	Total	435.82	447.13	458.43	481.04	492.35	503.65
	Sand	321.53	321.53	321.53	321.53	321.53	321.53
	FA	80.38	80.38	80.38	80.38	80.38	80.38
	Water	33.91	45.22	56.52	79.13	90.44	101.74

Results and Discussions

Hydraulic Conductivity

The hydraulic conductivity k was determined using results from trials on treated samples using fly ash with constant head permeability test. The hydraulic conductivity of the pure Pohara sand utilized in this study was also tested as a control. The result shows that the sand’s hydraulic conductivity is 4.76 × 10⁻⁴ m/s at DR 50%. Meanwhile, the results of hydraulic conductivity measurements of sand treated with fly ash in this study are summarized in Table 4 and shown graphically in Figs. 1, 2 and 3. The relationship between the hydraulic conductivity and the fly ash percentages, sample preparation saturations, and curing periods is plotted. The findings of those tests may be used to assess how fly ash-stabilized sand's hydraulic conductivity has changed over time. The main effect of the additive was a decrease in the treated sand's hydraulic conductivity.

Table 4. Hydraulic conductivity of fly ash-treated sand

Curing time (month)	FA (%)	Hydraulic conductivity, k (× 10−2 m/s)
		Sr 30%	40%	50%	70%	80%	90%
1	5	0.01362	0.01311	0.01275	0.01096	0.01088	0.00850
	10	0.00894	0.00888	0.00738	0.00678	0.00594	0.00577
	20	0.00850	0.00802	0.00561	0.00238	0.00216	0.00155
2	5	0.01160	0.01079	0.00923	0.00635	0.00624	0.00501
	10	0.00850	0.00738	0.00624	0.00468	0.00445	0.00203
	20	0.00508	0.00482	0.00413	0.00096	0.00078	0.00077
3	5	0.00638	0.00602	0.00558	0.00447	0.00369	0.00146
	10	0.00536	0.00505	0.00425	0.00390	0.00270	0.00124
	20	0.00447	0.00425	0.00385	0.00089	0.00069	0.00060
4	5	0.00590	0.00573	0.00501	0.00355	0.00296	0.00124
	10	0.00453	0.00407	0.00381	0.00281	0.00178	0.00105
	20	0.00251	0.00226	0.00221	0.00074	0.00029	0.00027

In general, the hydraulic conductivity k decreases as fly ash content in the sand increases. However, one of the crucial elements in liquefaction countermeasures is the hydraulic conductivity of the soil. The excess pore water pressure generated during an earthquake dissipates more quickly when the soil has greater hydraulic conductivity. The primary objective of this study is to determine the proportion of fly ash in a mixture that retains hydraulic conductivity over time as a consequence of the pozzolanic activity. Good hydraulic conductivity is produced by the sufficient pore space that is not covered by the fly ash gel (CSH and CAH), which can lessen the excessive pore water pressure generated during earthquakes. The fly ash gel generated might reduce soil pores, but on the other hand it will bind the soil grains more tightly, thereby reducing the probability of liquefaction occurrence. In this case, the formation of the fly ash gel generated is very important, namely binding the soil grains without excessively covering the pores.

Effect of Fly Ash on the Hydraulic Conductivity

The data from 72 treated samples with 5, 10, and 20% fly ash content are displayed in Figs. 2, 3, 4 and 5. According to an analysis of the test findings in Fig. 2, the hydraulic conductivity of the samples treated with fly ash varies with increasing percentage of fly ash content. It should be noted that the fly ash content causes the hydraulic conductivity k value in Fig. 2 to decrease. This is due to the cementitious gel generated as a consequence of the pozzolanic reaction. Cementitious gel partially covers the pores of the sand so that its hydraulic conductivity decreases. In the literature, similar occurrences have been recorded [29].

[See PDF for image]

Fig. 2

Hydraulic conductivity of fly ash-treated sand at different fly ash percentages and saturations: a CT 1 month, b CT 2 months, c CT 3 months, d CT 4 months

As the proportion of fly ash in mixtures grew, the hydraulic conductivity values of samples treated with fly ash invariably dropped. Generally, the decrease in permeability with an increase in curing time is explained by the development of the flocs due to pozzolanic reaction. The formation of particle aggregates results in the soil becoming more granular in nature and results in higher resistance to compression at comparable stress level. This produces a soil with a more open fabric and results in an increase in hydraulic conductivity. However, under high level stresses, the pore size gradually decreases leading to reduction of the permeability of the stabilized soils.

The use of fly ash as a liquefaction countermeasure based on the hydraulic conductivity investigated in this study seems agree with that based on the liquefaction resistance observed by other researchers. Kolay et al. [30] conducted a cyclic triaxial test on Ottawa sand for fly ash content ranging from 0 to 70% at a CSR range of 0.1–0.5 with a constant DR of 50%. They found that for fly ash content from 10 to 25%, the liquefaction resistance increased. However, as the fly ash content increases beyond 25%, the resistance to liquefaction decreases. Although the mechanical properties of fly ash-treated sand increase with increasing fly ash content, this does not appear the case for liquefaction resistance. The cause is thought to be a decrease in hydraulic conductivity along with increasing fly ash content in the mixture, as is the result of the current study. Excess pore water pressure that is blocked in the pores destroys the bonds between the sand grains that are formed. However, the case of liquefaction resistance of fly ash-treated sand is beyond the scope of the current study, so further research is needed.

The test results of other researchers, Keramatikerman et al. [4], also show the same thing. A series of cyclic triaxial tests were performed in order to determine the liquefaction resistance of fly ash-treated sand. The addition of fly ash from 4 to 6% with the same CSR and DR of 0.2 and 20% respectively shows an increase in the liquefaction resistance. As additional information, although it is beyond the scope of this paper, at various DRs ranging between 20 and 80% and at a constant fly ash content of 2%, the results of their research show an increase in liquefaction resistance along with DR.

Effect of Saturation on the Hydraulic Conductivity

Figure 3 depicts how saturation Sr during sample preparation affected the treated samples’ hydraulic conductivity. The analysis of the laboratory findings revealed that the mixture's saturation had a role in the hydraulic conductivity of samples that had been treated with fly ash. Greater saturation levels in treated samples resulted in lower hydraulic conductivity values; whereas, lower saturation levels in treated samples led to greater hydraulic conductivity values. According to Fig. 3, adding 5% fly ash reduced the treated sand's hydraulic conductivity by around 38 times at 90% saturation level Sr during the 4-month curing period. While fly ash treatment of sand at a 30% saturation level only resulted in an eight-fold decrease in hydraulic conductivity.

[See PDF for image]

Fig. 3

Hydraulic conductivity of fly ash-treated sand at different saturations and fly ash percentages: a CT 1 month, b CT 2 months, c CT 3 months, d CT 4 months

There was a reduction in the hydraulic conductivity of bare sand (k = 4.76 × 10⁻⁴ m/s) compared with sand treated with 20% fly ash at saturation Sr of 30% and 90% of around 19 and 174 times, respectively. The hydraulic conductivity decreased more than nine times, from 2.51 × 10⁻⁵ to 2.7 × 10⁻⁶ m/s, by increasing the saturation from 30 to 90%. This tendency of roughly 174 times reduction in hydraulic conductivity took place starting from 70% saturation to a full saturation during sample preparation. This means the reduction in hydraulic conductivity is greater at higher saturations and vice versa at lower saturations.

In order to determine the gel formation in the sand more clearly, SEM (Scanning Electron Microscope) images were performed on fly ash-treated sand. Figure 4 shows the microstructure of the 20% fly ash-treated sand at different saturation levels. The gel formation generated in the sand treated at the two different saturations of 30% and 90% was dissimilar. At lower saturation, the gel formation was concentrated at the contact point leaving pores between the grains. At higher saturation, however, the gel formation was spread evenly throughout the sand surface and covers the sand pores. The hydraulic conductivity will consequently drop to almost zero.

[See PDF for image]

Fig. 4

SEM images of 20% fly ash-treated sand at different saturation of a 30%, b 100% [10]

Therefore, it is not recommended to prepare the addition of fly ash into the soil as a liquefaction countermeasure at high saturation. In this situation, the liquefaction caused by blocking the pore spaces would prevent the release of extra pore water pressure. Based on the test results, adding roughly 5% fly ash could significantly impact the hydraulic conductivity of the fly ash-treated sand. This quantity of fly ash gave the highest hydraulic conductivity of the sand treated with fly ash observed in this study. As a result, the data show that adding more fly ash to the sand as an additive may result in less hydraulic conductivity, especially at higher saturation during preparation of the specimen.

Effect of Curing Time on the Hydraulic Conductivity

According to Fig. 5's graphs, the stable state in the evolution of hydraulic conductivity values was attained after 3 months of curing time for all samples treated with 5–10% fly ash and after 2 months for specimens treated with 20% fly ash. It denotes that the pozzolanic reaction producing cementitious gel was essentially completed, especially in specimens treated at greater fly ash content and saturation. Alternatively, it could be that the created cementitious gel had filled all of the pore space, providing a hydraulic conductivity close to zero. However, at lower saturation, the 20% fly ash-treated sand's hydraulic conductivity decreased steadily during 4 months of curing time. At lower saturations of 30–50%, the trend and value of the hydraulic conductivity evolution of sand treated with fly ash were nearly identical. However, there was a noticeable difference between the hydraulic conductivity of sand treated at lower and higher saturation, especially with a higher fly ash content of 20%, as shown in Fig. 5c. This was due to the effectiveness of fly ash gel formation in binding soil granules.

[See PDF for image]

Fig. 5

Hydraulic conductivity of fly ash-treated sand at sundry curing times and saturations: a FA 5%, b FA 10%, c FA 20%

Unconfined Compressive Strength

On the same fly ash-treated sand specimen, the Unconfined Compressive Strength (UCS), q_u, test was carried out immediately following the hydraulic conductivity test. The test findings are presented in Table 5 and depicted graphically in Figs. 6 and 7 at various curing time CTs, fly ash FA percentages, and saturation degrees Sr during sample preparation. The typical stress–strain curves of UCS tests at various parameters observed are shown in Fig. 6. It is clear that the UCS increases sharply until it reaches the peak strength at a small strain of about 1% and drops suddenly also in this strain range especially at lower FA content. It indicates the fragile behavior of FA bonds, where small stresses/strains can cause FA bonds to fail.

Table 5. UCS of fly ash-treated sand

CT (month)	FA (%)	Unconfined compressive strength (UCS), qu (kN/m2)
		Sr 30%	40%	50%	70%	80%	90%
1	5	1.46	1.11	1.44	0.70	0.95	0.79
	10	7.82	6.15	4.98	2.14	2.46	2.11
	20	75.54	70.21	22.41	15.47	12.04	11.38
2	5	1.73	1.61	1.77	1.51	1.56	1.55
	10	9.61	6.48	5.46	3.29	3.64	3.54
	20	80.12	74.92	43.63	17.71	18.56	11.48
3	5	2.11	1.87	2.11	2.14	1.73	1.66
	10	10.68	9.63	6.59	4.23	3.75	3.65
	20	95.08	78.78	72.40	62.59	31.04	26.71
4	5	5.40	4.01	2.97	2.57	2.52	1.96
	10	11.33	10.70	6.73	5.94	4.55	3.74
	20	117.84	93.65	73.05	63.73	40.67	40.18

[See PDF for image]

Fig. 6

Typical stress–strain relationship at different a CT, b Sr, c FA

[See PDF for image]

Fig. 7

Unconfined compressive strength of fly ash-treated sand at any: a fly ash percentages, b saturations, and c curing times

On the other hand, higher strains can be achieved at higher fly ash content and saturation of about 20% and 90%, respectively at a higher CT of 4 months. Higher strain at higher FA content is a consequence of the bigger and stronger bonds produced during the curing period. At higher saturation, however, the cementitious gel produced as a binder appears to be more ductile and thus capable of producing greater strains.

Table 5 and Fig. 7 show that the UCS rises with fly ash FA percentage and curing time CT and falls with saturation Sr. The UCS increases significantly from a specimen treated at 10–20% fly ash, notably at lower saturation Sr and higher curing time CT. For instance, the rise reached more than tenfold, from 11.33 to 117.84 kN/m² at 10–20% fly ash FA percentage, respectively, with 30% saturation Sr and 4 months curing time CT, as shown in Table 5 and Fig. 7a. However, this gain in the specimens treated at 20% fly ash, as seen in Fig. 7b, declines noticeably with higher saturation. At 4 months curing time CT, UCS at saturation Sr of 30% was 117.84 kN/m², which was higher than UCS at saturation Sr of 90%, which was 40.18 kN/m². As shown in Fig. 7c, soil treated at low saturation during curing was more effective than soil treated at high saturation. The lower the saturation, the greater the UCS of fly ash-treated sand. The same trend was observed in similar investigations conducted by other researchers, particularly tests on soils treated with chemicals [25, 28, 31].

The longer the curing time on the other side, the higher the UCS of the fly ash-treated sand produced, as shown in Fig. 7c and listed in Table 5. The pozzolanic reaction is what causes the UCS of fly ash-treated samples to rise gradually over time [25]. This phenomenon was particularly pronounced in the specimens treated at higher fly ash content, particularly at lower saturation. This was, however, not noticeable for specimens treated with decreased fly ash concentration. The UCS did not undergo much evolution in this instance.

Unconfined Compressive Strength on the Appropriate Hydraulic Conductivity

The corresponding correlation between hydraulic conductivity and unconfined compressive strength (UCS) of fly ash-treated sand is shown in Fig. 8. Following the completion of the permeability test on the same material, the UCS test was immediately performed. The test findings show that fly ash alters the strength and permeability of the soil and the size of the soil voids. A soil's strength typically rises, and its permeability falls with fly ash because of the inclusion of a cohesive strength component brought on by the bonding of soil grains with hydration products and interconnectivity of the voids [25, 27, 32].

[See PDF for image]

Fig. 8

Unconfined compressive strength on the corresponding hydraulic conductivity

Figure 8 and Tables 3 and 4 make it abundantly clear that, for sand treated with fly ash, the UCS increased from 1.46 to 75.54 kN/m². At the same time, the hydraulic conductivity decreased from 1.362 × 10⁻⁴ to 8.50 × 10⁻⁵ m/s with a percentage change from 5% (marked with a boxed black dot) to 20% (marked with a circled black dot), with a curing time of 1 month. These evolutions result from the cementitious gel produced by the pozzolanic reaction; the cementitious gel production increases with fly ash content. The development of cementitious gel is directly proportional to the rise in strength and decrease in permeability.

On the other side, the main finding was due to hydration during curing of the specimens, for example, specimens treated at the same fly ash content of 20% and saturation of 30%:

• The UCS of treated sand, cured from 1 month (marked with circled black dot) to 4 months (marked with rhombus black dot), increased from 75.54 to 117.84 kN/m², and

• The hydraulic conductivity decreased from 8.5 × 10⁻⁵ to 2.51 × 10⁻⁵ m/s.

The mechanical properties, such as UCS in this example, were strengthened, and the hydraulic conductivity was decreased as a consequence of the hydration process and pozzolanic activity over time. The mechanical and hydraulic properties of the specimens treated at 40% saturation (indicated with a red dot) followed the same trend as those previously used as case examples and treated at 30% saturation.

Given the ensuing hydraulic conductivity for liquefaction countermeasure, it is advised to utilize treated sand at 5–10% fly ash concentration and 30–50% degrees of saturation during curing. Even if the UCSs generated under those circumstances were considerably smaller than the UCS specimen treated at 20% fly ash, according to research by Simatupang et al. [18], their shear strength might be at least five times more than that of bare sand. The mix's qualities are improved by adding fly ash to the sand. As seen in Fig. 8 and Table 3, adding 5% fly ash can be determined as a mixture with the highest hydraulic conductivity to be used as an additive for liquefaction countermeasures. Sand treated with 10% fly ash also exhibited hydraulic parameters comparable to those treated with 5% fly ash. The hydraulic conductivity drops from 5.90 × 10⁻⁵ m/s in the prior case of 5% fly ash to 2.51 × 10⁻⁵ m/s when fly ash increases to 20%. These evolutions in hydraulic conductivity occurred in a specimen cured for 4 months and treated for 30% saturation.

The long-term durability and stability implications of the changes in hydraulic conductivity in fly ash-treated sand, especially when exposed to environmental factors such as rainfall and groundwater infiltration, are a decrease in suction due to increased pore water pressure [33]. Increasing pore water pressure due to decreasing hydraulic conductivity over time causes the shear resistance reduced. In this case, the function of releasing excess pore water pressure from the soil pores is lost.

Conclusions

Based on the findings of the hydraulic conductivity test of sand stabilized with fly ash, it can be concluded that:

The hydraulic conductivity of sand treated with fly ash is highly dependent on the amount of fly ash used. The greater the amount of fly ash used, the lower the hydraulic conductivity produced. This is due to the cementitious gel generated as a consequence of the pozzolanic reaction which gradually covers a portion of the pore space.

The formation of fly ash gel on the sand surface is greatly influenced by the degree of saturation during sample preparation. At low saturation, the fly ash gel generated will concentrate on the contact surface between sand grains and leave pore space. As a result, hydraulic conductivity will remain well maintained. On the other hand, at high saturation, the fly ash gel produced will cover the entire surface of the sand and cover the pores. The hydraulic conductivity resulted from the fly ash gel generated over time will move toward zero.

Hydraulic conductivity tends to decrease over time due to the pozzolanic reaction which produces cementitious gel before ceasing. There are two basic mechanisms that cause changes in hydraulic conductivity to cease, namely: the pozzolan reaction has been completed (hydraulic conductivity: k > 0), or the fly ash gel generated has covered all the pores (hydraulic conductivity: k ~ 0).

Soil mixtures, including sand and 5–10% fly ash at lower saturation degrees of 30–50% during sample preparation, perform better in terms of hydraulic conductivity than those at greater saturation levels. Accordingly, it can be anticipated that adding fly ash to a sand mixture at a concentration of 5–10% causes the smallest decrease in hydraulic conductivity and is advised for usage as an additive for liquefaction countermeasure.

According to test results, fly ash can be utilized as an additive in soil improvement techniques to increase the problematic soil's resistance to liquefaction. Additionally, by-product materials that could otherwise provide disposal issues will find a helpful end use by being used as additives, which will lower the cost of soil improvement.

Acknowledgements

The authors would like to thank Mr./Miss. Fidel Kalehangen Osok, Arjuna Bulo, Iin Andariska, and Auli Saputri for supporting this research. The researcher also thanks all those who have contributed to this research. We thank the Ministry of Education, Culture, Research and Technology of the Republic of Indonesia for these funds. We appreciate your support.

Funding

The Ministry of Education, Culture, Research and Technology of the Republic of Indonesia financed the study that produced these findings under Grant Agreement No. 26/UN29.20/PG/2023.

Declarations

Conflict of interest

The authors declare that they have no competing interests.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Parsons, RL; Kneebone, E. Field performance of fly ash stabilised subgrades. Ground Improv; 2005; 9, pp. 33-38. [DOI: https://dx.doi.org/10.1680/grim.2005.9.1.33] 1276.68151

2. Lirer, S; Liguori, B; Capasso, I; Flora, A; Caputo, D. Mechanical and chemical properties of composite materials made of dredged sediments in a fly-ash based geopolymer. J Environ Manag; 2017; 191, pp. 1-7. [DOI: https://dx.doi.org/10.1016/j.jenvman.2017.01.001]

3. Capasso, I; Lirer, S; Flora, A; Ferone, C; Cioffi, R; Caputo, D; Liguori, B. Reuse of mining waste as aggregates in fly ash-based geopolymers. J Clean Prod; 2019; 220, pp. 657-673. [DOI: https://dx.doi.org/10.1016/j.jclepro.2019.02.164]

4. Keramatikerman, M; Chegenizadeh, A; Nikraz, H. Experimental study on effect of fly ash on liquefaction resistance of sand. Soil Dyn Earthq Eng; 2017; 93, pp. 1-6. [DOI: https://dx.doi.org/10.1016/j.soildyn.2016.11.012]

5. Keramatikerman, M; Chegenizadeh, A; Nikraz, H; Sabbar, AS. Effect of flyash on liquefaction behaviour of sand-bentonite mixture. Soils Found; 2018; 58, pp. 1288-1296. [DOI: https://dx.doi.org/10.1016/j.sandf.2018.07.004]

6. Wei-Soon, N; Lee, M-L; Hii, S-L. An overview of the factors affecting microbial-induced calcite precipitation and its potential application in soil improvement. World Acad Sci Eng Technol; 2012; 6, pp. 685-691.

7. van Paassen, LA; Daza, CM; Staal, M; Sorokin, DY; van der Zon, W; van Loosdrecht, MCM. Potential soil reinforcement by biological denitrification. Ecol Eng; 2010; 36, pp. 168-175. [DOI: https://dx.doi.org/10.1016/j.ecoleng.2009.03.026]

8. Prabakar, J; Dendorkar, N; Morchhale, RK. Influence of fly ash on strength behavior of typical soils. Constr Build Mater; 2004; 18, pp. 263-267. [DOI: https://dx.doi.org/10.1016/j.conbuildmat.2003.11.003]

9. Trzebiatowski BBD, Edil TB, Benson CH (2004) Case study of subgrade stabilization using fly ash: state highway 32, Port Washington, Wisconsin. In: Recycled materials in geotechnics, pp 123–136 (2004)

10. Simatupang, M; Edwin, RS; Sulha,; Mustika, W; Putra, H; Yanto, DHY. Splitting tensile strength of fly ash-modified sand at various saturations and curing times. Infrastructures; 2022; 7, pp. 1-16. [DOI: https://dx.doi.org/10.3390/infrastructures7100126]

11. Shukla, BK; Gupta, A; Gowda, S; Srivastav, Y. Constructing a greener future: a comprehensive review on the sustainable use of fly ash in the construction industry and beyond. Mater Today; 2023; 93, pp. 257-264.0687.46043

12. Mishra, J; Nanda, B; Patro, SK; Krishna, RS. Sustainable fly ash based geopolymer binders: a review on compressive strength and microstructure properties. Sustainability; 2022; 14, pp. 1-41. [DOI: https://dx.doi.org/10.3390/su142215062]

13. Porbaha, A; Pradhan, TBS; Yamane, N. Time effect on shear strength and permeability of fly ash. J Energy Eng; 2000; 126, pp. 15-31. [DOI: https://dx.doi.org/10.1061/(asce)0733-9402(2000)126:1(15)] 1292.35161

14. Simatupang, M; Okamura, M. Liquefaction resistance of sand remediated with carbonate precipitation at different degrees of saturation during curing. Soils Found Jpn Geotech Soc; 2017; 57, pp. 619-631. [DOI: https://dx.doi.org/10.1016/j.sandf.2017.04.003] 1367.05147

15. Kaniraj, SR; Gayathri, V. Permeability and consolidation characteristics of compacted fly ash. J Energy Eng; 2004; 130, pp. 18-43. [DOI: https://dx.doi.org/10.1061/(asce)0733-9402(2004)130:1(18)] 1508.81450

16. Zabielska-Adamska, K. Shear strength parameters of compacted fly ash-HDPE geomembrane interfaces. Geotext Geomembr; 2006; 24, pp. 91-102. [DOI: https://dx.doi.org/10.1016/j.geotexmem.2005.11.006]

17. Palmer, BG; Edil, TB; Benson, CH. Liners for waste containment constructed with class F and C fly ashes. J Hazard Mater; 2000; 76, pp. 193-216. [DOI: https://dx.doi.org/10.1016/S0304-3894(00)00199-0] 0853.55018

18. Benson, CH; Daniel, DE. Minimum thickness of compacted soil liners: II. Analysis and case histories. J Geotech Eng; 1994; 120, pp. 129-152. [DOI: https://dx.doi.org/10.1061/(ASCE)0733-9410(1994)120:1(129)] 0453.65046

19. Pal, SK; Ghosh, A. Hydraulic conductivity of fly ash-montmorillonite clay mixtures. Indian Geotech J; 2012; 43, pp. 47-61. [DOI: https://dx.doi.org/10.1007/s40098-012-0033-3] 0902.73026

20. Zabielska-Adamska, K. Hydraulic conductivity of fly ash as a barrier material: some problems in determination. Environ Earth Sci; 2020; 79, pp. 1-13. [DOI: https://dx.doi.org/10.1007/s12665-020-09070-8] 1128.68403

21. Tokimatsu, K; Seed, B. Evaluation of settlements in sands due to earthquake shaking. J Geotech Eng; 1987; 113, pp. 861-878. [DOI: https://dx.doi.org/10.1061/(ASCE)0733-9410(1987)113:8(861)] 0860.68020

22. Williams, CG; Ojuri, OO. Predictive modelling of soils’ hydraulic conductivity using artificial neural network and multiple linear regression. Nat Appl Sci; 2021; 3, pp. 1-13.1490.16066

23. Tsuchida H (1970) Prediction and countermeasure against liquefaction in sand deposits. In: Abstract. Seminar in the Port and Harbor Research Institute. Ministry of Transportation, Yokosuka, Japan, pp 3.1–3.33

24. ASTM. Standard specification for coal fly ash and raw or calcined natural Pozzolan for use in concrete; 2005; West Conshohocken, ASTM:

25. Simatupang, M; Mangalla, LK; Edwin, RS; Putra, AA; Azikin, MT; Aswad, NH; Mustika, W. The mechanical properties of fly-ash-stabilized sands. Geosciences; 2020; 10, pp. 1-19. [DOI: https://dx.doi.org/10.3390/geosciences10040132]

26. SNI 03–3637–1994 (1994) Soil unit weight test method (in Bahasa Indonesia)

27. Simatupang, M; Mangalla, LK; Muhaimin, LO; Aldy, M; Aji, IC; Jalbanirah,; Jamin, M; Marteni, LOA; Rahmat,; Ardiansyah,. The ultimate shear strength of the fly ash-improved sands. IOP Conf Ser Earth Environ Sci; 2021; 871, pp. 1-8. [DOI: https://dx.doi.org/10.1088/1755-1315/871/1/012061]

28. Simatupang, M. Effectiveness of lowering saturation on residual shear strength of sand stabilized with fly-ash. IOP Conf Ser Earth Environ Sci; 2021; 622, 012003. [DOI: https://dx.doi.org/10.1088/1755-1315/622/1/012003]

29. Sobti, J; Singh, SK. Hydraulic conductivity and compressibility characteristics of bentonite enriched soils as a barrier material for landfills. Innov Infrastruct Solut; 2017; 2, pp. 1-13. [DOI: https://dx.doi.org/10.1007/s41062-017-0060-0]

30. Kolay, PK; Puri, VK; Lama Tamang, R; Regmi, G; Kumar, S. Effects of fly ash on liquefaction characteristics of Ottawa sand. Int J Geosynth Ground Eng; 2019; 5, pp. 1-14. [DOI: https://dx.doi.org/10.1007/s40891-019-0158-x]

31. Cheng, L; Cord-Ruwisch, R. In situ soil cementation with ureolytic bacteria by surface percolation. Ecol Eng; 2012; 42, pp. 64-72. [DOI: https://dx.doi.org/10.1016/j.ecoleng.2012.01.013]

32. Canakci, H; Sidik, W; Halil, IH. Effect of bacterial calcium carbonate precipitation on compressibility and shear strength of organic soil. Soils Found; 2015; 55, pp. 1211-1221. [DOI: https://dx.doi.org/10.1016/j.sandf.2015.09.020] 1375.13029

33. Fourie, AB. Relying on suction to maintain slope stability; 2016; Perth, Australian Centre for Geomechanics: pp. 3-12. [DOI: https://dx.doi.org/10.36487/ACG] 0938.47020