Evaluation of mechanical behavior and microstructural studies of paper sludge ash-treated fine grained soil: A step towards sustainable building materials

Abstract

It is well known that waste generation and its poor disposal practices have adverse environmental impacts. This study investigates a sustainable solution to improve soil properties using paper sludge ash (PSA), focusing on the physical, mechanical, and microstructural characterization of the stabilized material for building materials and field applications. Different tests were used to study the engineering characteristics of clay soil, which included standard proctor compaction, Atterberg limits, unconfined compressive strength (UCS) and pH. Besides, microstructural changes and mineralogical compositions were assessed using FESEM, XRD, and XRF. Effects of PSA content (0, 2.5, 5 and 10% by dry weight of soil), curing time (1, 7 and 28 days) and dry density (14, 15 and 16 kN/m³) on properties of treated soil were considered. The results indicated that maximum dry density (MDD) decreased with increasing PSA content and curing time, whereas optimum moisture content (OMC) increased. As the PSA content and curing time increased, the plasticity index (PI) decreased. The results indicated that the UCS of the treated soil increased with higher PSA content, curing time, and dry density. The maximum UCS was observed in the 10% PSA-treated soil. Driven by extended curing time and increased dry density, this value showed a 19.94-fold increase compared to untreated soil. The treated soil exhibited an increase in pH with higher PSA content, whereas a downward trend was observed with extended curing time. Microstructural investigations indicated that the addition of PSA has modified the soil structure. Finally, the various statistical methods such as correlation, regression and sensitivity analyses were applied.

Introduction

The scientific advancement has paved the way for greater awareness in creating conditions that balance the environment with economic and social considerations. It is sustainable development that aims to establish such a balance¹. Recently, sustainable development has been the common denominator between decision makers, academic researchers, and industry professionals. The construction companies apply these sustainable building practices in order to reduce pollution caused due to construction activities, minimize environmental degradation, and avoid the depletion of natural resources². Every construction in this world, such as a bridge, a building, a road, or a dam, stands on soil. Hence, soil must possess certain engineering properties sufficient to carry the structures imposed upon it. It needs to show certain degrees of resistance to the loads imposed and some levels of deformation and settlement acceptable to the designer. Soil treatment is one of the methods used to enhance the characteristics of weak soils, allowing structures to be safely constructed on them^3,4.

Soil chemical stabilization is a widely used method for soil treatement⁵. Under this method, cement and lime are used for soil strength development; after chemical reactions have formed, resistance, stiffness, and durability characteristics of soils are modified. For example, mixing water with cement yields calcium silicate hydrate (CSH) and calcium aluminate hydrate (CAH), which tightly bonds soil particles together⁶. With increasing the cement content in treated soil, increases in compressive strength resulted along with great drops in permeability^7,8. Application of lime to clay soil enhanced substantially the modulus of elasticity⁹. Addition of lime and natural pozzolanic material to silty sand improved the California bearing ratio¹⁰. The cement and lime industry greatly influences the improvement of the mechanical properties of soil; however, the production of cement and lime is a high energy consuming process accompanied by emissions of carbon dioxide and nitrogen oxides with air pollution and numerous environmental issues^5,11,12. Moreover, with global development and the industrialization of societies, a massive volume of waste has been generated by various industries over the years. Substantial quantities of this waste and problems faced during its improper disposal only gave rise to numerous environmental problems. Therefore, researchers are busy with utilizing these wastes in an attempt to make environmentally friendly building materials and reduce the existing environmental problems and attain sustainable development^13–17.

The pulp and paper industry, as one of the world’s major industries, produces tons of waste, including recycled paper, paper mill sludge, and lignosulfonates. Improper disposal of these wastes turned into significant costs and environment issues¹⁸. Europe is the second-largest paper producer in the world, accounting for 25% of global paper production, thereby generating possibly 11 million tons of paper wastes each year. In Italy alone, paper mills produced 600,000 tons of sludge in 2004^19,20. In 2020, more than 400 million tons of paper were produced worldwide by paper mills giving rise to 40–50 kg of dry sludge generated for every ton of paper production, thus indicating the large amount of wastes by the paper industry^21,22. Purification and its different effects on the environment are thousands of wastes generated by paper mill waste, like wastewater production, solid waste generation, and gas emissions¹⁸. The waste can be managed through various means like landfilling, but these traditional ways of waste management limit government resources and create environmental crises. Therefore, innovative ideas can provide better utilization of waste generated by paper mills. Using these wastes in the manufacture of construction materials has turned out to be a point of attraction among researchers in civil engineering^23–26.

Paper sludge is one of the most discarded wastes in paper mills, which contains numerous fibers with other organic and inorganic materials²⁷. Previous research results have indicated that using paper sludge reduced the swelling potential of the soil²⁸. It is generally shown that when the addition of paper sludge waste is increased in concrete, it is associated with a decrease in compressive strength as well as density²⁹. Incorporating paper sludge into clay bricks resulted in a more porous structure^30,31. microstructural investigations have revealed that utilizing waste paper sludge promotes a more compact, dense, and less porous matrix due to cation exchange and pozzolanic reactions³².

Burning paper sludge reduces the waste volume; the ash that the burning sludge produces can be an appropriate substitute for ordinary materials such as cement and lime, to form an eco-friendly construction material and civil engineering works because of its pozzolanic properties. The major chemical constituents of PSA are iron, alumina, silica, and a particular amount of calcium oxide³³. With the addition of PSA to the soil, a reaction will occur wherein cementitious compounds such as CSH and CAH are formed to alter the characteristics of soil “Equation (1–3)”. The relevant chemical reactions are as follows^34,35:

1

2

3

The addition of PSA to fine-grained soil reduced the LL and PI. Additionally, incorporating polypropylene fibers into the fine-grained soil treated with PSA increased the UCS^5,36. It was found that PSA-treated soil had a higher longitudinal wave velocity and better mechanical behavior compared to lime treated soil^37,38. The addition of PSA along with cement and lime in fine-grained soil actually improved the compressive strength by more than the addition of only cement and lime³⁹. The combination of cement with PSA enhanced the mechanical properties of well-graded gravel⁴⁰. Increasing the pulp mill fly ash content improved the compressive strength of silty sand⁴¹. Adding waste paper sludge ash to Sulphate-bearing clay and combining it with granulated blast furnace, lime and cement has improved the characteristics of the soil, such as increasing the strength of the soil⁴². Calcite formation has assuredly impacted upon the UCS of sand treated with PSA⁴³. The incorporation of geopolymer produced from paper mill sludge ash into clay can improve compressive strength by 220% over the original without stabilization using ordinary Portland cement⁴⁴. California bearing ratio of poorly graded sand soil improved following treatment with PSA based geopolymer⁴⁵. Recycled soil treated with PSA achieved higher strength and lower permeability even at lower dry density, mainly due to ettringite related hydration products formed by the stabilizer⁴⁶. Stabilizing soft soil with 10% waste paper sludge ash greatly increased shear strength, raising internal friction by 35% and cohesion by 38%⁴⁷. Environmental assessments of lime-based materials such as PSA and carbide lime demonstrated that heavy metal concentrations remained negligible and within international standard limits. Furthermore, the formation of cementitious products and the resulting improved inter particle bonding effectively prevented from leaching of heavy metals by encapsulating them within the stabilized soil matrix, confirming their potential use as environmentally friendly materials for sustainable soil stabilization^5,48. The incorporation of lime-based materials enhanced soil durability by transforming the microstructure into a flocculated fabric and reducing large cavities. This structural improvement led to an initial increase in UCS during early wet-dry cycles. However, with further cycling, the strength decreased due to crack propagation and the degradation of the cemented structure, reflecting a transition from chemical strengthening to physical deterioration^49,50. FSI dropped noticeably as lime sludge content increased, and adding 10% cement to the clay–lime sludge blends reduced it even further, making the blends almost non swelling⁵¹. Soil improvement using lime sludge and rice husk ash has been effective for structural and roadway applications⁵². Recent research has indicated that utilizing alkali-activated corn cob ash and eggshell powder as an ordinary Portland cement alternative can reduce carbon emissions by up to 30%⁵³. soil stabilization using volcanic ash geopolymer has minimized both the swelling potential and water absorption of the soil⁵⁴. incorporating pumice into cement-stabilized expansive subgrades has reduced brittle behavior by increasing failure strain, thereby enhancing soil ductility and deformation capacity under pavement loading⁵⁵. PSA was used in producing other building materials, such as concrete and clay bricks. One kind of action observed in an addition of PSA is the reduction of slump^56–58. According to another option is to employ hydrophobic PSA to decrease the permeability of concrete⁵⁹. Incorporation of PSA into self consolidating lightweight foamed concrete increased UPV⁶⁰. The addition of PSA to construction bricks decreased compressive strength⁶¹.

Researchers have indicated that developing sustainable construction materials from paper industry waste is a crucial way to minimize the environmental hazards associated with improperly disposing of them. Hence, paper industry wastes have attracted significant research interest in civil engineering, particularly regarding the use of paper sludge. The disposal of the huge amount of sludge produced in the paper mills poses multiple challenges, commonly coupled with a bad odor.

Targeting waste management and alignment with sustainable development goals, this current research has sought to reduce the quantity of paper sludge volumes and its associated pollution through incineration, and to utilize the resulting PSA as an additive in soil treatment, particularly as a substitute for traditional materials such as cement and lime. This approach highlights the practical importance of the study by providing a localized, environmentally-friendly solution for a major industrial plant in northern Iran, where the accumulation of this waste has caused severe environmental challenges. To achieve this, this research has provided a comprehensive approach by simultaneously evaluating the physical, mechanical, and microstructural properties of the PSA-treated soil. Additionally, integrating experimental results with statistical analyses has offered a better understanding of the relationships between variables and has identified the governing factors influencing soil performance, presenting a methodology rarely seen in previous literature. To this end, the effects of PSA content, curing time, and dry density on the clay soil were evaluated through standard proctor compaction, Atterberg limits, UCS, pH, and microstructural tests, followed by correlation, regression, and sensitivity analyses.

Materials and methods

Materials

Soil

The soil, used in this study, was provided from northern areas of Iran, Gorgan city, Golestan province, with coordinates to a specific site at 36°49’19.56"N and 54°30’18.59"E (Fig. 1 (a-c)). The particle size distribution curve was plotted following with ASTM D6913⁶² and ASTM D422⁶³(Fig. 2). The classification based on the unified soil classification system (USCS) proved that it was a low plasticity clay (CL). Table 1. lists the engineering characteristics of soil used for this study.

Fig. 1.

Geographic location and field view of the study area: (a) satellite map indicating the investigation region in Iran (b) close-up satellite view showing the soil sampling site in the vicinity of Gorgan city (c) field photograph of the soil sampling site. The maps presented in Fig. 1 (a) and 1 (b) were prepared by the authors using Google Earth Pro (Version 7.3.6; https://www.google.com/earth/).

Fig. 2.

Particle size distribution of the soil.

Table 1.

Engineering characteristics of the soil.

Parameter	Standard	Value
Soil classification (USCS)	ASTM D248764	CL
Specific gravity (Gs)	ASTM D85465	2.71
Maximum dry unit weight, MDD (kN/m3)	ASTM D69866	16.80
Optimum moisture content, OMC (%)		15.78
Liquid limit (LL) (%)	ASTM D431867	25.83
Plastic limit (PL) (%)		15.52
Plasticity index (PI) (%)		10.31
pH	ASTM D497268	8.54

Paper sludge ash

This is one of the active paper mills in Iran to manufacture white top cardboard from recycled white and brown paper used for different product packaging and carton making. The by-product of cellulose manufacturing in this factory is wastewater. The treatment of this wastewater yields fibrous solid waste, which forms a sludge that is deposited in ponds surrounding the factory (Fig. 3). An analysis of the collected paper sludge (Fig. 4(a)) revealed that it consists of three main components: coarse plastics (Fig. 4(b)), fine plastics (Fig. 4)c)), and fibers (Fig. 4(d)), the latter being fibrous and elongated structures. The analysis of these sludge components was carried out three-dimensionally using a binocular microscope, where materials such as fine plastics, fillers, and fibers can be observed. (Fig. 4(e)).

Fig. 3.

Paper sludge storage pond located around the factory.

Fig. 4.

Paper sludge collected from the pond and its components: (a) collected paper sludge (b) coarse plastics (c) fine plastics (d) FESEM image of paper sludge (e) binocular microscope image of paper sludge.

A substantial amount of sludge is generated by the paper mill and is incinerated in dedicated furnaces, where the organic fraction contributes to the fuel energy required for combustion. This process reduces the sludge volume and produces PSA as a stable by-product suitable for use in construction related research. Raw paper sludge is also associated with significant environmental and health concerns due to its contamination, high moisture content, presence of organic pollutants, unpleasant odor, and its tendency to attract pests at sludge disposal sites. Incineration eliminates these issues and yields a safer and more manageable material for engineering applications. Accordingly, the ash produced from burning paper sludge was utilized for soil treatment in this study. Its particle size distribution is presented in Fig. 5. In order to produce the PSA used in this study, Air dried paper sludge was burnt in a furnace (Fig. 6(a)) under a temperature of 850 °C, producing the PSA (Fig. 6(b)), which was then sieved using a sieve with a No. 40 mesh and prepared for testing. FESEM observations show that particles of PSA are agglomerated with irregular surface morphologies (Fig. 7). Table 2. presents properties of the PSA.

Fig. 5.

Particle size distribution of PSA.

Fig. 6.

Stages of producing PSA (a) placing paper sludge in the furnace (b) Resultant PSA.

Fig. 7.

FESEM image of PSA.

Table 2.

Properties of PSA.

Parameter	Reference	Value
Specific gravity (Gs)	ASTM D85465	2.44
pH	ASTM D497268	12.71
Ash content obtained from incineration of paper sludge (%)	TAPPI 41369	41.33

To establish the chemical composition and mineralogy of the studied materials, XRF and XRD tests were performed. The results of the XRF analysis for identifying the chemical compositions of the soil, paper sludge, and PSA are shown in Table 3. The results indicate that the PSA has high amounts of calcium oxide, indicating lime-based material. Furthermore, the results of the XRD analysis showed that the clay consists of quartz (SiO₂) and calcite (CaCO₃) and calcite magnesian ((Ca, Mg)CO₃) (Fig. 8(a)). The minerals found in the paper sludge were calcite (CaCO₃), calcitic magnesian ((Ca, Mg)CO₃), and portlandite (Ca(OH)₂) (Fig. 8(b)). Besides, the PSA consists of portlandite (Ca(OH)₂), lime (CaO), and calcite (CaCO₃) (Fig. 8(c)).

Table 3.

Chemical compositions of clay, paper sludge, and PSA.

Chemical compositions (%)	Soil	Paper sludge	PSA
SiO2	48.231	8.350	10.055
Al2O3	10.637	4.701	6.084
Fe2O3	5.940	0.846	1.169
CaO	11.680	45.741	49.053
MgO	3.397	1.444	1.993
Na2O	0.994	0.488	0.036
K2O	2.359	0.202	0.117
TiO2	0.517	0.193	0.204
MnO	0.103	0.015	0.018
P2O5	0.128	0.294	0.366
LOI	13.440	36.770	29.600

Fig. 8.

XRD test results (a) clay (b) paper sludge (c) PSA.

Sample preparation and test procedures

To prepare soil and PSA for testing, the two materials were first dried inside an oven at temperature of 110℃ ± 5^℃. After that, the specified amount of dried soil was weighed and PSA was added to the soil at contents of 0, 2.5, 5 and 10% by dry weight of soil, then the mixture was stirred to achieve homogeneity. Subsequently, water was added at the OMC. The selected range of PSA contents was extended beyond the average ranges reported in previous studies on soil treated with lime-based materials in order to obtain a more comprehensive assessment of the behavior of treated soil^70–72. The mixture of soil, PSA, and water was kept in cellophane to retain its moisture for curing (1, 7 and 28 days). The curing time was selected to account for both the short term and long term effects of chemical reactions, based on common practices in previous studies⁷³. Within this research, the impact of PSA on the various characteristics of fine-grained soil was assessed using standard proctor compaction, Atterberg limits, UCS, pH, and microstructural analyses like XRD and FESEM. Also The effect of dry density (14, 15, and 16 kN/m³) was evaluated in the UCS test. Based on the MDD of 16.8 kN/m³ determined for the untreated soil, three dry density levels of 14, 15, and 16 kN/m³ were selected. The value of 16 kN/m³ corresponds to 95% of the MDD of the untreated soil, while 15 and 14 kN/m³ correspond to approximately 89.29 and 83.33% of the MDD of the untreated soil, respectively. This range was selected to evaluate the mechanical behavior of the soil under different compaction degrees. Another reason for choosing this range was the laboratory limitations and the feasibility of preparing specimens in the standard 38 × 76 mm mold. Similar studies have also reported the effect of dry density on UCS⁷⁴. Further evaluations were carried out with respect to the secant modulus, brittleness index, and energy absorption capacity of the treated soil, whereas dry density, PSA content, and curing time, as variables, were carefully monitored. The details of the laboratory program are summarized below (Table 4).

Table 4.

Summary of the experimental program.

Test	Variables
	PSA content (%)	Curing time (days)	Dry density (kN/m3)
Standard proctor compaction	0, 2.5, 5, 10	1, 7, 28	-
Atterberg limits	0, 2.5, 5, 10	1, 7, 28	-
UCS	0, 2.5, 5, 10	1, 7, 28	14, 15, 16
pH	0, 2.5, 5, 10	1, 7, 28	-
XRD	2.5, 5, 10	28	-
FESEM	0, 10	28	-

Standard proctor compaction

Standard proctor compaction was used to determine values of MDD and OMC of untreated and treated soil. Testing was conducted per the ASTM D698⁶⁶. Curing time was regarded as an important factor in completing chemical reactions and in affecting the compaction parameters of treated soil. Thus, this parameter was considered.

Atterberg limits

The Atterberg limits test was conducted on untreated and treated soil using ASTM D4318⁶⁷ to evaluated the plasticity characteristics and parameters such as LL, PL, and PI. In soils stabilized with lime‑based materials, the formation of chemical reactions between the soil and the additive alters the soil’s plastic behavior. Since curing time influences the progression of these chemical reactions, the effect of this parameter on the Atterberg limits of the treated soil has been investigated. In accordance with the ASTM D4609⁷⁵, considering the effect of curing time on the Atterberg limits is justified.

Unconfined compressive strength

The UCS test was conducted, as described in ASTM D2166⁷⁶, to ascertain the mechanical behavior of the soil and the effect of PSA on it. After mixing the soil and PSA and adding water at the OMC (Figs. 9(a-c)), the mixture prepared for the UCS test were molded in a 38 mm diameter and 76 mm height mold (Fig. 9(d)) with a dry density of 14, 15 and 16 kN/m³ in 3 layers. After curing, the prepared samples (Fig. 9(e)) were tested at a loading rate of 1.27 mm/min (Fig. 9(f)). for each dry density level and mold volume, the dry weight of the soil was calculated, and the PSA was added as a percentage of this weight. By keeping the dry weight of soil constant for each dry density level, any mechanical behavioral changes can be attributed to the effect of the additive, making the comparison of different PSA contents more reliable. The sample preparation method used has also been reported in previous studies⁷⁴. From the results of this test, the secant modulus, brittleness index, and energy absorption capacity were calculated.

Fig. 9.

Stages of preparing the UCS sample (a) dried soil (b) dried PSA (c) mixture of water, soil, and PSA (d) equipment for preparing UCS sample (e) prepared UCS test samples (f) sample undergoing UCS testing.

pH

The pH testing in soil treatment is intended to assess the chemical conditions of the soil environment, which would enable pozzolanic and cementitious reactions between soil particles and PSA. Therefore, the pH test was used to assess the effect of adding PSA on the acidity or alkalinity of the soil in accordance with ASTM D4972⁶⁸. Using a pH meter, the pH was measured with an accuracy of ± 0.05 units. For each test, 10 g of the mixture were mixed with distilled water at a weight ratio of 1:2.5 for soil to water⁷⁷.

Microstructural studies

The purpose of microstructural testing, including X-ray diffraction (XRD) and field emission scanning electron microscopy (FESEM), is to establish internal changes, chemical processes, and mineral composition that have taken place in soil beyond what can be seen with the naked eye. The usage of the XRD was for analysis of modifications brought about by the addition of PSA to the clay. Angle ranges for setting the XRD test were limited from 5.291 °2Th to 79.991 °2Th, with step sizes of 0.05 °2Th in between. By examining the imaging results from this high-magnification and high-resolution imaging technique, the FESEM test was used to analyze the continuity between particles, structural changes within soil treated with PSA and the formation of new compounds.

Results and discussion

Standard proctor compaction

According to the results, the MDD and OMC of the untreated soil were 16.80 kN/m³ and 15.78%, respectively. The MDD of the soil treated with 2.5, 5 and 10% PSA after 1 day curing was 16.21, 15.95, and 15.63 kN/m³, respectively, while the OMC was 16.69, 20.50 and 21.30%, respectively (Fig. 10). Comparatively, the MDD obtained for the 10% PSA-treated specimen was 24.54% higher than that reported in a previous study, while its OMC exhibited a 8.67% increase⁷². The decrease in MDD with increasing PSA content was attributed to the fact that PSA has a lower specific gravity than clay. Also, PSA readily absorbed excess water, and thus the OMC increased^5,78. It can be seen from the plotted diagram that increased curing time resulted in an increase in OMC and a decrease in MDD (Figs. 11(a-b)). For instance, the OMC of 10% PSA-treated soil increased by 34.98, 67.30, and 74.27% with curing at 1, 7 and 28 days, respectively. In contrast, the MDD of 10% PSA-treated soil after curing for 1, 7 and 28 days decreased by 6.96, 13.10, and 14.29%, respectively, in comparison with untreated soil. Therefore, an OMC increase with lessening of MDD was observed. With increasing curing time and the progressive development of chemical reactions, the cementitious compounds formed within the soil–PSA mixture gradually modified the internal structure of the soil. These newly formed products penetrated into the spaces between soil particles, bonded them together, and consequently altered the particle arrangement, and induced significant microstructural changes. Since compaction behavior is strongly governed by these factors⁷⁹, the curing time caused the mixture to respond differently to compaction compared to the fresh mixture. This change has demonstrated how the incorporation of curing time into the standard compaction test influenced the soil behavior, directly explaining the observed reduction in MDD and the subsequent increase in OMC. Moreover, part of the initially added water was consumed during curing for hydration and the formation of cementitious compounds, leaving less water available for achieving maximum compaction; hence a higher moisture content was required to reach the optimum condition. Altogether, as the PSA content increased, the OMC increased while the MDD decreased. In addition, an increase in curing time resulted in a higher OMC, while the MDD decreased. Similar findings have already been reported in literature^70,71.

Fig. 10.

Effect of adding PSA on the compaction of fine-grained soil.

Fig. 11.

Effect of curing time on compaction parameters of treated soil (a) OMC (b) MDD.

Atterberg limits

the results obtained from the Atterberg limits test, it appears that the soil treated with 0, 2.5, 5 and 10% PSA under curing time of 1 day had PL values of 15.52, 16.1, 17 and 18.35%, respectively. The LL of soil treated with 0, 2.5, 5 and 10% PSA under curing time of 1 day was found to be 25.83, 25.1, 24.4 and 23.62%, respectively; while the PI of the soil treated with 0, 2.5, 5, and 10% PSA under curing time of 1 day was measured to be 10.31, 9, 7.4 and 5.27%, respectively. Overall, as indicated with increasing PSA content the LL and PI decreased while PL increased. The observed decrease in plasticity was attributed to the release of calcium ions (Ca²⁺) from the PSA, which replaced monovalent cations on the clay particle surface, thereby leading to flocculation and aggregation^9,72. Beyond the immediate effects of PSA on soil plasticity properties, the influence of curing time on variations in the Atterberg limits provides a clearer understanding of the stabilization process and the effectiveness of such additives. This evaluation serves as a practical criterion for controlling the quality of the treated soil and for estimating key parameters such as strength⁷. The PL and LL of the 10% PSA-treated soil, increased by 22.42% and decreased by 15.21%, respectively, after curing for 28 days, as compared to the untreated soil (Figs. 12(a-b)). The PI of the soil treated with 10% PSA after curing for 1, 7, and 28 days decreased by 48.88, 65.08 and 71.87%, respectively, compared to untreated soil (Fig. 12(c)), and the PI value of the 28-day cured 10% PSA-treated soil exhibited a 27.5% greater reduction compared to a similar study⁵. During the first 7 days, the PI of the treated soil changed significantly; however, it remained almost stationary thereafter. The rapid cation exchange implies that this process could not appreciably affect the PI of the soil in the long term⁵. So in general, with increasing PSA content, LL and PI decreased, while PL increased; also with longer curing time, the values of LL and PI showed a downward trend while the PL showed an upward trend. Similar observations were made in earlier studies^72,73.

Fig. 12.

Effect of PSA content and curing time on Atterberg limits:.(a) Plastic Limit, PL (b) Liquid Limit, LL (c) Plasticity Index, PI.

Unconfined compressive strength

The use of lime-based materials such as PSA for soil treatment has caused pozzolanic reactions to occur and ultimately improving the mechanical behavior. UCS test directly showed three factors such as higher PSA content, increasing dry density, and extending curing time affecting UCS. Regarding the UCS results, the strength of the treated soil with a dry density of 16 kN/m³ and a curing time of 1 day significantly improved with the addition of PSA. Specifically, as the PSA content increased, the UCS values grew by 2.32 to 7.25 times compared to the untreated soil. The UCS of the treated soil was raised with the duration of curing time such that the 5% PSA-treated soil and a dry density of 16 kN/m³ under curing for 1, 7 and 28 days increased by 3.64, 5.22, and 7.95 times, respectively, compared to the untreated soil. The UCS of soil cured for 28 days at a dry density of 16 kN/m³ and treated with 2.5 (D16_PSA2.5%_C28), 5 (D16_PSA5%_C28) and 10% (D16_PSA10%_C28) PSA improved by 3.82, 7.95 and 19.94 times, respectively, compared to untreated soil (Fig. 13). The maximum obtained strength exhibited a 1.75-fold increase compared to a similar study⁷³. The significant increase in the strength of the PSA-treated soil, which consequently resulted in up to a 19.94-fold increase in UCS, was due to the simultaneous effect of the applied physical and chemical mechanisms. Regarding the chemical mechanisms, PSA gave rise to the formation of initial cementitious compounds such as CSH and CAH in the clay, while prolonged curing time nurtured the chemical reactions and the emergence of secondary cementitious compounds like calcite. These cementitious products infiltrated the voids, creating stronger bonds between particles and providing a more integrated matrix, thereby increasing the UCS of the treated soil⁵. On the other hand, in the fabricated specimens, as the dry weight of the soil was kept constant for each target dry density and the PSA was added to the dry soil, physical mechanisms also contributed to the strength improvement. Due to the smaller particle size of the PSA compared to the soil, these particles penetrated the soil pores, acting as a filler that reduced porosity and created a denser soil fabric. This structural enhancement increased the contact points between particles, leading to more efficient stress transfer throughout the soil matrix. As a result of the improved mechanical behavior achieved through both chemical and physical mechanisms, the formed bonds contribute to a denser and more stable structure, enhancing the material’s ability to withstand stress without sudden failure. According to previous studies, the chemical and physical bonds in lime stabilized soil remained stable, and the pozzolanic reactions continued. As a result, the strength of the treated soil continued to increase during long-term curing, indicating long‑term strength stability⁸⁰. The soil cured for 28 days and treated with 5% PSA, the UCS value for the sample having a dry density of 16 kN/m³ was 1.31 times higher than that for a dry density of 15 kN/m³ and 1.85 times greater than that for 14 kN/m³ (Fig. 14). This improvement can be attributed to a decrease in porosity due to the increase of dry density, hence an increase in UCS. The large deep crack that existed in the untreated soil reflects the instability after failure, whereas the narrow crack that occurred in the treated soil indicates better stability (Figs. 15(a, b)). Bigger improvement in post-failure stability of the soil results in reduction of engineering risks, along with construction structure reliability. Such observations have been previously reported by many^{5,37,72,73,81,82}.

Fig. 13.

Stress-strain curve showing the effect of PSA content on the UCS of soil with a dry density of 16 kN/m³ at 28 days curing.

Fig. 14.

Evaluation of the effect of dry density on the UCS of soil treated with various PSA contents under different curing times.

Fig. 15.

Samples after failure (a) Untreated soil (b) treated soil.

Secant modulus (E₅₀)

The secant modulus defines the ratio of half maximum stress to strain corresponding to that stress, which indicates the stiffness of the material⁷⁴. An impact assessment of the PSA content on the secant modulus of fine-grained soil for different dry density and curing time has been carried out based on the results obtained (Table 5). The secant modulus of untreated soil with 14, 15 and 16 kN/m³ of dry density was observed to be 4.60, 8.08 and 10.07 MPa, respectively. At 28 days of curing and a dry density of 16 kN/m³, as shown in the table, the secant modulus of soil treated with 2.5%, 5%, and 10% PSA content increased 2.96, 5.34, and 11.79 times, respectively, compared with the untreated soil at the same dry density of 16 kN/m³. Overall, higher secant modulus to treated soil and consequently greater stiffness developed as increased PSA content, curing time, and dry density, respectively, all of which were attributed to a better structure derived from the treatment. Similar findings have been observed in prior studies performed on materials based on lime^82,83.

Table 5.

Secant modulus results (MPa).

		(γd = 14 (kN/m3			(γd = 15 (kN/m3			(γd = 16 (kN/m3
Curing Time (days)		1	7	28	1	7	28	1	7	28
PSA (%)	0	4.60	4.60	4.60	8.08	8.08	8.08	10.07	10.07	10.07
	2.5	14.09	15.76	19.97	15.83	18.29	26.57	23.30	26.99	29.84
	5	26.99	28.68	36.61	38.61	41.65	47.62	44.84	46.37	53.77
	10	30.44	44.82	63.38	42.70	57.62	76.04	47.81	61.78	118.77

Brittleness index

One of the criteria for assessing the ductility or brittleness of the soil and rock at failure is the brittleness index. The brittleness index of the soil is calculated using “Equation (4)”:

4

Where and represent the failure stress and ultimate stress, respectively³⁶. A higher brittleness index corresponds to more brittle soil behavior, whereas a lower index indicates more ductile performance. The brittleness index of untreated soil with dry density of 14, 15 and 16 kN/m³ was 0.57, 0.45, and 0.28, respectively. The brittleness index of the soil treated with 2.5, 5 and 10% PSA at a curing time of 28 days and a dry density of 16 kN/m³ decreased by 39.29, 60.71, and 71.43%, respectively, as compared to untreated soil at dry density 16 kN/m³. For either the higher PSA content or more dry density level, porosity was reduced and, therefore, denser packing of particles with improved stress transfer occurred by which treated soil attained its maximum strength at higher strains resulting in more flexible soil behavior as a result. Over time the chemical reactions matured and were rendered more complete stabilizing the soil structure, by which the ability of the treated soil to withstand stress without sudden failure improved, accounting for its more flexible behavior (Table 6).

Table 6.

Brittleness index of the studied samples.

		(γd = 14 (kN/m3			(γd = 15 (kN/m3			(γd = 16 (kN/m3
Curing Time (days)		1	7	28	1	7	28	1	7	28
PSA (%)	0	0.57	0.57	0.57	0.45	0.45	0.45	0.28	0.28	0.28
	2.5	0.49	0.45	0.43	0.36	0.34	0.30	0.22	0.20	0.17
	5	0.42	0.39	0.33	0.32	0.29	0.26	0.15	0.12	0.11
	10	0.37	0.34	0.31	0.28	0.27	0.24	0.13	0.11	0.08

Energy absorption capacity

The amount of energy absorbed by the soil before the failure under loading is defined by this parameter. A high energy absorption capacity indicates that soil has more ductility to be able to sustain deformation without a catastrophic type of failure. This value is equal to the area behind the stress-strain curve and represents the work done on the soil. The results obtained suggest that the energy absorption capacity for untreated soil at dry densities of 14, 15 and 16 kN/m³ is 61.599, 158.838, and 307.509 kJ/m³, respectively. Overall, increasing PSA content, curing time, and dry density made the treated soil demonstrated more ductile behavior and increased energy absorption capacity (Fig. 16). The energy absorption capacity for soil treated with 2.5, 5 and 10% PSA under a curing time of 1 day and a dry density of 16 kN/m³ were found to increase by 2.63, 4.15 and 7.59 times, respectively, as compared to untreated soil with the same dry density of 16 kN/m³.

Fig. 16.

Influence of dry density, curing time, and PSA content on the soil’s energy absorption capacity.

The brittleness index and energy absorption capacity are two parameters that, beyond merely describing mechanical behavior, can directly guide the selection of the type and amount of stabilizer and help predict the structural performance of stabilized soils in practical geotechnical design contexts. Overall, the increase in energy absorption capacity and the decrease in the brittleness index of the soil treated with PSA indicate that the addition of this material has improved the flexibility of the soil. These findings also show that the stabilized soil exhibited a more ductile behavior, absorbed more energy before failure under loading, allowed for a more uniform stress distribution, and consequently did not undergo sudden failure. From a stabilizer selection standpoint, materials that simultaneously produce a lower brittleness index and a higher energy absorption capacity are preferable for applications where ductile behavior and controlled deformation are critical, such as embankments and backfills supporting retaining structures.

pH test

Data and observations post adding PSA on clay determined that resulting pH values for the soil environment would be alkaline. The soil treated with 2.5, 5 and 10% PSA at the end of 1 day curing showed increased pH values over untreated soil by 45.43, 48.36 and 50.12%, respectively (Fig. 17). Increased soil pH is sometimes viewed unfavorably from an environmental aspect as it negatively influences the growth and development of plants and will result in problems^23,84. However, the addition of lime-based materials by increasing pH provide suitable environment for the formation of cementitious compounds. Considering the presence of portlandite (Ca(OH)₂) in the XRD test results of PSA (Fig. 8(c)), this mineral reacted with water to form hydroxyl ions and calcium ions “Equation (1)”, which increased the pH of the soil and alkalized the soil’s chemical environment³⁵. The released calcium ions then reacted with the silica and alumina dissolved from the clay minerals to form CSH “Equation (2)” and CAH “Equation (3)” compounds. As illustrated in Fig. 17, further curing time led to a tendency for the pH of the treated soil to decrease. thus, the pH of the 10% PSA-treated soil was found to be down 2.50% and 3.67% after 7 and 28 days of curing, respectively, as compared with 10% PSA-treated soil with only 1 day curing. As the curing time increased, the concentration of OH⁻ ions gradually decreased, leading to a reduction in pH. This reduction, in turn, triggered the carbonation of CSH and its subsequent transformation into calcite as a secondary cementitious material, which further contributed to the stabilization of the soil matrix and resulted in increased soil strength⁸⁵. The same results have been reported in other studies^85–87.

Fig. 17.

Effects of PSA content and curing time on the pH.

Microstructural studies

XRD analysis

The quantitative XRD analysis demonstrated a progressive increase in the intensity of calcite peaks as the PSA content was raised (Fig. 18). The primary calcite peak intensity for the untreated soil was recorded at 2616 counts. Following the addition of PSA and a 28 days curing time, these values increased to 3610, 3694, and 3992 for the soil treated with 2.5, 5, and 10% PSA, respectively. This substantial growth in peak intensity (up to a 52% increase compared to the untreated soil) provides clear evidence of enhanced calcite formation, which plays a key role in the significant strength improvement observed in the treated soil. The increase in calcite peaks with increasing amount of PSA is an indication of increasing UCS of the treated soil. Calcite is one of those cementing compounds that greatly influence strength of treated soil. The usage of PSA has caused the soil environment become alkaline, increasing pH and forming initial cementitious compounds such as CSH and CAH. As curing progressed, the pH decreased leading to the carbonation of CSH and its disintegration to calcite as a secondary cementitious material^82,88.

Fig. 18.

Comparative XRD results of untreated and PSA-treated soil.

FESEM analysis

The FESEM analysis showed that untreated soil (Fig. 19(a)) has voids that indicate porosity in the texture. PSA incorporated in the soil particles formed cementitious compounds resulting in a denser soil texture (Fig. 19(b)). With the dense structure of soil particles, the compressive strength of the treated soil was increased due to the formed cementitious compounds covering the soil particles and increasing interparticle forces^89,90.

Fig. 19.

FESEM images (a) untreated soil (b) soil treated with PSA.

Statistical studies

Statistical studies in the field of civil engineering serve to analyze laboratory data, to model complex phenomena, and to carry out optimization in various design scenarios. Statistical methods are adopted to predict the behavior of structures and building materials under different loading conditions, to analyze slope stability, and to effectively evaluate the influence of various factors on soil and structural performances. Statistical studies have been conducted to analyze the results. In this regard, Python was used to study correlation between the dependent variables (pH, OMC, MDD, PI, and UCS) and the independent variables (PSA content, dry density, and curing time) and some preliminary relationships were established. Subsequently, regression analyses were carried out using the UCS test results to quantitatively model the effect of independent variables on the UCS and to obtain a highly accurate equation predicting the UCS of PSA-treated soil. Finally, a sensitivity analysis was used to evaluate the relative importance of each independent variable with respect to the model output.

Correlation analysis

In the present study, a correlation analysis was conducted to assess the intensity and direction of the statistical association between the variables.The magnitude of the correlation coefficient quantifies the intensity of the relationship, while its sign (positive or negative) determines the direction of the mutual association between the variables. The results showed that PSA content was the most influential independent variable (Fig. 20). In this regard, independent variables such as PSA content, curing time, and dry density correlate at values of 73, 51, and 31%, respectively, with UCS. Thus, the positive coefficients indicated a direct association between UCS and the parameters of PSA content, curing time, and dry density. Among these, the one with a strong positive correlation was PSA, which demonstrated a significantly greater impact on UCS than curing time and dry density.

Fig. 20.

Correlation between parameters.

Regression analysis

Regression analysis generally describes the class of models and statistical methods used to model and analyze data. This method gives a quantitative description of the relationship between independent variables and the dependent variable⁹¹. Since assessing the mechanical behavior of soil treated with PSA is quite relevant, a linear-logarithmic regression analysis was utilized in analyzing the relationship between the three independent variables and one dependent variable. This model was specifically selected because preliminary analyses indicated a non-linear relationship between the independent and dependent variables, thereby enhancing the overall model quality and prediction results. Considering the exponential behavior of the dependent variable with respect to the independent variables, applying a logarithmic transformation simplified the model into a linear form and made the interpretation of the coefficients clearer. The analysis will furnish the model and accurate prediction of the UCS of soil treated with PSA. By applying regression analysis, Eq. (5) was derived. In “Equation (5)”  P,  C and  D represent the PSA content, curing time, and dry density, respectively. The regression coefficients represent the theoretical effect of a one unit change in each independent variable on the dependent variable, assuming all other variables remain constant. For example, a one unit increase in the PSA content resulted in approximately a 21% increase in the UCS value (e0.1883 ≈ 1.21). It should be noted that, due to differences in scale and range among the independent variables, directly comparing the coefficients to identify the most influential independent variable on the dependent variable is not appropriate; therefore, a sensitivity analysis is necessary to resolve this issue.

5

According to the developed regression model, the accuracy of the model was found to be , indicating a high level of accuracy for the developed model (Fig. 21). To ensure the reliability of the regression model, the dataset (108 samples) was randomly divided into training (70%) and testing (30%) subsets, resulting in R² values of 95.5 and 93.4% for the training and testing sets, respectively, indicating strong predictive performance without overfitting. The five fold cross validation yielded an average R² of 88.9%, supporting the stability of the model. Error metrics on the test data (MAE = 0.141 and RMSE = 0.172) demonstrated the high accuracy of the predictions. Additionally, residual analysis using the Jarque–Bera (P = 0.257>0.05) and Omnibus (P = 0.136>0.05) tests verified the normality of the errors. It is worth noting that the proposed regression equation is limited to the experimental ranges of this study: PSA content (0, 2.5, 5, and 10% by dry weight of soil), dry density (14, 15, and16 kN/m³), and curing time (1, 7, and 28 days). Utilizing the model beyond these ranges may reduce its accuracy.

Fig. 21.

Predicted of experimental values and estimated values.

Sensitivity analysis

The sensitivity analysis assessed how each variation in the independent variables influenced the dependent variable⁹². In this method, the maximum, minimum, and average values were first established for each independent variable, after which the value for that variable was calculated “Equation (6)”.

6

Where and are the maximum and minimum output values of the regression model, respectively “Equation (5)” and the maximum and minimum values of one of the independent variables are substituted into the regression model, while other independent variables are set to their average values. Then, the effective degree of each independent variable on the dependent variable were calculated “Equation (7)”⁹¹.

7

Where n is the number of input variables and is the effective degree. According to the results obtained, the effective degree of the independent variables, including PSA content, curing time, and dry density, were 54.71, 24.78 and 20.51%, respectively. The PSA content had a greater effect on the UCS compared to other independent variables (Table 7).

Table 7.

Effective degree of the independent variables on the dependent variable.

Dependent variable	Independent variables
	PSA content (%)	Curing time (%)	Dry density (%)
UCS	54.71	24.78	20.51

Conclusions

This research successfully evaluated the feasibility of utilizing PSA as a sustainable and eco-friendly stabilizer for fine-grained soil in construction projects. By examining the effects of three major variables, namely, PSA content, dry density, and curing time, this approach not only mitigates the environmental burdens of paper sludge disposal but also significantly enhances the engineering properties of the treated soil. Based on the comprehensive experimental and statistical analyses, the key findings are summarized as follows:

• With the increase in the PSA content, the MDD decreased by 6.96% because PSA has a lower specific gravity compared to clay. Additionally, the OMC of the treated soil increased by 34.98% due to the higher water absorption capacity of PSA. As the curing time increased, the MDD exhibited a 14.29% reduction, whereas the OMC demonstrated a 74.27% increase.

• A reduction of approximately 48.9% in the PI was obtained with increasing PSA content, while an overall decrease of about 71.9% was achieved as the curing time was prolonged.

• For PSA-treated soil with a dry density of 16 kN/m³, the formation of cementitious compounds driven by increasing PSA content and prolonged curing time resulted in up to 19.94-fold improvement in UCS. However, the UCS eventually declined due to the lower dry density resulting from increased porosity.

• Higher PSA content, longer curing time, and increased dry density consistently enhanced the secant modulus and energy absorption capacity of the treated soil, while simultaneously reducing its brittleness index. Consequently, the treated specimens achieved peak strength at higher strains, demonstrating a distinct transition toward a more ductile and flexible failure mechanism.

• The PSA-treated soil exhibited a substantially higher pH than the untreated soil; the 10% PSA-treated soil showed a 50.12% increase, thereby providing the necessary chemical conditions for the formation of cementitious compounds. However, over the curing period, the pH of the soil treated with 10% PSA decreased by about 3.67% from 1 day to 28 days.

• XRD and FESEM analyses confirmed that adding PSA to clay led to the formation of cementitious compounds and the crystallization of calcite. Specifically, the calcite peak intensity in the treated soil increased by 52% compared to the untreated soil, and microstructural imagery revealed a dense and interconnected structure within the treated soil matrix.

• During the statistical investigations, it was determined that PSA presented the strongest correlation with dependent variables. The regression analysis carried out on the UCS test results generated an equation which produced a 95.8% accuracy in predicting the UCS of soil treated with PSA. Based on the sensitivity analysis, PSA content had the highest influence on UCS values.

In terms of practical implications, the utilization of PSA-treated soil provides a sustainable and eco-friendly material for enhancing the soil quality under the foundation of building structures, serving as a alternative to traditional stabilizers such as cement and lime. For future research directions, it is recommended to evaluate the durability of the treated soil under freeze-thaw cycles cycles and to investigate its dynamic behaviors.

Author contributions

Nima Mohammadizadeh: Investigation, Methodology, Resources, Writing - original draft. Alireza Tabarsa: Conceptualization, Validation, Supervision, Project administration. Mohammadhadi Aryaie Monfared: Writing - review & editing.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

All data, used in this investigation, will be shared upon request by the editor.

Declarations

Competing interests

The authors declare no competing interests.