Effect of Dispersed Polypropylene Fibers on the Strength and Stiffness of Cement-Stabilized Clayey Sand

Abstract

Soil stabilization with hydraulic binders like cement is widely used in road construction but significantly contributes to CO₂ emissions. This study investigates a more sustainable alternative involving the use of dispersed polypropylene fiber reinforcement to improve the mechanical properties of stabilized soils while reducing cement consumption. Nine clay sand mixtures with varying cement (2–6%) and fiber (0–0.5%) contents were tested using unconfined compressive strength (UCS) and ultrasonic pulse velocity (UPV) methods. Fiber addition improved UCS by 5.59% in a mix with 2% cement and 0.25% fibers and by 25.45% in one with 4% cement and 0.25% fibers. This shows that fibers can improve strength at different cement levels. A novel reinforcement index ($R_i$) was introduced to predict UCS empirically. The model showed that using 0.5% fibers ($R_i=1.0\%$) enabled a 25.12% reduction in cement without compromising strength. However, this improvement came at the cost of stiffness: deformation modulus $E_{50}$ decreased by up to 67.51% at 0.5% fiber content. Statistical validation using MAE, RMSE, and MAPE confirmed the model’s accuracy. Although the results were based on a single soil type, they showed that polypropylene fibers can support decarbonization efforts by reducing cement demand and represent a technically feasible approach to more sustainable geotechnical engineering applications.

1. Introduction

Soil stabilization is a method to improve the physical and mechanical properties of soil, which is achieved by blending and mixing it with other materials [1]. The method has been in use since 1915, when cement stabilization was used to build a road in the city of Saracosta, Florida [2]. In subsequent years, the method grew in popularity, resulting in the construction of more than 150 miles of cement-stabilized soil roads in the United States alone by 1938 [3]. Over time, binders have been developed for soil stabilization, so today, we can identify three main binders: traditional stabilizers, by-product stabilizers, and non-traditional stabilizers [4]. However, cement remains the most commonly used material for enhancing the mechanical properties of soil [5,6,7].

Continued growth in global demand for binders in various industries has led to a 30-fold increase in cement production since 1950 and an almost 4-fold increase since 1990 [8]. This growth has resulted in an increase in the contribution of cement producers to global carbon dioxide (CO₂) emissions, which now account for the world’s third-largest source of anthropogenic CO₂ emissions [9]. Therefore, engineering solutions that support sustainable development principles are gaining importance in the context of increasing environmental awareness and the need to reduce greenhouse gas emissions. Thus, although effective, traditional soil stabilization using cement is associated with a high carbon footprint, which is becoming increasingly pressing in light of global climate challenges. Consequently, research is being conducted on ways to reduce the amount of cement while improving the mechanical properties of the soil. In particular, the possibilities of using construction and demolition waste (CDW) [10,11,12], slag and fly ash [13,14,15], geopolymers [16,17,18], dispersed reinforcement [19,20,21], or a combination of the methods above, i.e., hybrid methods [22,23,24], are being studied.

The addition of dispersed reinforcement in the form of polypropylene fibers to cement-stabilized soils represents a promising method for reducing the demand for cement while maintaining or even improving the mechanical performance of the mixture [19]. This type of reinforcement is added to the soil and mixed with it, similar to cement, lime, or other additives. The process of preparing the mixture is similar to admixture stabilization [25]. One main advantage of randomly distributed reinforcement is ensuring the isotropy of the reinforced soil and the absence of potential planes of weakness [26,27,28]. In addition, the use of such reinforcement, in many respects, resembles traditional soil reinforcement [29]. Polypropylene fibers make it possible to achieve comparable strength parameters with lower cement content, which directly translates into a reduction in CO₂ emissions. Furthermore, due to their resistance to degradation and wide availability, polypropylene fibers offer a durable and efficient way to reinforce soils, aligning with the concepts of infrastructure longevity and responsible resource management.

Soil stabilization with dispersed reinforcement makes the soil medium behave like a composite material, where fibers are embedded in the soil matrix. The reinforcement mechanism of such a composite is the mobilization of the fibers’ tensile strength, which occurs during the onset of shear stresses [30,31,32]. So far, in laboratory studies of stabilized soil, researchers have mainly used polypropylene fibers as a reinforcing element [33,34,35]. Numerous studies in the scientific literature have demonstrated the effect of dispersed reinforcement on the stabilized soil’s strength parameters, including unconfined compression strength (UCS) [33,35,36,37,38], indirect tensile strength (ITS) [33,37,38,39,40], shear strength [35,41,42,43,44], or flexural strength (FS) [45,46,47]. Among the listed parameters, UCS is the primary parameter that measures strength and is often used to assess the material’s properties. Studies have shown that the increase in strength is related to the amount of reinforcement used in the mix. Even when a small amount of fiber was used, an increase in compressive strength was observed, which, as the fiber content increased, stopped or led to a decrease in strength. Some researchers have observed the most significant increase in strength with the use of fibers in the range of 0.25% to 0.50% of the dry weight of the soil [35,36,37,39,48]. However, some papers indicate an increase in strength of up to 2.50% [36,49,50]. The strength of stabilized soil is affected not only by the amount of reinforcement used but also by the length of the fibers. Some researchers indicate that the enhancement effect intensifies as the fibers’ length increases [20]. In contrast, others note that the most significant improvement in strength is obtained when 12 mm fibers are used [38,47,51]. However, it should also be noted that the optimal fiber content is not a universal value and may vary depending on the type of soil, the type and amount of binder used, and the type and proportion of dispersed reinforcement.

Dispersed reinforcement, in addition to affecting the strength of stabilized soil, also influences its deformability. The most noticeable change that occurs is the improvement in the plasticity of the material as a result of the addition of fibers [21]. The linear deformation modulus is measured to assess changes in the material’s deformability during testing, as it describes the material’s elasticity under compression or tension. Available studies are ambiguous in determining the effect of dispersed reinforcement. Some studies indicate the possibility of reducing the stiffness of stabilized soil [21,39,43], while others indicate the possibility of increasing stiffness [39,52]. In both cases, the authors obtained improvements in strength parameters. In addition, it is worth noting that distributed reinforcement can contribute to increasing the total and absorbed energy of elastic deformation in the stabilized soil while reducing dissipated energy and reducing energy losses. This effect is due to the fiber’s crack-bridging effect, which inhibits crack growth and reduces local stress concentrations [53]. However, selected studies indicate that glass fibers can exhibit higher dissipation energy at failure than samples enhanced with polypropylene fibers [47].

An important aspect of scientific research is determining the effect of dispersed reinforcement on the mechanical properties of stabilized soil and the ability to predict these properties. One method, based on the relationship between porosity ($n$) and the cement index ($C_i$), defined as the ratio of cement volume to sample volume, was proposed by Consoli et al. [7,54]. The presented approach proves that there is a relationship in the form of a power function between UCS and the ${n/C}_i$ ratio. This relationship will vary depending on the type of soil, binder used, curing time, and type of reinforcement. Numerous papers in the literature show this relationship [55,56,57,58,59,60,61,62].

This paper analyzes the effect of polypropylene fibers on the mechanical parameters of cement-stabilized soil, taking into account both non-destructive testing (NDT) and destructive testing (DT). A literature review indicates that polypropylene fibers can significantly improve the strength of the material. Still, the available empirical relationships, based mainly on the porosity and cement index, do not consider the presence of dispersed reinforcement. Consequently, separate empirical relationships must be developed for mixtures with added fibers to predict UCS.

The objectives of the study are, therefore, to (1) develop an empirical relationship to predict the compressive strength of cement-stabilized clayey sand, both without and with the addition of polypropylene fibers, (2) investigate the effect of these fibers on the material’s deformation parameters, and (3) evaluate the potential for reducing the amount of cement in the mix while maintaining the required strength through the use of fibers.

Non-destructive methods are used to measure ultrasonic pulse velocity (UPV) and shear wave velocity ($V_S$), which allows the determination of elastic parameters at very low strains, such as Young’s modulus ($E_{UPV}$), shear modulus ($G_{UPV}$), and Poisson’s ratio ($v_{UPV}$). Unconfined compressive strength (UCS) and the secant modulus at half the failure deviator stress ($E_{50}$) are determined in DT. Based on the results, correlations between UCS and strain parameters are developed.

The proposed approach’s novelty is the inclusion of polypropylene fibers in the general empirical relationship for predicting soil strength. For this purpose, a so-called reinforcement index ($R_i$), based on the ratio of the volume of fibers to the volume of stabilized soil, is introduced, which makes it possible to include dispersed reinforcement in the empirical relationship.

2. Materials and Methods

The research was conducted on nine mixtures consisting of soil, binder, and dispersed reinforcement. All mixes were developed based on soil classified as clayey sand (clSa), the detailed characteristics of which are presented in Section 2.1. As a binder, CEM V composite cement (Górażdże Cement S.A., Chorula, Poland) was used, the basic properties of which are discussed in Section 2.2. The dispersed reinforcement was polypropylene fibers, the characteristics of which are presented in Section 2.3. The prepared mixtures of stabilized soil were classified in accordance with a European standard. The samples were compacted at optimum moisture content and had a slenderness coefficient of 1. A detailed description of the procedure for the preparation of mixtures and samples and their treatment is provided in Section 2.4. After the treatment, non-destructive testing was conducted to determine the UPV and $V_S$. A full description of the testing procedure and the apparatus used is included in Section 2.5. Subsequently, strength testing of the stabilized soil was carried out, including an unconfined compression test. A detailed description of the testing procedure and the apparatus used is presented in Section 2.6. Section 2.7 shows the apparatus used to take the SEM images. Section 2.8 presents the methodology for determining the porosity of samples and the cement and reinforcement index, which are then used in the analysis.

2.1. Soil

The soil used in the study was passed through a 2 mm mesh sieve before the main tests, and the resulting compacted lumps were mechanically fragmented. The purpose of this procedure was to ensure homogeneous distribution of the dispersed reinforcement in the mix. Once the material was prepared, its basic physical properties were determined. As part of the study of physical properties, aerometric analysis and sieve analysis were carried out using a complete set of sieves. In addition, the plasticity index (PI) was determined based on the measured liquid limit (W_L) and plastic limit (W_P). The W_L was measured using a Casagrande apparatus, while the W_P was determined using the rolling method. Determination of the specific density (G_S) of soils was carried out using a Micromeritics gas pycnometer, model Accupyc II 1340. The device is a fully automated material density and volume analyzer, in accordance with ISO 12154:2014 [63]. The tests were carried out in a chamber with a volume of 100 cm³. The grain size distribution curve of the soil used in the study is presented in Figure 1. The figure shows the boundaries between clay (Cl), silt (Si), sand (Sa), and gravel (Gr) fractions. All determined properties are summarized in

Table 1. Properties of the soil.

Properties of Soil	Value	Units
Cc	5.33	(-)
Cu	75	(-)
PI	9.15	(-)
WL	19.03	(%)
WP	9.88	(%)
MDD	1.98	(g∙cm−3)
OMC	8.64	(%)
GS	2.66	(g∙cm−3)

. The soil was classified as clayey sand (clSa).

2.2. Binder

The samples were prepared using composite cement (CEM V) produced at the Górażdże cement plant. The cement was delivered to the laboratory in sealed bags. The cement was stored in the laboratory room and protected from moisture during the testing period. The cement has a declared compressive strength of 10 MPa after 2 days of curing and 32.5 MPa after 28 days. In addition, the specific density of the cement (G_C) was determined using a gas pycnometer. This cement is characterized by low heat of hydration and high resistance to aggressive chemical agents, which makes it applicable in soil stabilization processes. This aspect is particularly important in environments exposed to chemical aggression, as demonstrated in studies on the durability of prefabricated concrete under acid exposure, where the water-to-cement ratio significantly influenced long-term performance [64]. Basic information about the binder is given in

Table 2. Properties of the binder.

Properties	Value	Units
Binder type	Composite cement	(-)
Designation	CEM V/A (S-V) 32,5 R-LH	(-)
GC	2.97	(g∙cm−3)
Cement component	Content Range	Units
Portland cement clinker	40–64	(%)
Granulated blast furnace slag	18–30	(%)
Siliceous fly ash	18–30	(%)
Minor constituent	0–5	(%)
Required compressive strength	Value	Units
After 2 days	≥10.0	(MPa)
After 28 days	≥32.5	(MPa)

.

2.3. Fibers

During the study, dispersed reinforcement was applied, using synthetic fibers. These fibers were produced from organic chemical compounds, specifically polypropylene (PP). The fibers were delivered to the laboratory in airtight packaging, in which they were stored throughout the entire duration of the study. The specific density of polypropylene fibers (G_F) was determined using a gas pycnometer, model Accupyc II 1340 (Micromeritics Instrument Corporation, Norcross, GA, USA). Basic information regarding the fibers is presented in

Table 3. Properties of the fibers.

Properties	Value	Units
Manufacturer	Belgian Fibers NV	(-)
Material	Polypropylene (PP)	(-)
Length	12	(mm)
Diameter	34	(μm)
Tensile strength	40	$(\mathrm{c}\mathrm{N}/\mathrm{T}\mathrm{e}\mathrm{x})$
GF	0.92	(g∙cm−3)

.

2.4. Mixtures, Sample Preparation, and Curing

Nine different mixtures were used to prepare stabilized soil samples for the conducted studies. Each prepared mixture was a combination of soil, binder, water, and, if necessary, dispersed reinforcement. The mixes differed in the percentage of each component relative to the dry weight of the soil. Since the maximum dry density (MDD) in all mixtures was achieved at a moisture content of about 10%, a uniform water content at this level was assumed for sample preparation. Although the approximation curves indicated local differences in optimum moisture content (OMC), this value ensured comparability of results and optimal compaction conditions. The compaction curves of the individual mixtures are presented in Figure 2. In order to uniquely identify the mixtures used in this study, each was assigned a unique designation. Detailed identification of the mixes used is summarized in

Table 4. Mixture types with weight percentages of cement and fibers.

Indication of the Mixture (-)	Cement—C (%)	Fiber—F (%)
C20	2.0	0.00
C21	2.0	0.25
C22	2.0	0.50
C40	4.0	0.00
C41	4.0	0.25
C42	4.0	0.50
C60	6.0	0.00
C61	6.0	0.25
C62	6.0	0.50

.

Based on the mixtures mentioned, 27 samples were prepared according to a uniform procedure to ensure reproducibility and homogeneity of the material. The mixing process began by adding the binder to the previously prepared soil and mixing them using an electric mixer for at least 60 s. For mixtures containing fibers, reinforcement was added, and mixing was conducted again for at least 60 s. After the dry components were thoroughly mixed, the water was added. The final mixing was carried out, which included three stages: preliminary mixing for a minimum of 60 s, manual separation of the mixture from the walls of the container, and basic mixing until a homogeneous consistency was achieved.

After the mixture was prepared, it was first sampled to determine the moisture content and then proceeded to mold samples in accordance with European standards [65]. The samples were prepared in cylindrical molds with a diameter and height equal to 8 cm. Before compaction, each mold was weighed. All samples were compacted at optimum moisture content using a constant energy of 0.59 J·cm⁻³. After the compaction process was completed, the molds were weighed again to determine the initial weight of the samples. Immediately after the samples were compacted, the curing process began, which continued for 28 days. This process consisted of curing the samples in a room with constant humidity and a temperature of 22 °C ± 2 °C.

2.5. Measuring Device—Ultrasonic Pulse Velocity

Non-destructive testing measured the propagation velocity of longitudinal and shear waves in cement-stabilized soil. A Pundit Lab+ device, equipped with two types of transducers with different frequencies, was used for the tests. Longitudinal wave velocities were determined using transducers with a frequency of 54 kHz. A layer of non-toxic gel was used to ensure their proper adhesion to the sample surface and to allow the wave to pass efficiently between the transducer and the sample. Transducers with a frequency of 40 kHz were used to measure the propagation of shear waves. Due to Dry Point Contact (DPC) technology, they did not require a gel layer. The measuring device was calibrated each time before use to ensure the accuracy of the measurements. During non-destructive testing, the direct method of measuring the wave speed was used. In this type of setup, the energy transferred is maximized and the path length is known [65]. In addition, Figure 3 shows a schematic representation of the transmitted pulse and the received signal, from which the wave propagation speed was determined. To interpret the results, a measurement method was used in which the wave travel time through the material is determined as the time difference between the emitted pulse and the received signal. This method allows for determining the signal propagation time in the tested medium and is a standard procedure in material diagnostics, as described by the device manufacturer [66]. All tests were conducted after 28 days of sample treatment, in accordance with European standards [65,67]. Details of the measuring device used are shown in

Table 5. Basic parameters of the Pundit Lab+ device and transducers.

Parameter	Value	Units
Measurement length	0.1–9999	(µs)
Resolution	0.1	(µs)
Frequency	24, 37, 54, 82, 150, 200, 220, 250, 500	(kHz)
Voltage	125, 250, 350, 500	(v)
Used excitation voltage	250	(v)
Calibration time offset	5.6	(µs)

.

Based on the measured longitudinal ($V_P$) and shear wave ($V_S$) velocities and measured densities ($\rho$), deformation parameters were determined in the range of very small strain. Young’s modulus ($E_{UPV}$) was determined according to Equation (1), shear modulus ($G_{UPV}$) was determined according to Equation (2), and Poisson’s ratio (${\nu }_{UPV}$) was determined according to Equation (3):

$$E_{UPV}=2·\rho ·V_S^2·\left(1+{\nu }_{UPV}\right)$$

(1)

$$G_{UPV}=\rho ·V_S^2$$

(2)

$${\nu }_{UPV}={\displaystyle \frac{{V_P}^2-2{V_s}^2}{2\left({V_P}^2-{V_s}^2\right)}}$$

(3)

2.6. Measuring Device—Unconfined Compressive Strength

Unconfined compression tests were conducted using an Instron Universal Testing Machine, model 5982 (Instron, Norwood, MA, USA). During the test, the pressure on the specimens was applied continuously by steel pressure plates, which ensured uniform loading of the test specimens. The machine was equipped with a force sensor and a displacement recorder. The installed sensors allowed the recording of the stress–strain relationship. From the recorded highest compressive force ($F_C$) and contact area ($A_C$), the unconfined compression strength ($R_C$) of the cement-stabilized soil was determined according to Equation (4). In addition, the secant modulus of elasticity ($E_{50}$) was determined, defined as the ratio of half the maximum compressive stress (${\sigma }_{0.5}$) and the corresponding strain (${\epsilon }_{0.5}$), according to Equation (5). Basic information on the measuring apparatus and the accuracy of the measurements is summarized in

Table 6. Basic measurement parameters of the universal testing machine.

Parameter	Value	Units
Accuracy of force measurement	±0.5	(%)
Accuracy of displacement measurement	±0.01	(mm)
Accuracy of load speed	±0.1	(%)
Frequency of data recording	2.5	(kHz)
Load speed	0.05	(N·mm−2·s−1)

. In addition, the effect of fiber addition on the strength classes (C_X/Y) of the stabilized soil was analyzed, according to [68], where X is the unconfined compressive strength of specimens with slenderness 2, and Y is the strength of specimens with slenderness 1:

$$R_C={\displaystyle \frac{F_C}{A_C}}$$

(4)

$$E_{50}={\displaystyle \frac{{\sigma }_{0.5}}{{\epsilon }_{0.5}}}$$

(5)

2.7. Microstructure Tests

The microstructure of the soil stabilized with the addition of fibers was analyzed using a ZEISS AURIGA 60 scanning electron microscope (SEM) (Carl Zeiss AG, Oberkochen, Germany) after the UCS test.

2.8. Indicators Analyzed

Classical soil mechanics treats the soil as a three-phase medium consisting of a solid ($V_{solid}$), liquid ($V_{water}$), and gaseous phase ($V_{air}$). When the soil is stabilized with a binder, such as cement, an additional binding phase ($V_c$) is introduced into the mixture. Furthermore, adding dispersed reinforcement, such as fibers, introduces a structural phase ($V_f$), which changes the nature of the material. As a result, the stabilized and reinforced soil can be considered a five-phase system, consisting of soil particles $\left(V_{sp}\right)$, cement ($V_c$), fibers ($V_f$), water ($V_{water}$), and air ($V_{air}$). In this model, the solid phase ($V_{solid}$) includes the combined volume of soil particles, cement, and fibers. Mixtures that do not contain dispersed reinforcement constitute a four-phase medium. However, due to the chemical reactions between the binder and water, the volumes of the individual phases change over time. At the same time, a new strengthening phase in the form of a C-S-H gel is formed during this process. Due to these changes, it is impossible to control the volume of the individual phases at later stages. For this reason, an analysis based on the volume of the phases in samples prepared before the start of the binding process was conducted in this research. Therefore, the volume of soil stabilized with polypropylene fibers can be expressed as the sum of the volumes of its individual phases. In the conducted analysis, the parameters that determine the ratio of the phase volume to the total volume of the medium were used. The study identified three key parameters, which were then determined for each analyzed sample: porosity (n), cement index ($C_i$), and reinforcement index ($R_i$). Porosity, which is defined as the ratio of the volume of pores ($V_V=V_{air}+V_{water}$) to the volume of sample ($V$), is described by Equation (6). The cement index determines the ratio of $V_C$ to $V$, according to Equation (7). In turn, the reinforcement index describes the quotient of $V_f$ to $V$, as expressed in Equation (8). In the presented formulas, the symbols C and F represent the percentage content of cement and fibers, respectively. The value ${\rho }_d$ refers to the dry density of the sample:

$$n={\displaystyle \frac{V_V}{V}}={\displaystyle \frac{Volume of pores}{Volume of sample}}=100-100\left[\left({\displaystyle \frac{{\rho }_d}{1+\frac{C}{100}+\frac{F}{100}}}\right)\left({\displaystyle \frac{1}{G_S}}+{\displaystyle \frac{\frac{C}{100}}{G_C}}+{\displaystyle \frac{\frac{F}{100}}{G_F}}\right)\right]$$

(6)

$$C_i={\displaystyle \frac{V_C}{V}}={\displaystyle \frac{Volume of cement}{Volume of sample}}=\left[{\displaystyle \frac{{\rho }_d·C}{\left(100+F+C\right)·G_C}}\right]·100$$

(7)

$$R_i={\displaystyle \frac{V_f}{V}}={\displaystyle \frac{Volume of fibres}{Volume of sample}}=\left[{\displaystyle \frac{{\rho }_d·F}{\left(100+F+C\right)·G_F}}\right]·100$$

(8)

3. Results

This section is divided into three subsections to systematize the presented results. Section 3.1 analyzes the effect of the addition of polypropylene fibers on the compressive strength ($R_C$) and deformation parameters ($E_{50}$, $E_{UPV}$, $G_{UPV}$, and ${\nu }_{UPV}$) of cement-stabilized soils. To better demonstrate the influence of polypropylene fibers, the tested material was classified according to the standard guidelines [68]. Furthermore, SEM images illustrating the microstructure of the tested material are presented. Section 3.2 presents the possibility of applying the five-phase medium concept to predict the strength of stabilized soil. Meanwhile, Section 3.3 compares the ability to predict the deformation parameters of stabilized soils based on unconfined compressive strength.

3.1. Impact of Polypropylene Fibers on Stabilized Soil

The results presented in this section represent average values obtained from tests. In addition, the figures show the standard deviations, illustrating the variability of the results obtained for each series. The exceptions are the results for Poisson’s ratio, which are presented without averaging to show more clearly the dependence of this parameter on the strength of the materials analyzed. The stress–strain characteristics obtained during the UCS test, shown in Figure 4, confirmed that adding polypropylene fibers positively affected the compressive strength of cement-stabilized soil. Using fibers in the amount of 0.25% led to a rise in the strength of mixtures containing 2%, 4%, and 6% cement by 5.59%, 25.45%, and 16.08%, respectively. In contrast, adding fibers in the amount of 0.50% increased the strength by 14.1%, 23.76%, and 7.39%, as can be seen in Figure 5. It can be seen that the most significant increase was obtained in mixtures containing 4% cement. On the other hand, the smallest increase was recorded in mixtures containing 2% cement.

This phenomenon may be due to the insufficient strength of the stabilized soil, which limited the ability to form strong connections at the fiber-stabilized soil interface, resulting in lower enhancement efficiency. The situation differed for mixtures characterized by a higher binder content, where strong bonds were formed between the fibers and the stabilized soil due to the increased cement content. In addition, the use of 0.50% fibers in a mixture containing 2% cement resulted in an increase in strength class from C_0.8/1 to C_1.2/1.5. For a mixture with a cement content of 4%, an increase in strength class from C_1.2/1.5 to C_1.5/2 was already achieved by adding 0.25% fibers. In contrast, no change in strength class was observed for mixtures with 6% cement content.

The next parameter analyzed was the $E_{50}$ modulus, determined during unconfined compression testing. During the determination of the value of the $E_{50}$ modulus, an additional correction of the stress–strain curve was conducted. This correction aimed to eliminate the influence of the initial adjustment of the press plate to the surface of the specimen. It was performed according to the procedure described in [69]. Based on the data shown in Figure 6, it was observed that the use of dispersed reinforcement contributed to a reduction in the value of the $E_{50}$ modulus up to 32.49% of the value of this parameter in the unreinforced mix. The exceptions were the C21 and C61 series, which showed an increase in the value of the strain modulus ranging from 9.94% to 10.37%. However, it should be noted that the presented results were characterized by a much larger standard deviation than the results obtained during strength tests of stabilized soils. The observed phenomenon may be due to the fact that the use of polypropylene fibers changed the characteristics of the stabilized soil, converting it from a brittle material to one with more plastic properties [70].

Subsequently, the values of the elastic moduli $E_{UPV}$, $G_{UPV}$, and ${\nu }_{UPV}$ were determined based on the measured ultrasonic pulse velocities. Analogously to the $E_{50}$ modulus, it was observed that using dispersed reinforcement reduced the values of the mentioned parameters. The effect of adding polypropylene fibers on the $E_{UPV}$ modulus is shown in Figure 7, while the impact on the $G_{UPV}$ modulus is shown in Figure 8. The results obtained were characterized by high repeatability, as reflected by the small values of standard deviations for the analyzed series. The most significant decrease in modulus values was observed for the mix containing 2% cement and 0.50% fiber, where the value of the $E_{UPV}$ modulus decreased to 76.72% and the value of the $G_{UPV}$ modulus to 77.41% compared to the mix without dispersed reinforcement. Module increases were observed in some series: the $E_{UPV}$ module increased by 1.18% in series C61 and by 6.94% in series C41, while the $G_{UPV}$ module increased by 2.32% in series C62 and by 8.43% in series C41.

The results shown in Figure 9 illustrate the effect of polypropylene fibers on the value of Poisson’s ratio and its dependence on unconfined compressive strength. A decrease in Poisson’s ratio with increasing compressive strength was observed in stabilized soil without the addition of fibers and with the addition of 0.25% fibers. In contrast, in specimens containing 0.5% fibers, no significant changes in this coefficient were observed with increasing strength. Regardless of the amount of fiber used, their addition to the mixtures led to an overall decrease in the Poisson’s ratio. This phenomenon may be due to more interfaces between the fibers and the stabilized soil, which the propagating seismic wave must overcome [52]. One sample from the C20 series, which was considered an outlier, was excluded from the analysis.

The SEM images shown in Figure 10 were taken after the UCS test. The purpose of the analysis was to evaluate the material’s microstructure, with particular emphasis on the characteristics of the connection between the polypropylene fibers and the cement matrix and the presence of their local clusters. The images show polypropylene fibers embedded in the surrounding structure, but their contact surface with the matrix was not entirely homogeneous. In many places, micro-gaps and small spaces separating the fibers from the cement were observed, which may indicate limited adhesion and incomplete stress transfer between these components. Such an effect could result from insufficient mechanical bonding or weakened contact caused by stress during UCS testing. In addition, the accumulation of several fibers nearby is noticeable in Figure 10a, indicating the formation of local clusters. Such inhomogeneous distribution of fibers can lead to local disruption of the structure and, consequently, to a decrease in the homogeneity of the material. In the zones of clustering, there were voids and disturbances in the continuity of the cement matrix, which can negatively affect the strength and deformation parameters of the stabilized soil. In comparison, Figure 10b shows a fiber–cement matrix bond of a more homogeneous nature, with fewer gaps and voids, indicating better cooperation between the fiber and the surrounding material. These observations suggest that despite the anchoring of the fibers in the structure, their effectiveness in transferring stresses depends on both the quality of the connection to the matrix and the uniformity of their distribution in the stabilized soil.

3.2. Predicting the Strength of Stabilized Soil

The five-phase medium concept used is a development of the method for predicting the strength of stabilized soil based on the porosity and cement index [71], which is described by Equation (9). The effectiveness of this method has been confirmed for both stabilized soil without and with dispersed reinforcement. However, empirical relationships based only on the porosity and cement index do not take into account the effect of the content of dispersed reinforcement. Therefore, it is impossible to determine a universal relationship describing the strength of the material, which is why some authors have proposed normalized relationships that can also be successfully applied:

$$R_C=X{\left[{\displaystyle \frac{n}{C_i}}\right]}^Y$$

(9)

In the analysis conducted, dispersed reinforcement was included in the relationship described by Equation (9), which was taken into account in the form of a reinforcement index. This index was included as the difference between 1 and $R_i$. This procedure is because a value of $R_i=0$ is assumed for mixtures that do not contain polypropylene fibers. In addition, the exponent A of the reinforcement index was introduced to calibrate the coefficient. After the implementation of the mentioned factors, the new relationship was described by Equation (10). In order to illustrate the impact of the introduced index, an analysis of Equation (9) was additionally conducted:

$$R_C=X{\left[{\displaystyle \frac{n \left(1-{\left(R_i/100\right)}^A\right)}{C_i}}\right]}^Y$$

(10)

The data presented in Figure 11 show that the coefficient of determination for the relationship between $R_C$ and $n/C_i$ obtained a very high value for both mixtures without and with polypropylene fibers. In addition, a strong influence of dispersed reinforcement could be observed, which increased the strength of uniaxial compression. This result confirmed the possibility of using Equation (9) to predict the strength of stabilized soil and highlighted some limitations of such an approach.

Subsequently, Figure 12 shows the empirical relationship determined from Equation (10), where the coefficient $A=0.3$ was used. The value of the A coefficient was determined by an analysis to identify the value that provides the maximum coefficient of determination. The data presented confirmed that using the five-phase medium concept made it possible to determine the relationship for predicting the strength of the material, both with and without fibers. The determined empirical relationship also made it possible to achieve a very high coefficient of determination, which was 0.97. In addition, the presented relationship had a MAPE = 5.37%, MAE = 0.12 MPa, and RMSE 0.14 MPa. The stats confirmed the trend line’s good fit.

Figure 13 shows the enhancement surfaces of stabilized soil as a function of the reinforcement index and cement index. These surfaces can be described by Equation (11), where the coefficients were $a=-0.50493182$, $b=-0.014086175$, $c=0.43853229$, $d=0.27073498$, and $e=0.87542443$, respectively, assuming the porosity was constant. Based on the data presented, it can be concluded that using $R_i=0.5\%$ increased the strength of the stabilized soil by about 22.15%. In comparison, using $R_i=1.0\%$ increased it by up to 28.83%. In addition, further increases in fiber content above the analyzed range can lead to disproportionate increases in the compressive strength of the stabilized soil:

$${\mathrm{l}\mathrm{n} (R}_C)=a+b{R_i}^2+c{R_i}^{0.5}+dEXP(-R_i)+eln\left(C_i\right)$$

(11)

3.3. Prediction of Deformation Parameters

In engineering practice, deformation parameters are most often correlated with the unconfined compressive strength of the stabilized soil. This approach has been used successfully for years. However, based on the presented research results, there are indications that the addition of polypropylene fibers increased unconfined compressive strength and decreased strain parameters. In order to assess whether the fibers did not significantly affect the stiffness of the material, an additional analysis was carried out. During the analysis, linear relationships between moduli ($E_{50}$, $E_{UPV}$, and $G_{UPV}$) and R_C were investigated. Relationships are described using Equation (12). This approach will make it possible to assess whether polypropylene fibers affect the deformation parameters of stabilized soil. It should be noted that the analysis did not consider Poisson’s ratio, which showed a decreasing trend as the strength of the tested material increased. The analysis was not conducted due to the very small variation in the results obtained, which ranged from 0.298 to 0.335, which can be considered a typical range for the tested material:

$$E_{50},E_{UPV},G_{UPV}=X\left[R_C\right]+Y$$

(12)

Based on the analysis of the relationship between the $E_{50}$ modulus and $R_C$, it can be concluded that polypropylene fibers reduced the value of this modulus in relation to the unit strength of the material. An analysis of Figure 14 revealed that the rate of increase in the $E_{50}$ modulus was similar in both the series with 0.50% dispersed reinforcement and the series without it. However, due to the use of fibers, the intercept coefficient decreased significantly while the slope coefficient increased slightly in the series of tests with fibers. The use of 0.25% fibers also showed a reduction in modulus values concerning the series without dispersed reinforcement, but the slope coefficient decreased this time. Among the determined relationships, the trend line had the highest accuracy for the series containing 0.50% polypropylene fibers: MAPE = 9.88%, MAE = 17.26 MPa, and RMSE = 20.04 MPa. Next, the series with 0.25% fibers were characterized by MAPE = 11.26%, MAE = 30.49 MPa, and RMSE = 37.04 MPa. The lowest accuracy of $E_{50}$ modulus predictions was obtained for the series without dispersed reinforcement: MAPE = 13.18%, MAE = 39.65 MPa, and RMSE = 48.82 MPa.

Similar relationships were observed during analysis of the moduli determined from UPV tests. The use of polypropylene fibers also contributed to a reduction in $E_{UPV}$, as shown in Figure 15. However, in this case, all the determined relationships were almost parallel. In addition, the determined dependencies were characterized by a very accurate fit to the experimental data. This was confirmed by the results of the statistical analysis, where the series without fibers obtained: MAPE = 2.10%, MAE = 0.15 GPa, and RMSE = 0.18 GPa. The series with 0.25% dispersed reinforcement obtained: MAPE = 2.49%, MAE = 0.17 GPa, and RMSE = 0.19 GPa, while the series with 0.5% fibers recorded: MAPE = 1.88%, MAE = 0.12 GPa, and RMSE = 0.14 GPa. Since the $G_{UPV}$ was determined based on the same velocities as the $E_{UPV}$, the obtained relationships were also characterized by very high accuracy and similar influence of polypropylene fibers, as shown in Figure 16. As confirmed by the results of statistical analysis, the series without fibers obtained: MAPE = 2.64%, MAE = 0.07 GPa, and RMSE = 0.08 GPa. The series with 0.25% fibers obtained: MAPE = 2.31%, MAE = 0.06 GPa, and RMSE = 0.07 GPa, while the series with 0.50% fibers recorded: MAPE = 1.65%, MAE = 0.04 GPa, and RMSE = 0.05 GPa.

Subsequently, using the determined relationships, an analysis was conducted to determine the degree of modulus reduction induced by polypropylene fibers. Analyzing Figure 17, which shows the effect of 0.25% polypropylene fibers, it can be seen that the trend of modulus change was similar for each module considered. It can be assumed that the $E_{50}$ modulus decreased on average 4.73% more than the module determined by wave speed. Meanwhile, Figure 18 shows a significant decrease in the $E_{50}$ modulus at lower $R_C$ values. In addition, it should be noted that due to the addition of polypropylene fibers, the $E_{50}$ modulus decreased to a greater extent than the $E_{UPV}$ and $G_{UPV}$ moduli. This was particularly evident when 0.50% fibers were used. In addition, as the strength of the stabilized soil increased, this effect weakened.

The results presented refer to the strength level of the stabilized soil, which allowed for the evaluation of changes in modulus values in mixtures where the binder content was reduced and polypropylene fiber reinforcement was introduced. This way of analysis revealed a critical problem associated with using dispersed reinforcement to compensate for the reduction in cement content. This issue arose from the fact that the cement matrix played the primary role in determining the material’s stiffness. As a result, although reducing the cement content and adding fibers can help achieve a compressive strength comparable to that of mixtures with more binder but no fibers, it led to a reduction in the elastic modulus of the stabilized soil. However, it should be noted that this effect, when considered in percentage terms, diminished as the strength level of the material increased.

4. Discussion

The presented results proved that it is possible to predict the unconfined compressive strength by considering the soil stabilized by cement with the addition of polypropylene fibers as a five-phase medium. Implementing a reinforcement index into the equation allowed the effect of distributed reinforcement to be included in the empirical strength relationships. The described method is more universally applicable than one based solely on the binder and dispersed reinforcement amount. What follows is the inclusion of porosity in the equation. This parameter depends, among other things, on the type of soil, moisture content, and degree of compaction of the mixture. The listed parameters showed significant variability resulting from soil-related and technological conditions.

Based on the described concept of a five-phase medium, a relation was determined for predicting unconfined compressive strength, which had an average absolute percentage error of 5.37%. Due to such a small error, this paper decided to omit the cement index exponent, which numerous authors have successfully used [7,58,72,73,74]. Based on the relationship used to predict the unconfined compressive strength, shown in Figure 12, it was found that adding polypropylene fibers in the amount of $R_i=0.5\%$ increased the strength of the material by 22.11%. In contrast, using $R_i=1.0\%$ increased this strength by 28.82%, regardless of the porosity adopted. After transforming this relationship, as illustrated by Equation (13), it was possible to calculate the necessary value of $C_i$ to achieve the assumed strength. The analysis showed that $R_C=2.5$ MPa can be achieved without dispersed reinforcement at a porosity of n = 30% and a cement index of $C_i=3.72\%$. Adding fibers corresponding to a reinforcement index of $R_i=0.5\%$ (≈0.25% fibers) allowed for achieving the same strength with a reduced cement index to $C_i=2.96\%$, representing a 20.4% reduction in cement content. Similarly, using a reinforcement index of $R_i=1.0\%$ (≈0.50% fibers) further reduced $C_i$ to 2.79% while maintaining the same strength, corresponding to a 25.12% cement reduction:

$$C_i={\displaystyle \frac{n\left(1-{\left(R_i/100\right)}^{0.3}\right)}{{\left(\frac{R_C}{15.54}\right)}^{\left(\frac{1}{-0.88}\right)}}}$$

(13)

The increase in the unconfined compressive strength of stabilized soil due to the addition of fibers has also been confirmed by other authors [33,35,36,37,38]. The study showed a relationship between the increase in strength and the amount of dispersed reinforcement used in the mix. A small addition of polypropylene fibers could contribute to an increase in unconfined compressive strength. However, as the amount of fibers increased, this effect disappeared. Therefore, it can be considered that polypropylene fibers in the amount of 0.25% was the optimal amount in the present study. Other authors also made similar observations, indicating that a polypropylene fiber content of 0.25% led to the most pronounced increase in strength, suggesting that this may be the optimal value [35,37]. Furthermore, the researchers noted that higher cement content reduces the optimal amount of dispersed reinforcement [36].

Therefore, it can be concluded that the increase in strength was the result of two soil enhancement mechanisms: cement hydration and the addition of polypropylene fibers. The strength gained during the hydration process is due to the chemical reactions that lead to the formation of calcium silicate hydrate (C-S-H), which is responsible for the strength development of the stabilized soil. On the other hand, the strength gain resulting from adding polypropylene fibers is because stabilized soil with fiber behaves like a composite material, in which high-tensile-strength fibers are embedded in the soil–cement matrix. The generated shear stresses mobilize tensile resistance in the fibers, which led to an overall increase in the material’s strength.

Combining the two soil improvement methods can lead to different improvement effects depending on the proportion of ingredients used. The cement matrix is relatively weak and has limited strength at low binder content. Adding polypropylene fibers can improve the overall strength of the stabilized soil in this case. Still, the effectiveness of this improvement will be limited due to the weak bond between the fibers and the cement matrix. Subsequently, as the amount of binder increases, the cement matrix’s role in shaping the material’s strength becomes dominant. At the same time, stronger fiber–matrix bonds begin to form, which can increase the effectiveness of their application. However, it should be noted that an excessive amount of fibers can lead to the formation of local clusters of fibers, which disrupts the homogeneity of the material’s structure and, consequently, can weaken the cement matrix. This effect is particularly noticeable in higher cement contents, where disruption of the structure can negatively affect the quality of the material. In such cases, the weakening of the matrix may exceed the benefits of dispersed reinforcement. This phenomenon may explain the observed decrease in the improvement efficiency of the C42 and C62 mixes.

Variability in the efficiency of soil enhancement is manifested not only in its strength but also in its deformation properties. The analysis conducted showed a strong correlation between unconfined compressive strength and deformation parameters of stabilized soil. The deformation moduli, such as $E_{50}$, $E_{UPV}$, and $G_{UPV}$, increased with increasing strength. Analysis of regression plots showed a significant effect of dispersed reinforcement on the relationship between unconfined compressive strength and the moduli $E_{50}$, $E_{UPV}$, and $G_{UPV}$. As the polypropylene fiber content increased, there was a systematic decrease in modulus values for the same compressive strength, although the slopes of the regression lines remained comparable. Although the fibers contributed to the strength, they did not cause a proportional increase in the material’s stiffness. The observed trend indicated that stiffness depended primarily on the binder content in the mixture, while the presence of polypropylene fibers may reduce the material’s ability to transfer deformation. The phenomenon may result from the following factors:

• The inclusion of polypropylene fibers in the cement-treated soil mixture may lead to a loosening of its microstructure. This is a likely cause, as the use of polypropylene fibers decreased the dry unit weight, which in turn increased the porosity of the compacted mixture. Consequently, a more porous and less homogeneous cement-ground structure may have a lower stiffness, which translates into a reduction in the value of the deformation modulus.
• A smooth and hydrophobic surface characterizes polypropylene fibers, so the bond between the fiber and the cement matrix may be limited. Such factors can form a zone with weaker cohesion, making it more susceptible to microcracks. As a consequence, the fibers may slip out or detach from the cement matrix before the test material reaches complete failure.
• Polypropylene fibers can affect the homogeneity of the cement-ground mixture, especially when using a higher amount of dispersed reinforcement or longer fibers. In such cases, the risk of forming local aggregations of dispersed reinforcement increases, which can lead to local weakening of the material. The effect of such a phenomenon can be a reduction in the modulus of elasticity of the test material.

Therefore, it can be concluded that polypropylene fibers primarily had a positive effect on the strength of the stabilized soil, and their use allowed the amount of binder in the mix to be reduced without reducing the unconfined compressive strength. The consequence of reducing the amount of binder was a decrease in the stiffness of the stabilized soil. However, this effect disappeared in percentage terms as the strength of the stabilized soil increased, as shown in Figure 17 and Figure 18. Similar observations on the impact of adding polypropylene fibers on the material’s properties have also been reported in other research works, confirming the consistency of the results obtained with the observations of other authors [42,43,75]. Additionally, an increase in $E_{UPV}$ and $G_{UPV}$ moduli values was recorded in the C41 and C61 mixtures. This may be due to the fact that, in these cases, the cement matrix was already sufficiently developed and rigid enough, and the amount of fiber used was small enough not to cause adverse structural effects, such as increased porosity or local weakening. It is also worth noting that a significant increase in unconfined compressive strength was recorded in these series simultaneously, confirming the effective cooperation between the composite components. In turn, for the $E_{50}$ modulus, an increase was observed in the C21 and C61 mixtures; however, the results in this case showed greater variability, which may have affected the obtained average values.

Also worth noting is that despite the widespread use of correlations of deformation parameters with compressive strength, there are indications for the search for new relationships to predict the deformation parameters of stabilized soil both with and without the addition of fibers. Considering the results presented, it should be noted that although the strength of the stabilized soil was the primary evaluation criterion, it may be equally important to consider its deformation properties at the design stage of the mix composition. This issue becomes critical in the context of the use of polypropylene fibers in the mix composition. Therefore, it is necessary to determine the parameters that significantly affect the deformability of soil stabilized by adding polypropylene fibers.

5. Conclusions

The conducted experimental study allowed for the assessment of the influence of dispersed reinforcement in the form of polypropylene fibers on the mechanical properties of stabilized soil. In addition, the analysis evaluated the possibility of using a five-phase medium to predict unconfined compressive strength. A novelty introduced in the paper was the inclusion of a reinforcement index in the empirical relationship. This index is a component of material description that captures the effect of adding polypropylene fibers. The use of this index allowed a more precise prediction of the strength of stabilized soil, taking into account both the cement content and the dispersed reinforcement. The proposed approach extends existing predictive models and increases their usefulness in designing the composition of a stabilized soil mixture.

Above that, the relationship between unconfined compressive strength and deformation parameters was investigated, considering the influence of dispersed reinforcement. An additional statistical analysis was conducted to determine the empirical relationships, during which MAE, RMSE, and MAPE were determined. However, it should be noted that the empirical relationships presented have limited applicability due to the use of a single type of soil, fiber, and cement. The variability of soil properties makes it likely that the effects of using dispersed reinforcement can vary significantly from one type of soil to another. Therefore, in the following stages of the research, it will be necessary to extend the analyses to a wider range of materials, which will allow us to verify the results obtained and increase the universality and practical applicability of the developed relationships. The following are the main conclusions drawn from the results obtained in this study:

• The addition of polypropylene fibers reduced the maximum dry density of the mixtures, regardless of the cement content used.
• The use of polypropylene fibers increased the unconfined compressive strength of cement-stabilized soil. The increase ranged from 5.59% to 25.45%, with the greatest enhancement observed in mixtures containing 4% cement.
• The use of polypropylene fibers allowed for a significant reduction in the amount of cement without reducing the compressive strength of the tested material. The implementation of $R_i=0.5\%$ reduced the cement content by 20.4%, while increasing $R_i$ to 1.0% reduced it by as much as 25.12%.
• The addition of polypropylene fibers in the amount of 0.5% caused a decrease in the values of deformation moduli: $E_{50}$ (7.46–67.51%), $E_{UPV}$ (7.21–23.28%), and $G_{UPV}$ (5.8–22.59%). For a fiber content of 0.25%, the results were inconclusive.
• The Poisson’s ratio decreased slightly as the compressive strength of the stabilized soil increased, with the addition of polypropylene fibers further contributing to its decrease.
• The use of porosity, cement, and reinforcement indexes allowed accurate prediction of the compressive strength of soil stabilized with the addition of polypropylene fibers, which confirmed the validity of this approach.
• Polypropylene fibers significantly affected the relationship between unconfined compressive strength and deformation parameters. Their presence decreased the value of elastic moduli with respect to the unit compressive strength, which indicated a modification of the material’s deformation properties. This effect intensified as the content of dispersed reinforcement increased.

These findings contribute to improved mechanical performance of stabilized soils and align with global efforts toward sustainable construction practices. By enabling a significant reduction in cement usage—one of the major contributors to CO₂ emissions in the construction sector—the application of polypropylene fibers supports the decarbonization of ground improvement techniques. The proposed predictive models based on porosity and reinforcement indices also promote material efficiency and resource optimization, further reinforcing the sustainability potential of this approach. Therefore, dispersed reinforcement can be considered a technical enhancement and an eco-friendly strategy for modern geotechnical engineering.