Enhancement of concrete performance and sustainability through incorporation of diverse waste carpet fibres

Carpet fibres have demonstrated the potential to mitigate early-age cracking and improve tensile properties in concrete. However, a detailed analysis of the varied types of standard carpet fibres in reinforced concrete has been lacking. This study aims to bridge this gap by investigating the performance of concrete reinforced with widely used waste carpet fibres, namely Nylon, Polypropylene, Polytrimethylene terephthalate, and Polyester. The study employs fibres at 0.3 % and 0.5 % volume fractions with a 12 mm length. The research examines mechanical properties, shrinkage and cracking behaviour, pore structure, microstructure, and the ITZ. Results show that 0.3 % fibre volume yielded optimal performance based on GRA analysis. All fibre types reduced shrinkage compared to the control with no fibres. Nylon T1 at 0.3 % achieved a 22.3 % reduction at 90 days. Furthermore, fibre inclusion enhanced flexural and splitting tensile strengths up to 12 % and 39 % respectively due to fibre bridging, pore refinement, and reduced porosity. Notably, individual fibre mechanical properties influenced concrete performance significantly. Hydrophilic fibres exhibited a thinner 10 µ m ITZ compared to 15 µ m for hydrophobic fibres, contributing to denser interfacial regions and improved bonding. This study demonstrates the potential of carpet fibre-reinforced concrete as a sustainable solution, offering enhanced mechanical properties, shrinkage mitigation, and effective utilization of carpet waste, addressing critical issues in construction and waste management sectors.


Introduction
Carpets serve as flooring materials designed to enhance comfort and find extensive use in various settings such as offices, commercial complexes, industries etc. [1], with over 70 % of floor coverings are carpets [2].The extensive utilization of carpets, however, results in a massive annual generation of carpet waste.For an example, in Europe, the total annual carpet waste generated is approximately 1.6 million tons, with 60 % directed to landfills and the remainder sent to be incinerated in specialized plants [3].In the United Kingdom, there is an annual generation of 400,000 tons of carpet waste, with half of this amount ending up in landfills [4].The United States, which has the largest carpet market, generated 3.4 million tons of carpet waste in 2018, and it is 3.5 % from the total waste disposed in U.S landfills each year.Meanwhile, in New South Wales, Australia, the disposal of carpets alone account for 2 % of the materials sent to landfills [3].Despite carpets having a lifespan of 8-10 years, a majority are composed of synthetic fibres that last for several hundred years [1].The non-compressible nature of carpets contributes to disorder in waste disposal when they are landfilled.Moreover, incinerating carpet waste releases various toxic gases, creating environmental concerns [4].Consequently, the recycling of carpet waste has become a crucial strategy to mitigate environmental impacts and strengthen the economies of nations.
On the other hand, concrete a predominant material in the construction industry, has a number of drawbacks, particularly its weakness in tension, resulting in the development of cracks.These limitations, cause early-age cracking which can significantly impact the durability of structures.In the Australian context, the annual cost of corrosion repair for reinforced concrete structures is estimated at 8 billion dollars, whereas in the United States, it is 76 billion dollars [5]. Furthermore, the process of demolition and subsequent repairs generates a substantial amount of waste, approximately 27 million tons in Australia for the 2018-2019 period [6].In addressing these challenges, past research has highlighted the potential to incorporate artificial fibres to enhance concrete properties, particularly in delaying cracking and improving overall structural durability [7][8][9][10][11][12].In addition, natural fibres, originating from plants or animals, have shown potential for use in cementitious composites.[13][14][15][16][17][18].Moreover, to enhance the durability of natural fibres such as vegetable fibres, flax fibres, and luffa fibres, surface treatment methods should be applied.The modifications can be broadly divided into chemical, physical and combined methods [19,20].While the inclusion of virgin fibres may cause significant costs, the utilization of waste fibres presents a sustainable solution, aligning with principles of a circular economy [4].Wang et al. [21,22] demonstrated that the addition of 1 % carpet fibres significantly improved compressive strength by between 12.9 % and 17.5 % at 28 days.Moreover, this research indicated that concrete strengthened with 2 % carpet fibres exhibited superior flexural toughness compared to concrete incorporating 0.5 % virgin polypropylene (PP) fibres.Additionally, the inclusion of 1-2 % carpet fibres by volume demonstrated a substantial reduction in shrinkage, ranging from 15 % to 30 %.Similarly, according to Zareei et al. [23], introducing 1 % polypropylene (PP) fibres resulted in an 11 % decrease in compressive strength at 28 days, but it enhanced flexural strength by 5 %.The aspect ratio of the fibres was maintained at 0.45.Furthermore, Mohammadhosseini et al. [24] analysed the characteristics of concrete reinforced with fibres using 0.25 % and 1.25 % PP carpet waste fibre volumes.Their findings revealed that an increase in fibre volume caused in a reduction of the compressive strength while concurrently enhancing tensile strength.The study also observed that fibre inclusion positively influenced impact resistance and energy absorption capacity.There are many studies that focused on investigating the behaviour of PP fibres in concrete.Ghanim et al. [11] focused on the behaviour of PP fibres in lightweight concrete.Mohammadhosseini et al. [25] observed the mechanical characteristics of PP fibres in preplaced aggregate fibre-reinforced concrete, and in another study [26], they identified the effect of PP fibres at elevated temperatures.Similarly, many other researchers have analysed the influence of PP fibres on concrete under various conditions [3,[27][28][29][30][31].In addition, Wu et al. [32] noted that a 0.12 % volume of fibre content in recycled aggregate concrete resulted in an impressive 43 % increment in flexural strength while simultaneously achieving a significant 9.72 % reduction in shrinkage.These results underscore the capability of fibre reinforcement as a realistic strategy for improving the mechanical properties and sustainability of concrete structures.Building on this concept, carpet fibre-reinforced concrete emerges as a promising material with potential applications in various construction sectors.Its shrinkage and crack reduction properties make it particularly suitable for on-ground slabs, airport runways, pavements, sidewalks, and flooring.Furthermore, it could be used for shotcrete applications in soil stabilization and tunnel construction.It also demonstrates potential for use in sound barriers along highways and similar locations.However, it is crucial to emphasize that the properties and performance characteristics of this material must undergo thorough testing and verification before it can be considered for critical structural applications.
The literature demonstrates the significant impact of utilizing carpet waste fibres in concrete, with the majority of previous studies focusing predominantly on the behaviour of polypropylene (PP) carpet fibres.However, a critical gap in the research exists regarding the utilization and performance of the wide range of fibre types, each with distinct characteristics, commonly employed in the carpet industry.There is a notable scarcity of studies examining the microstructural characteristics and pore structure development in concrete mixtures reinforced with diverse carpet fibres.Addressing these research gaps could significantly advance the potential for incorporating a range of waste carpet fibres in construction, potentially leading to more sustainable and resourceefficient building practices.This study aims to bridge this critical knowledge gap by exploring the behaviour of different carpet fibres, namely Nylon, PP, Polytrimethylene terephthalate (PTT), and Polyester, which are among the most commonly used fibre types in carpet manufacturing.Notably, Nylon and Polyester rank among the highest in global production, while PTT is recognized for its exceptional strength and stiffness.The investigation provides a comprehensive understanding of the effects of these diverse carpet fibre types on the pore structure, microstructure, and interfacial transition zone (ITZ) thickness between the fibres and the cement matrix.By examining with different fibre volume fractions (0.3 % and 0.5 %) with a consistent fibre length of 12 mm, the research investigates the mechanical properties, shrinkage behaviour, microstructural evolution, pore structure, and ITZ thickness alterations.This study holds significance as it contributes to the sustainable utilization of carpet waste in concrete applications by informing the optimized selection and utilization of carpet fibres and provides an insight for more innovative and sustainable practices.

Material preparation
In this investigation, General-purpose Portland cement (Type 1), in accordance with AS 3972-2010, was used as the primary binding material.ASTM Class-F fly ash with low calcium content, meeting the specifications of AS 3582.1:2016, was utilized, sourced from the Eraring power station in Australia, having a specific gravity of 2.4.XRF analysis of the fly ash revealed the following chemical compositions: SiO 2 (65.73 %), Al 2 O 3 (24.21%), Fe 2 O 3 (2.95%), K 2 O (1.79 %), CaO (1.60 %), and others (3.72 %).The chosen basalt coarse aggregates comprised two types, one with a maximum size of 7 mm and the other 10 mm, both sourced from Pyramid Hill, Victoria, Australia.The fine aggregates comprised uncrushed river sand, possessing a fineness modulus of 2.65 (according to ASTM C33/C33M-23) and a specific gravity of 2.63 (in accordance with ASTM C128-22).To maintain consistency in slump and enhance early strength, Sika ViscoCrete 20 HE high-range water-reducing admixture (HRWRA) was incorporated into the concrete mix.This study incorporated four types of carpet fibre, namely Nylon, Polypropylene (PP), Polyester, and Polytrimethylene terephthalate (PTT).Both Nylon and Polytrimethylene terephthalate fibres have two types with different diameters.Fig. 1 shows the physical and SEM images of each type of fibre.The fibres were sourced from Godfrey Hirst Australia Pty Ltd.These fibres were in the form of bundles of yarns and were cut into a standardized length of 12 mm, based on prior research by Tran et al. [3].The physical properties of the fibres were observed via Scanning Electron Microscopy (SEM).Meanwhile, the mechanical properties of individual fibres were characterized using an Instron machine with a 2.5 N load cell, and detailed fibre characteristics are presented in Table 1.Five samples from each fibre type were tested to obtain the characteristics of the fibres.

Mix design & casting
In this study, twenty percent by weight of the total binder content consisted of added fly ash, while maintaining a coarse aggregate to fine aggregates ratio of 1.56.The water-to-binder ratio was fixed at 0.4, determined through trials to obtain consistency with a slump range of 100-140 mm.The amount of cement, sand, fly ash, 7  ), were incorporated in two volumes (0.3 % and 0.5 %) for performance evaluation.The concrete mixtures were prepared using a horizontal drum mixer with a capacity of 120 litters.
Coarse and fine aggregates were initially blended for 2 minutes, after which Portland cement was added, and fly ash with a subsequent 3-minute mixing.Fibres were gently added to prevent fibre-balling behaviour and mixed for an additional 1.5 minutes.The superplasticizer was mixed with water and added to the mixture, with continuous mixing for another 1.5 minutes.Fig. 2 illustrates the production process of the fibre reinforced concrete.The homogeneous concrete was poured into 200 mm × 100 mm cylindrical moulds for compressive strength, 100 mm × 100 mm × 350 mm prism moulds for flexural strength, (Fig. 3(a)).The moulds, filled in three layers, were underwent compaction using a vibrating table in accordance with AS 1012.8.1 (ASTM C1170) without a delay.The samples were kept in controlled laboratory surroundings (23 ) for a duration of 24 hours prior to the demoulding process and subsequently cured in lime-saturated water until testing.In the case of shrinkage specimens, initial lengths were measured after 7 days of curing, after which they were transferred to a humidity chamber set at 50 % ± 2 % relative humidity and 23 Optimized samples were analysed by SEM, nanoindentation, and X-ray micro-CT at 7 and 28 days.Specimens for these tests were cut into 15 mm × 15 mm × 15 mm sections using a diamond saw, immersed in acetone for 24 hours, and kept in a desiccator for another 24 hours.These were then embedded in epoxy resin in 25 mm diameter Teflon moulds and dried for one day.The upper surface underwent grinding using silicon carbide grinding papers ranging from 400 to 1200 grit, followed by polishing with the Struers Tegramin -25 instruments.The samples went through a three-step polishing process.Firstly, MD -Largo plates were used with DiaPro Allegro/ Largo 9 µm abrasive spray at a speed of 150 rpm and a force of 15 N for 6 minutes.Subsequently, the samples were polished using MD-Dac plate with DiaPro Dac 3 µm abrasive at a speed of 150 rpm and a force of 15 N for 5 minutes.Finally, the samples were polished using MD -Nap plate with DiaPro Nap B 1 µm abrasive at a speed of 150 rpm with a force of 10 N for 2 minutes.Samples designated for nanoindentation were utilized without further treatment [33,34], while those intended for scanning electron microscopy (SEM) underwent a final preparation stage involving the application of a 5 nm thick iridium coating to minimize charging.Concrete specimens for Micro-CT scans were cut into 20 × 20 × 20 mm 3 size and immersed in acetone for 24 hours.N. Gamage et al.

Experimental testing
In this study, workability of the mixes was identified by conducting the slump test complying with ASTM C143/C143M -20 and a Trueforce 5000 kN compression machine was employed to conduct compressive strength, splitting tensile strength, and flexural strength tests on the specimens.(Fig. 3

(b), (c) & (d))
The tests were performed at 7, 28, and 90 days, with reported values representing the average of four-cylinder samples for each mix in the compressive strength test, three prism specimens for flexural and three-cylinder specimens for splitting tensile strength tests.The compressive test was conducted following AS 1012.9 standards, applying a loading rate of 2620 N/s.For the splitting tensile strength test, a loading rate of 830 N/s was applied, following compliance with AS 1012.11.The flexural strength test is done through a fourpoint bending test with a loading rate of 500 N/s, employing a steel loading apparatus.To measure unrestrained drying shrinkage, length changes were monitored using a horizontal comparator and it was tested from 7 to 90 days at 7-day intervals.Shrinkage rates were assessed following the guidelines of ASTM C157/C157M-17 (ASTM 2017).(Fig. 3 (e)) Computed tomography (CT) scans were performed using a Bruker SkyScan 1275 with specific parameters: 15 mm pixel size, a 0.2 • rotation step, 100 kV voltage, 100μA current and 1.00 mm thick copper filter.After the data acquisition, the reconstruction of the structure was carried out using the Nrecon software supplied by Bruker.The CTan program was employed for data analysis and characterization of porous internal structures, utilizing a carefully determined threshold value to ensure precise classification of pores and the cement matrix.SEM analysis was carried out using an FEI Nova Nano SEM.It was operated at a voltage of 15 kV and with a 5 mm working distance.The Hysitron T1-950 nano indenter was employed for depth-sensing indentation, utilizing a loadcontrolled approach (Pmax = 1000 µN) with a Berkovich tip.The loading and unloading rate were set at 0.2 mN/s, and a 10-second holding phase was implemented to prevent creep and plastic effects.A grid of 10 × 3 indentations with a 5 µm spacing was utilized on the fibrematrix interface in the optimized mixtures at both 7 and 28 days.Using the data, hardness and elastic modulus values at the indentation points were obtained to determine the thickness of the ITZ and observe the behaviour of H and E values from close to the fibre surface to the cement matrix.

Fibre reinforced concrete optimization method
The optimization and ranking of the performance of fibre-reinforced concrete mixtures were done through the application of the Grey Relation Analysis (GRA) method.This methodology is suitable for analysing multiple performance characteristics simultaneously, offering the advantage of transforming diverse performance metrics into a singular objective optimization problem [35].In the initial step, the data is transfer from the original sequence to a comparable sequence, wherein numerical data were normalized to a range between zero and one.The normalization process employed distinct methods based on the nature of the data series, utilizing Eq. ( 1) for "smaller-the-better" datasets and Eq. ( 2) for datasets that use the "larger-the-better" concept.After normalization, the Grey Relation Coefficient was determined using Eq. ( 3), serving as an expression of the relationship between the ideal and normalized values.Subsequently, the grey relational grade, a weighted sum of the grey relational coefficients, was calculated through Eq. ( 4).Finally, the concrete mixtures were ranked based on their respective grey relational grades, facilitating a comprehensive assessment of their optimized performance.(1) where y αβ are original data.
where, γ (x 0β, x αβ ) is the grey relation coefficient between x αβ and x 0β , The distinguishing coefficient, denoted as ξ, is a parameter with values ranging between 0 and 1.Based on past literature, it was taken as 0.5 [35][36][37].
The weight assigned to response β is represented by w β and is typically based on the preferences of the decision-maker.For this study, an equal weight was assigned to all factors.

Workability
The workability of the concrete mixes is illustrated in Fig. 4. The data indicates a clear reduction in slump with the inclusion of carpet fibres.Similarly, Tran et al. [38] also observed around 23.5-58.8% reduction in flowability when adding different textile fibres to cementitious composites.This reduction in workability exhibits a direct correlation with the increase in fibre dosage.Specifically, increasing the fibre dosage from 0.3 % to 0.5 % results in a significant decrease in workability ranging from 30.43 % to 34.78 %, 30.43-43.48 %, 21.74-47.83%, and 39.13-52.17% for hydrophilic fibres PTT T1, PTT T2, Nylon T1, and Nylon T2, respectively.Meantime, hydrophobic Polyester, and PP fibres show a reduction in workability of 39.13-47.83% and 30.43-47.83%, respectively, when increasing the fibre volume fraction from 0.3 % to 0.5 %.These results do not exhibit any significant relationship with the hydrophilic and hydrophobic nature of the fibres.Moreover, Mohammadhosseini et al. [39] observed that increasing PP fibre volume from 0.25 % to 1.25 % resulted in reduced workability of the concrete from 38.1 % to 88 %.Furthermore, Nylon Thickness 1 fibres have the least impact on the reduction of workability at a 0.3 % fibre volume compared to other fibres in the Fig. 4. Variation of workability after the inclusion of selected carpet fibres and volumes.

Compressive strength
The compressive strength development of the carpet fibre-reinforced concrete is illustrated in Fig. 5 for 7, 28, and 90 days.The results indicate that, at 28 and 90 days, the compressive strength of the control surpasses that of all fibre-reinforced concretes.However, at 7 days, Nylon T1 and PTT T1 show a slight elevation in compressive strength compared to the control with an increment of 3.43 % and 3.17 %, respectively.Additionally, concrete mixtures with 0.3 % fibre volume demonstrate higher compressive strength than those with 0.5 % fibre volume at 7, 28 and 90 days.This observation shows that an increased fibre volume fraction tends to decrease the compressive strength of the concrete.This phenomenon can be ascribed to the ability of 0.3 % fibre dosage to absorb energy induced by tensile stress without compromising the dense nature of the concrete matrix to a greater degree than the 0.5 % fibre dosage, where the additional fibres have a more pronounced impact on the matrix's integrity, which reduce the compressive strength.This behaviour was observed in past studies which used different types of textile fibres [22,28,40,41].Among the fibre-reinforced concretes, Nylon T1 with 0.3 % fibre volume exhibits the highest compressive strength, reaching 46.69 MPa at 28 days, which is only 4.52 % less than that of the control.Significantly, Nylon T1 with 0.5 % fibre volume and PP with 0.3 % fibre volume rank as the second and third highest in compressive strength at 28 days, with reductions of 4.52 % and 12.15 %, respectively, compared to the control.Conversely, Nylon T2 with 0.5 % fibre volume displays the lowest compressive strength at 28 days, recording a strength of 34.30MPa, which is 29.86 % less than the control.None of the fibre-reinforced concrete types exceed the 90-day compressive strength of the control.However, Nylon T1 and T2 with 0.3 % fibre volume show compressive strengths at 90 days close to that of the control, with a reduction of 2.82 % and 4.26 %, respectively.

Flexural strength
The flexural strength development at 7, 28, and 90 days is illustrated in Fig. 6.Generally, the inclusion of carpet fibres in the concrete enhances its flexural strength after curing for 28 days.The results indicate that a 0.5 % dosage of fibres improves the flexural strength of concrete more than the 0.3 % fibre dosage.Conversely, Polyester fibre-reinforced concrete exhibits a minor decrease in flexural strength when increasing the dosage from 0.3 % to 0.5 %.Specifically, PTT T1 with a 0.5 % fibre volume in concrete presents the highest flexural strength, being 11.8 % higher than the control (6.26 MPa) at 28 days.Moreover, PP 0.5 % and PP 0.3 % fibre-reinforced concrete show the second and third highest flexural strengths, measuring 6.2 MPa and 6.11 MPa, respectively.These values are 10.7 % and 9.11 % higher than the control.Furthermore, PP fibre-reinforced concrete exhibits the highest strength gain from 7 to 28 days (PP 0.3 % -1.16 MPa & PP 0.5 % -1.6 MPa), more than doubling the strength gain observed in the control during the same period.This behaviour was observed in studies by X. Wu et al. [32] and Awal and Mohammadhosseini et al. [41].However, other studies showed that excessive fibre dosage (beyond about 0.5 %) may lead to a reduction in the flexural strength of the concrete [23,24,27,42].

Splitting tensile strength
The splitting tensile strength development of carpet fibre-reinforced concrete was investigated at 7, 28, and 90 days, as illustrated in Fig. 7.The findings indicate that the incorporation of fibres enhances the splitting tensile strength of the concrete, regardless of fibre type or volume fraction.Zareei et al. [23] also observed that inclusion of 1 % carpet fibres enhances the splitting tensile strength of the concrete by 9 %.In addition, Mohammadhosseiniet al. [43] and Awal et al. [28] observed that higher amounts of carpet fibres enhance the splitting tensile strength of the concrete.Significantly, Polyester 0.5 % fibre concrete exhibited a slightly different behaviour, showing a marginal reduction in splitting strength with a 0.8 % strength decrease compared to the control.The inclusion of 0.5 % fibre dosage resulted in increased splitting tensile strength compared to the 0.3 % fibre volume fraction at 28 days.On the other hand, PTT T2 and Polyester fibre-reinforced concrete exhibited adverse behaviour.PP 0.5 % fibre-reinforced concrete displayed the highest splitting tensile strength at 28 days, reaching 3.53 MPa, marking a notable 38.98 % improvement over the control.In addition, Nylon T1 0.5 % and Nylon T1 0.3 % fibre-reinforced concrete demonstrated the second and third highest splitting tensile strengths, measuring 3.34 MPa and 3.19 MPa.These strengths represented enhancements of 31.5 % and 25.59 % over the control.Examining strength gain, PP 0.5 % exhibited the highest increase from 7 days to 28 days curing, marking a gain of 1.25 MPa.This figure exceeds the control's strength gain by more than seven times, as the control showed a gain of 0.15 MPa during the same period.These results highlight the positive impact of fibre reinforcement on splitting tensile strength, with PP 0.5 % standing out as a particularly effective enhancer in the tested concrete mixtures.

Total shrinkage
Fig. 8 illustrates the total shrinkage observed within the initial 90 days for the carpet fibre-reinforced concrete specimens.The test results clearly indicate that the inclusion of all types of carpet fibres reduces concrete shrinkage compared to normal concrete.Significantly, when considering fibre dosage, concrete with 0.3 % fibre content displays enhanced performance concerning shrinkage in comparison to 0.5 % fibre volume.For instance, in Nylon T1 fibre-reinforced concrete mixtures, the 0.3 % fibre dosage exhibits a shrinkage rate of 0.164 % at 90 days, while the 0.5 % fibre concrete shows a higher shrinkage rate of 0.18 %.An 8.9 % increase in the shrinkage rate for 0.5 % Nylon T1 fibrereinforced concrete compared to the 0.3 % fibre in Nylon T1.Furthermore, the lowest shrinkage is observed in Nylon T1 0.3 % fibrereinforced concrete, recording a shrinkage rate of 0.164 % at 90 days, representing a significant 22.3 % reduction compared to the control shrinkage rate at the same time.Among the various types of carpet fibrereinforced concrete mixtures, PP 0.3 %, and PTT T1 0.3 % fibrereinforced concrete exhibit the second and third lowest shrinkage rates at 0.17 % and 0.173 % at 90 days.These values are 19.43 % and 18.01 % lower than the control.However, PTT T2 fibre-reinforced concrete has the highest shrinkage rates among all types, measuring 0.191 % and 0.205 % for 0.3 % and 0.5 % fibre-reinforced concrete, Fig. 6.Flexural strength at 7, 28 and 90 days for the concrete mixtures with: (a) 0.3 %, (b) 0.5 % fibre volume.
respectively.These values are still 0.09 % and 0.03 % less than the control at 90 days.Wang et al. [21] also observed that the use of recycled PP carpet fibres minimizes shrinkage by about 15 % -30 % compared to the control at 21 days.In addition, Xiaoxin et al. [32] reported that 19 mm-length recycled PP carpet fibre had 30.2 % and 9.72 % lower shrinkage than that of control samples at 10 and 150 days, respectively.

Optimization of fibre reinforced concrete
Table 2 presents the grey relation coefficients and rankings of each fibre-reinforced concrete mixture, based on their performance at 28 days using the GRA method.The analysis considers all the factors such as compressive strength splitting tensile strength, flexural strength, and total shrinkage of the concrete mixes.The findings from the GRA method indicate that, overall, the 0.3 % fibre-reinforced concrete perform better than the 0.5 % fibre volume concrete, except for PP fibre-reinforced concrete.Specifically, the results highlight that Nylon T1 fibrereinforced concrete demonstrates superior performance compared to other fibre-reinforced concrete types.

Discussion
The compressive strength, splitting tensile strength, flexural strength, and total shrinkage results, shows that the optimal performance for Nylon, Polyester, and PTT carpet fibre-reinforced concrete is achieved at a 0.3 % fibre dosage with a fibre length of 12 mm.In the case of PP fibre mix, an overall superior performance is observed at a 0.5 % dosage compared to a 0.3 % fibre dosage.However, compressive strength and shrinkage results in PP fibre concrete shows that the 0.3 % fibre dosage has better outcomes than the 0.5 % dosage.To understand the underlying mechanisms influencing their performance, the microstructure, pore structure, and properties of the interfacial zone in these optimized mixtures were observed.

Correlations of fibre characteristics with the performance of concrete
The findings from the investigation observes that tensile strength of the fibres act as a main element for the performance of the concrete.The incorporation of fibres into the concrete mixture enables the absorption of energy through fibre bridging action, directly enhancing both splitting tensile strength and flexural strength.Significantly, fibres with high tensile strength exhibit a superior capacity to absorb energy, compared to others with low tensile strength.Test results illustrate that Nylon T1, PP, and PTT T1 fibres, characterized by high tensile strength, significantly enhance flexural strength and splitting tensile strength.Moreover, higher fibre dosage increases the energy absorption capacity which tends to enhance the splitting tensile and flexural strength of the concrete.Generally, the concrete matrix exhibits low tensile strength, and the high tensile properties of the fibres play a crucial role in withstanding stress at shear planes, thereby enhancing the overall concrete properties.Zhu et al. [44] also observed this effect in their study.Additionally, the elastic properties of fibres also directly influence on concrete performance.Fibres with a high Young's modulus possess the ability to mitigate concrete shrinkage by distributing stress, thereby minimizing crack propagation and overall shrinkage.Shrinkage test results demonstrate that Nylon T1 and PP fibres, characterized by a high Young's modulus, reduce concrete shrinkage more effectively compared to fibres with lower elasticity.This behaviour matches with similar observations made by Wang et al. [45] in their research.
The influence of the hydrophilic and hydrophobic characteristics of fibres on concrete properties is found to be comparatively insignificant when compared with the effects cause from the mechanical properties of the fibres.Polyester fibres, characterized by hydrophobic behaviour, exhibit low performance in comparison to other types of fibre-reinforced concrete mixes.In contrast, PP fibres, which possess hydrophobic behaviour with high Young's modulus and tensile strength properties, demonstrate superior performance in flexural strength, splitting tensile strength, and shrinkage reduction when compared to Nylon T2 and PTT T2 fibre concrete, which exhibit hydrophilic behaviour.This variability highlights that the mechanical properties of the fibres exert more influence on concrete performance than the hydrophilic or hydrophobic behaviour of the fibres.This highlights the critical role played by the mechanical characteristics of the fibres, namely tensile strength, and Young's modulus, in determining the overall effectiveness of fibrereinforced concrete.While the hydrophilic or hydrophobic nature of fibres may contribute to certain aspects of performance, the dominant influence is attributed to the mechanical characteristics of the fibres.Therefore, when optimizing the formulation of fibre-reinforced concrete mixes, a focus on enhancing mechanical properties is more effective to achieve enhanced splitting tensile strength, flexural strength, and shrinkage reduction.Furthermore, the noncircular geometry of fibres plays a significant impact on reducing the workability, as these fibres act as three-dimensional barriers obstructing the gravitational movement of free water, thereby densifying the concrete mixture.This phenomenon was observed by the past research conducted by Tran et al. [3], and his study highlighted the challenge posed by noncircular fibre shapes to the fluidity of concrete.Additionally, Nylon T1 and PP fibre bundles have knots at specific locations within the bundle while other fibre bundles display a twisted behaviour (Fig. 1).Fibres with knots exhibit a better dispersion within the concrete matrix compared to their twisted fibre bundles.The knotted fibres' ability to uniformly disperse throughout the concrete proves to be a contributing factor to their enhanced performance, thus highlighting the significance of fibre geometry in optimizing concrete properties.Moreover, the fibre dimensions play a crucial role in influencing the pore formation within concrete structures.The micro-sized fibres can subdivide larger macro-sized pores into smaller entities, as observed by various past studies [32,46,47].Additionally, the presence of these flexible fibres may interrupt moisture transfer from the concrete to the external environment, which helps minimize shrinkage [40].

Pore distribution and microstructure
The investigation of pore characteristics in optimal concrete mixtures was conducted through analysis of X-ray micro-CT data.Fig. 9 represents the porosity (open and closed) for each mix, while Table 3 presents the pore diameter distribution of concrete mixtures at 7 and 28 days.The inclusion of fibres demonstrated a reduction in total porosity compared to the control, at both 7 and 28 days.Furthermore, an overall reduction in total porosity was observed with extended curing times.Significantly, Nylon T1 exhibited the lowest porosity among fibrereinforced concretes, demonstrating an impressive reduction in total porosity, ranging from 76.6 % to 81.12 % less than the control after 7 and 28 days.Additionally, the 0.3 % PP mix demonstrated a reduction in total porosity compared to the control (Table 3).Among all 0.3 % fibre dosage concrete mixtures, Polyester fibre-reinforced concrete exhibited the highest porosity, yet still showed a reduction of 45.3-20.5 % compared to the control at 7 and 28 days.Significantly, when considering the fibre dosage in PP 0.3 % and 0.5 % concrete mixtures, the total porosity results show that the addition of a higher fibre dosage actually increased the total porosity by 68.11-66.17% at 7 and 28 days.However, still 0.5 % fibre dosage shows less porosity than the control.This behaviour was observed in studies conducted by Zareei et al. [23] and Meddah et al. [48].It shows that a higher fibre volume tends to form a greater number of voids due to the low workability, which in turn reduces the compactness of the concrete.In addition to the densification observed in the concrete mixtures, the reduction in average pore diameter, pore refining effect, can be observed from the pore distribution data (Table 3).Xiaoxin et al. [32] also observed that the inclusion of waste fibres tends to subdivide macropores into mesopores, which helps minimize cracks in the concrete.Significantly, Nylon T1 shows the highest reduction in pore diameter among the varied fibre-reinforced concretes, showing average pore diameters of 0.12 mm and 0.108 mm at 7 and 28 days, respectively.On the other hand, concrete compositions incorporating Polyester and Nylon T2 fibres exhibit higher average pore diameters compared to other fibre-reinforced concretes.Specifically, at 7 and 28 days, the average pore diameter for Polyester fibre-reinforced concrete ranges from 1.08 mm to 0.87 mm, while Nylon T2 fibre-reinforced concrete shows a range of 1.02 mm to 0.77 mm.
In the examination of open and closed porosity, Fig. 9 illustrate that the inclusion of fibres tends to enhance the percentage of closed porosity than open porosity.Open pores provide an important role to increase the drying shrinkage of concrete by facilitating the movement of water, leading to higher evaporation.Therefore, the reduction in open pores and the increase in closed pores contribute to the overall improvement in the shrinkage reduction performance of fibre-reinforced concrete.Nylon T1 and PP 0.3 % fibre-reinforced concrete mixes exhibit the densest pore structures, characterized by lower porosity, higher closed pore percentages, and a more pronounced pore refining effect when compared to other concrete mixtures at both 7 and 28 days (Fig. 9 and Table 3).Furthermore, the PP 0.5 % concrete mixture, displays a lower percentage of closed pores compared to PP 0.3 %.Specifically, at 7 days, the 0.5 % PP fibre-reinforced concrete exhibits 40.1 % closed pores, while the 0.3 % PP showed a higher 80.3 % closed pore percentage.This variation in closed pores suggests that the 0.3 % PP fibre-reinforced concrete may exhibit superior performance in shrinkage reduction compared to the 0.5 % fibre dosage mix.This pore refining effect and the reduction in the porosity has been highlighted in several studies that used other types of fibres such as, PP, polyacrylonitrile and PTT [3,49,50].
In addition to illustrating the pore structure, Fig. 10(a) reveals the presence of micro gaps adjacent to hydrophobic fibres.These create a loose interface and bonding between the fibre and matrix.These gaps have the potential to diminish the strength of the concrete, and these weak bonds can cause fibre pullouts, negatively affecting crack propagation control [24,40].However, it is worth noting that some of these voids may have originated from inadequate bonding of the matrix with the resin.As a result, this lack of proper bonding could have led to fibre movement during the polishing process, contributing to the formation of certain voids.Furthermore, Fig. 10(c) illustrates strong bonding between the fibre and matrix in hydrophilic fibres.The hydrophilic nature of certain fibres can contribute positively to concrete performance, particularly in terms of crack control and hydration.These fibres can absorb and retain water, which facilitates internal curing [20].Additionally, the trilobal cross-sectional geometry of carpet fibres contributes to enhanced physical interlocking, thereby increasing the stress-induced energy of the fibres.Furthermore, it is observed that the higher fibre dosage leads to fibre agglomeration.Agglomeration tends to disturb the compactness of the concrete, creating larger voids, which negatively impact the shrinkage performance and mechanical properties of the concrete.This has been observed in a number of previous studies [23,27,40,48].Formation of large voids was observed in the PP 0.5 % concrete mix.(Fig. 10(b)).Additionally, the entanglement of fibres further contributes to the formation of large voids, weakening the connection between the fibres and the cement matrix.Micro cracks which were initiated from the fibres were observed in the areas containing bundles of fibres (Fig. 10(d) & Fig. 10(f)).Furthermore, fibre bridging action was observed in SEM images (Fig. 10(e)) that make the concrete stiffer and increase the strength of the fibre reinforced concrete.This was observed in all the types of concrete mixtures.Fibre bridging enhances the mechanical properties of concrete as these fibres absorb the stress placed on the matrix during the hydration process.When a crack occurs, the randomly distributed fibres bridge the crack and transmit the tension concentrated at the crack tip.This helps to minimise both the crack width and the number of cracks [32,40,42].

ITZ zone variations and the relationship of it for the performance
Nanoindentation data analysis offers valuable insight into the thickness of the interfacial transition zone (ITZ).The suitable locations for conducting the indentation process were determined using the SEM images.In Fig. 11, the mean value distribution of the elastic modulus (E) and hardness (H) for various optimized concrete samples is presented.Significantly, all samples exhibit similar hardness and elastic values.The elasticity ranges from 1.5 GPa to 35 GPa, while hardness falls within the range 0.    38,46].For example, in 28-day cured Nylon T1 fibre-reinforced concrete, the average H and E values within the ITZ were 43.3 % and 46.93 % lower than the bulk matrix.Similarly, for 28-day cured Polyester fibre-reinforced concrete, these values were 45.4 % and 47 % lower than the bulk matrix.This shows that a wider ITZ area correlated with a higher proportion of voids [38,46], potentially explaining the poorer mechanical performance of hydrophobic Polyester fibre-reinforced concrete compared to hydrophilic Nylon T1 fibre-reinforced concrete.Moreover, a smaller ITZ area, being the weak zone between fibre and matrix, positively affected bonding compared to fibres with thicker ITZ zones.This observation was supported by SEM images and additionally, these images revealed the initiation of several microcracks near the fibres, possibly due to the highly porous nature of the ITZ.Furthermore, increasing the curing age enhanced the H and E values of the matrix for all mixes.Extended curing time (from 7 to 28 days) resulted in a minor reduction in ITZ thickness, potentially reducing the number of voids and enhancing concrete performance at 28 days compared to 7 days.

Conclusion
The key conclusions drawn from this study are as follows: • Carpet fibres slightly reduced compressive strength compared to standard concrete.Nylon fibres with 0.3 % volume gave a 4.52 % decrease in compressive strength at 28 days, the highest strength of the mixtures.This decrease is likely due to poor dispersion and air voids.Increasing to 0.5 % volume further lowered strength, due to reduced workability and compaction.• Carpet fibres markedly improved splitting and flexural strengths, with 0.5 % volume outperforming 0.3 %.PTT T1 fibres at 0.5 % volume excelled, increasing 28-day flexural and splitting tensile strengths by 11.8 % and 20 % respectively.This improvement is due to fibre bridging action, which transfers loads across cracks and absorbs stress from the cement matrix.• All the types of carpet fibres effectively mitigated concrete shrinkage.Nylon T1 fibres at a 0.3 % dosage demonstrated the most significant, achieving a 22.3 % decrease at 90 days when compared to the control.However, increasing the fibre volume from 0.3 % to 0.5 % led to a slight increase in shrinkage, but still maintained a lower shrinkage value than the control specimen.• GRA analysis showed that Nylon T1, with the highest tensile strength (0.84 GPa) and Young's modulus (2.18 GPa), exhibited the overall best performance, indicating that fibres with superior individual characteristics contribute most significantly to the performance.• Carpet fibres reduce total porosity, exhibits a pore-refining effect, and increase closed pores relative to open pores.This decrease in the percentage of open pores assists in minimizing the overall drying shrinkage of the concrete.However, increasing fibre dosage from 0.3 % to 0.5 % raises total and open porosity, reducing shrinkage performance compared to 0.3 % fibre-reinforced concrete.• The reduction in the interfacial transition zone (ITZ) thickness between the carpet fibres and the cement matrix played a pivotal role in enhancing the performance of the fibre-reinforced concrete.Hydrophilic fibres exhibited an ITZ of 10 µm, compared to 15 µm for hydrophobic fibres.
The findings of this study, while promising, underscore several limitations that should be considered before implementing carpet fibre reinforced concrete in commercial applications.Future research should address these limitations by investigating the impact of fibre distribution and the impact of different carpet fibre combinations, which were not examined in the present work.A deeper understanding of using blended carpet fibre combinations will help identify how low-performance fibres can be utilized in concrete alongside high-performance fibres.This approach could potentially allow for the reuse of a larger amount of carpet waste in concrete, even when certain fibre types cannot be used individually.Additionally, evaluating the long-term performance of carpet fibre reinforced concrete is crucial to ensure durability in largescale applications.Researchers should also focus on developing innovative methods to incorporate higher volumes of fibres while maintaining acceptable workability and mechanical properties, as well as improving the cost-effectiveness of the material.Furthermore, formulating predictive models would be beneficial for future applications.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Fig. 1 .
Fig. 1.Physical and Scanning electron microscope images of the carpet fibres utilized in the investigation.

Fig. 9 .
Fig. 9. Percentage of open pores and closed pores in concrete mixtures.

Fig. 10 .
Fig. 10.SEM images of fibres.(a) micro gaps near PP fibre surface; (b) fibre agglomeration in PP 0.5 % mix; (c) strong bonding in Nylon T1 fibres; (d) micro cracks and trilobal cross-sectional of Nylon fibre; (e) fibre bridging in Polyester fibre; and (f) formation of micro cracks in PP fibre reinforced concrete at 28 days.

Fig. 11 .
Fig. 11.The mean distribution of hardness and elastic modulus at ITZ area for (a and b) Nylon T1; (c and d) Nylon T2; (e and f) PTT T1; (g and h) PTT T2; (i and j) Nylon T2; and Polyester (k and l) at 7, and 28 days.

Table 1
Specific characteristics of carpet fibres.

Table 2
Grey relation coefficient and ranking for carpet fibre reinforced concrete mixtures.

Table 3
Quantitative analysis of pore characteristics.