​​Influence of coir and flax fiber lengths on fracture toughness of fly ash, slag, and silica fume-based geopolymer concrete

Abstract

Geopolymer concrete (GC), which is produced from industrial by-products rich in aluminosilicates such as fly ash, ground granulated blast furnace slag, and silica fume, serves as an environmentally sustainable substitute for conventional Portland cement-based concrete. Fracture toughness is vital for GC, as its brittle matrix is prone to crack initiation and propagation, affecting structural safety. Enhancing fracture resistance ensures reliable performance under various loading modes. Coir and flax fibers were chosen for their availability, sustainability, and ability to bridge cracks and improve post-crack energy absorption, providing insights for optimizing natural fiber reinforcement. Researchers have turned to natural fibers such as coir and flax to overcome the limited ductility and mechanical shortcomings typical of geopolymer composites. These fibers were specifically chosen for their potential to enhance both toughness and post-cracking energy absorption in the matrix. This investigation focused on how different aspect ratios of coir and flax fibers affected the fracture properties of GC when exposed to various loading modes: Mode I, Mode III, and their combination. To ensure a controlled comparison, GC specimens were cast with fiber lengths set at 20 mm, 40 mm, and 60 mm, while maintaining a fixed fiber volume fraction of 0.5%. Additionally, the microstructure of the geopolymer composite was characterized using scanning electron microscopy, X-ray Diffraction, thermogravimetric analysis, and fourier transform infrared spectroscopy analyses. The results demonstrated that the incorporation of 40 mm coir and flax fibres enhanced the fracture toughness of the GC by up to 20.65% under mixed-mode loading (loading angle = 20°), 18.96% under Mode I, and 9.70% under Mode III. In contrast, extending the fibre length to 60 mm led to a deterioration in performance, with FRTS values falling by as much as 2.64% below those of the control composite, a decline attributable to fibre agglomeration and the consequent disruption of effective stress transfer. Microstructural analyses reveal that a dense, continuous sodium aluminosilicate hydrate gel network augmented by residual crystalline phases such as quartz and mullite substantially reinforces the geopolymer matrix. Coir fibres exhibit superior interfacial bonding and more stable debonding characteristics than flax fibres, thereby promoting more effective crack bridging and yielding greater fracture toughness. Complementary thermal and chemical characterisation further indicates enhanced matrix stability, reduced porosity, and improved load-transfer efficiency, with partial carbonation imparting additional microstructural densification.

Introduction

The construction industry is experiencing a major shift as it increasingly seeks eco-friendly alternatives to conventional cement-based concrete^1,2. The Portland cement production is believed to contribute to approximately 8% of the world’s carbon dioxide emissions, exerting a significant environmental burden on the Earth^3,4. In this scenario, geopolymer concrete (GC), which is made from industrial waste derivatives: silica fume (SF), ground granulated blast furnace slag (GGBS), and fly ash (FA)⁵, metakaolin^6,7 has become a promising and environmentally friendly alternative⁸. GC offers numerous advantages, such as reduced carbon emissions, enhanced durability, and excellent resistance to chemical degradation. Nonetheless, GC inherently possesses mesoscale microcracks that can expand and coalesce into larger macrocracks when subjected to external forces, affecting both its structural integrity and lifespan. Fracture toughness (FRTS) evaluates this phenomenon by measuring the peak stress region at the crack tip and forecasting when the crack will begin to propagate uncontrollably⁹. Material fracture can occur under Mode I (tensile)¹⁰, Mode II (in-plane shear)¹¹, Mode III (out-of-plane shear)¹², or a mixed-modes¹³. In practical scenarios, cracks typically form under mixed-mode conditions, which vary with the specimen geometry, the patterns of loading, and the direction of pre-existing flaws. This often results in failure through either individual or combined fracture modes, highlighting the intricate interactions of internal stresses¹⁴. Extensive experimental configurations have been developed to characterize the fracture response of concrete and other brittle substances. Among these, common approaches include three-point bending¹⁵, semi-circular bend tests¹⁶, edge-notched disc bending (ENDB)¹⁷, split-tension cube¹⁸ and wedge-splitting¹⁹. Each method offers distinct perspectives on how cracks begin, spread, and the material’s ability to withstand fractures under different loading scenarios. Nonetheless, the material’s natural brittleness restricts its structural applications, particularly in situations where crack development and fracture behavior are critical²⁰. To address this limitation, researchers have explored the incorporation of natural and other type of fibers; such as hemp, sisal, jute, coconut, bamboo, flax²¹, steel²², polyvinyl alcohol^23,24 to boost the mechanical properties FRTS of GC.

These fibers are widely studied for their sustainability, availability, and mechanical enhancement of concrete^21,25.

Numerous studies have explored how both natural and hybrid fibers affect the mechanical properties and rheological behavior of GC, underscoring the significance of fiber type, amount, and structure in defining the material’s performance. For instance, Wongsa et al.²⁶ discovered that incorporating 0.5% of either sisal or coconut fibers optimized the compressive strength of high-calcium FA geopolymer mortars, while increasing the fiber content led to a decline in performance. In a related study, Korniejenko et al.²⁷ observed increases in compressive strength of 2%, 15%, and 26% for fibers made from 1% sisal, coir, and cotton, respectively. However, raffia fibers did not perform as well owing to inadequate fiber-matrix interface bonding, despite having similar geometry and a higher elastic modulus. Abbas et al.²⁸ found that the addition of kenaf fibers to GC significantly decreased workability, with the slump reducing as both fiber length and content increased. The workability dropped by 70–85% compared to the control mix across different lengths and volume fractions. Similarly, Mohammed et al.²⁹ observed that GC reinforced with 35 mm jute fibers at a 0.70% volume fraction had a minimum slump of 28 mm while achieving a maximum 28-day compressive strength of 31.5 MPa. Bazarkhankyzy³⁰ found that flax fibers measuring 20 mm yielded the highest compressive strength in GC, reaching 45.87 MPa, an increase of 6.5%. Sisal, jute, and coir fibers exhibited a clear trend of decreasing composite strength as the fiber length was increased from 20 mm to 60 mm. Coir fibers significantly enhanced impact resistance, although longer fibers compromised workability. The best results were observed with 40 mm flax, jute, and sisal fibers, while 60 mm sisal fibers maximized failure impact but slightly reduced first-crack resistance. Ductile fibers like coir and jute facilitated gradual failure, whereas stiffer fibers such as flax and sisal induced greater crack widths and a loss of toughness. In a similar vein, Guruswamy et al.³¹ utilized the coir fibers with a length of 1.0 cm and a 4% volume fraction resulted in a maximum compressive strength of 34 MPa, indicating a notable improvement in the mechanical performance of the concrete matrix. However, as the fiber volume fraction increased from 4% to 12%, a gradual reduction in compressive strength was noted, primarily due to the degradation of the matrix-aggregate interfacial bond resulting from the excessive moisture absorption characteristics of the coir fibers. Firas et al.³² noted that kenaf fiber-reinforced GC exhibited compressive strengths that were 2.4% and 5% higher than those of coconut fiber-reinforced GC at 7 and 28 days, respectively. Additionally, kenaf fibers enhanced thermal performance by reducing thermal conductivity by 7.1% compared to fiber-free GC, surpassing coconut fiber-reinforced specimens, which showed a 3.8% reduction. Expanding on the advantages of hybridization, Rajendran et al.³³ observed that incorporating both coconut coir and Polyvinyl alcohol fibers in GC resulted in a 60% increase in impact resistance and promoted ductile deformation behavior. These results highlight the importance of optimizing fiber type, length, and content to improve both the mechanical properties and workability of GC.

The FRTS of GC that includes natural fibers is not well understood and requires focused investigation, unlike other types of fibers that have been thoroughly studied in the existing literature. For example, Zargaleh et al.³⁴ discovered that substituting 30% of the binder with GGBS greatly enhanced the fracture performance of GCs. This was evidenced by a 284% increase in critical fracture energy and a rise in FRTS from 16.73 to 27.49 MPa·√mm, indicating improved resistance to the initiation and spread of cracks. In a similar vein, Wang et al.³⁵ found that FA-based GC containing 0.05% basalt fibers reached a maximum FRTS of 0.81 MPa.m^0.5 and a fracture energy of 86.43 N/m, suggesting an optimal fiber content for enhancing fracture properties. Furthermore, the addition of 1.5% steel fibers to heavyweight GC significantly boosted fracture performance, with FRTS nearly tripling, initial fracture energy increasing by a factor of eight, and total fracture energy rising approximately 95 times compared to specimens without reinforcement³⁶. Moreover, Li et al.³⁷ found that raising the fiber dosage from 1% to 2% led to a 70% increase in the fracture energies of polyethylene fiber-reinforced GC, compared to a 45% increase with steel fibers, attributed to the higher aspect ratio of polyethylene. Nonetheless, for both fiber types, the toughening effect decreases at higher concentrations due to fiber agglomeration and uneven distribution, with polyethylene fibers being more susceptible to clustering beyond 2%. Furthermore, microsilica enhances fracture energy by reinforcing the fiber–matrix bond and reducing the adverse effects of excessive fiber addition³⁸. Conversely, Samadi et al.¹³ detected that the effective FRTS of preplaced aggregate GC reduced from 0.86 MPa.m^0.5 at a mixed-mode angle of γ = 50° to 0.79 MPa.m^0.5 at γ = 20° for the specimens, signifying a reduction in resistance to fracture as the influence of mixed-mode I/III loading conditions increased. Incorporating reinforcement meshes significantly boosted effective FRTS, with the M5-type steel wire mesh achieving toughness enhancements of 9.09% and 20.10% for specimens measuring 75 mm and 150 mm in diameter, respectively, at γ = 20°. These improvements exceeded those achieved with the M40-type mesh, which resulted in respective increases of 4.61% and 18.86%. In their study, Karthik et al.³⁹ found that GC has a lower FRTS in pure mode III (0.50-1.0 MPa.m^0.5) compared to mode I (1.40–2.19 MPa.m^0.5). The addition of short steel fibers increased mode III toughness by 67.05%, while long fibers resulted in an even more significant improvement of 102.35%. Although there has been extensive research on the fracture behavior of GC with various synthetic and industrial fibers, there are still considerable gaps in understanding the performance of GC when natural fibers are used.

Although numerous investigations have explored the influence of natural fibres on the mechanical and rheological responses of GC, the existing literature exhibits considerable inconsistencies regarding the optimal fibre type, length, and dosage, alongside divergent findings on workability, interfacial bonding, and strength enhancement. More importantly, despite extensive research on fracture toughness using synthetic and industrial fibres, the fracture behaviour of natural fibre-reinforced GC particularly under realistic mixed-mode loading conditions remains inadequately characterised. There is still limited consensus on how mesoscale microcracks interact with the distinct morphologies of natural fibres to govern crack initiation, energy absorption, and post-peak mechanical response. These knowledge gaps underscore the necessity for a focused, mechanistic investigation into the ways in which specific natural fibres modify the fracture response of GC. Accordingly, the present study aims to elucidate the fracture mechanisms of natural fibre-reinforced GC by evaluating their influence on FRTS, crack-propagation characteristics under controlled loading conditions. The novelty of this work lies in its systematic interrogation of natural fibres within a fracture-mechanics framework an area in which current evidence remains sparse thereby offering new insights into the optimisation of fibre-reinforced GC for enhanced toughness and structural performance.

Experimental design

Materials

• The binders adopted for this research: Class F-FA, GGBS and SF sourced from Kazakhstan. Figure 1a; Table 1 illustrate the visual representation and detailed oxide compositions of FA, GGBS, and SF. FA is notable for its higher Al₂O₃ (28.94%) and SiO₂ (61.23%) and content, which contributes to its strong pozzolanic reactivity, making it ideal for geopolymer synthesis. In contrast, GGBS is rich in CaO (38.41%), promoting early strength development, although its SiO₂ (34.48%) and Al₂O₃ (11.81%) levels are lower compared to FA. SF, with a significantly high SiO₂ content (96.3%), shows exceptional reactivity and improves the microstructural densification of the geopolymer matrix through secondary pozzolanic interactions. The particle size distribution of the binder materials are depicted in Fig. 1b.

• Natural river sand, used as the fine aggregate, was characterized by a specific gravity (2.60) and a fineness modulus (2.71), meeting the grading and quality standards outlined in IS: 383-2016⁴⁰. The coarse aggregate in the mix design consisted of crushed granite, exhibiting a water absorption rate of 0.6%, a specific gravity of 2.66 and maximum nominal particle size of 12.5 mm and the other properties meets the requirements of ASTM-C131⁴¹, BIS: 812-112⁴².

• The activation of the pozzolanic binder system was accomplished using a combined alkaline solution consisting of 12 M sodium hydroxide (NaOH) and sodium silicate (Na₂SiO₃) at a mass ratio of 1.45. To prepare the NaOH solution, analytical-grade pellets were dissolved in distilled water, and then Na₂SiO₃ was introduced gradually to reach the desired composition. The resulting activator solution was then conditioned for 24 h before mixing to achieve thermal and chemical balance, ensuring consistent reactivity and promoting optimal geopolymerization during the casting process.

• In this study, coir and flax were introduced to the GC mixtures in lengths of 20, 40, and 60 mm, while keeping a constant volumetric fraction of 0.5% relative to the GC matrix. Figure 1a provides a schematic illustration of these fibers. The tensile strengths of the coir and flax fibers were 250 MPa and 1000 MPa, respectively, with diameters of 500 μm and 21 μm, underscoring their unique mechanical properties and potential effectiveness as reinforcing elements within the GC matrix.

Fig. 1.

Materials used for GC.

Table 1.

Chemical properties of materials³⁰.

Composition	MgO	Al2O3	Na2O	K2O	CaO	MnO	SiO2	P2O5	FeO	TiO2	SO3
FA (%)	0.45	28.94	0.41	0.74	1.28	0.15	61.23	0.54	5.18	1.09	–
GGBS (%)	9.98	11.81	0.65	0.84	38.41	0.27	34.48	–	0.47	1.45	1.66
SF (%)	0.63	0.57	0.55	1.17	0.24		96.3	–	0.54	–	–

SEM analysis of binders

The SEM image of FA primarily reveals smooth, spherical cenospheres that are produced during the high-temperature combustion of coal (Fig. 2a). Additionally, it shows some irregular porous particles originating from unburned carbon and mineral fragments, which could negatively affect the workability of concrete. The particle size varies from several micrometers to approximately 100 μm, emphasizing its dual function as both a microfiller and a pozzolanic material. Smaller particles clinging to larger ones enhance packing density and facilitate the “ball-bearing” effect. GGBS predominantly consists of angular, irregular particles with rough exteriors and dense, low-porosity cores, characteristic of materials that have been mechanically ground (Fig. 2b). Particle sizes vary from approximately 1 to 100 μm, with smaller particles boosting reactivity and the development of strength. In contrast to the smooth spherical shapes of FA, GGFS features sharp-edged forms that influence both hydration and workability. The smooth, non-crystalline surface indicates its potential for hydraulic and pozzolanic activity when used in mixed cement systems. SF is characterized by tightly packed, extremely fine spherical particles, primarily measuring less than 1 μm (Fig. 2c). These nanospheres, which are a by-product of silicon alloy manufacturing, possess a high specific surface area due to their strong tendency to agglomerate, which enhances their pozzolanic activity. Their amorphous, glass-like structure increases reactivity with calcium hydroxide, resulting in enhanced matrix density and lower permeability.

Fig. 2.

SEM analysis of materials (a) FA, (b) GGBS and (c) SF.

SEM analysis of fibers

The SEM images of coir fiber display a coarse, uneven surface characterized by distinct fibrillar structures and layered textures, indicative of its multicellular natural makeup (Fig. 3a). The presence of visible lumens and pores suggests a porous, low-density structure that enhances energy absorption and contributes to the creation of lightweight composites. The surface’s protrusions, pits, and longitudinal grooves facilitate strong mechanical interlocking and interfacial bonding within cementitious or polymeric matrices. SEM images of flax fiber display parallel, rod-like fibrillar bundles characterized by longitudinal striations and grooves created by cellulose microfibrils (Fig. 3b). Minor delamination of fibrils results in localized weak points, while the generally smooth yet slightly rough surface enhances matrix adhesion. This organized, unidirectional configuration contributes to flax fiber’s high strength-to-weight ratio and its effectiveness as a sustainable reinforcement in composites.

Fig. 3.

SEM analysis of fibers (a) Coir, (b) Flax.

XRD analysis of materials

The XRD examination of GGBS reveals peaks associated with quartz, calcite, rankinite, gehlenit, and magnetite, suggesting the presence of minor crystalline inclusions within the slag matrix (Fig. 4a). The emergence of additional peaks linked to calcium silicate and calcium carbide indicates the existence of calcium-rich reactive phases that play a crucial role in the material’s latent hydraulic and pozzolanic properties. The XRD pattern of FA shows significant reflections for quartz, along with clear peaks for mullite, hematite, and magnetite, indicating the presence of crystalline aluminosilicate and iron oxide phases (Fig. 4b). The increased background intensity suggests a significant presence of amorphous material, which is crucial for the processes of geopolymerisation and alkali activation. The distinct reflections of quartz and mullite reveal their relatively low reactivity, while the amorphous aluminosilicate component mainly facilitates the formation of geopolymeric gel. The XRD pattern of SF is marked by broad, low-intensity humps, indicating its mainly amorphous and highly reactive characteristics. Small peaks associated with quartz, opal, and cristobalite suggest the presence of trace amounts of crystalline or semi-crystalline silica phases (Fig. 4c). The absence of sharp diffraction peaks indicates the glassy nature of SF, making it highly appropriate for alkali activation and the formation of geopolymeric gels.

Fig. 4.

XRD analysis (a) GGBS, (b) FA and (c) SF.

Mixing combination and specimen preparation

In this study, seven GC formulations were created (Table 2), incorporating natural fibers such as coir and flax at a constant volumetric fraction of 0.5% in relation to the binder content. Each type of fiber was used in lengths: 20 mm, 40 mm, and 60 mm. A basic no fiber GC mixture served as the control. Mixtures reinforced with coir fibers were labeled as GC-CF20, GC-CF40, and GC-CF60, corresponding to the different fiber lengths. Similarly, flax fiber-reinforced mixtures were named in the same manner, with numerical suffixes indicating the length of the fibers used. A cylindrical mold with a diameter of 150 mm and a thickness of 40 mm was utilized to prepare specimens for the fracture evaluation test (Fig. 5a). The GC mixtures, which include FA, GGBS, SF, and natural fibers, were prepared following a meticulously controlled sequence to ensure even fiber distribution and enhance mechanical properties. The alkaline activator was created by mixing NaOH solution with Na₂SiO₃ in a mass ratio of 1:1.45 and was then set aside for later use. Initially, the dry pozzolanic materials FA, GGBS, and SF were placed in a pan-type mixer and blended for about three minutes to achieve uniformity. Afterward, the fine and coarse aggregates were added and mixed for an extra 3 min to ensure they were evenly distributed within the binder matrix. Natural fibers were then gradually introduced to the dry mixture, with careful attention to avoid clumping, and mixing continued until they were uniformly dispersed (Fig. 5b). The dry composite was slowly combined with the pre-prepared alkaline solution while being continuously stirred for about 2 min, allowing the creation of a geopolymer matrix with adequate cohesiveness and workability (Fig. 5c). The newly prepared GC was poured into molds and vibrated to remove any trapped air. A notch-forming plate was inserted into the upper surface of the fresh specimen and withdrawn after approximately 2 h, thereby producing a controlled pre-notch in the disc samples. The resulting notched specimen before and after demoulding is presented in Fig. 5d,e. After an initial 24-h curing period, the samples were taken out of the molds and allowed to cure further in ambient conditions until the specified testing times.

Table 2.

Blend compositions of GC (kg/m³).

Mixture id	SF	GGBS	FA	Na2SiO3	NaOH	Type of fiber	Length of fiber (mm)	Volume of fiber (%)	Superplasticizer (%)
GC-R	41	163	204	70	101	–	0	–	0.5
GC-CF20						Coir	20	0.5
GC-CF40						Coir	40	0.5
GC-CF60						Coir	60	0.5
GC-FF20						Flax	20	0.5
GC-FF40						Flax	40	0.5
GC-FF60						Flax	60	0.5

Fig. 5.

Casting procedure.

Experimental test setup

The geometric configuration and loading arrangement of the ENDB specimen provide essential insights into the experimental setup and the resulting fracture behavior (Fig. 6). The optimized geometry of the ENDB specimen provides a reliable framework for exploring fracture mechanics under combined-mode loading conditions. Among the parameters analyzed, the notch inclination angle (γ) the angle formed between the notch and the specimen’s horizontal axis plays a crucial role in determining fracture behavior. The chosen γ values of 0°, 20°, 50°, and 63° enable a structured assessment of the transition from mode I to I/III fracture response. Each specimen features a 16 mm notch depth and a 40 mm overall thickness. In a three-point bending setup, the specimen is placed on two lower supports that are symmetrically aligned, while an upper fixture applies an increasing load at a rate of 0.2 mm/min. This setup induces both tensile and shear stresses, effectively simulating real-world fracture scenarios and allowing for accurate measurement of stress intensity factors. Earlier research^43–45 suggests that the fracture behavior of ENDB specimens is mainly affected by the arrangement of the lower supports and the angle of the crack. When γ is at 0°, crack growth is controlled by pure mode I, which is marked by an opening that is perpendicular to the tensile stress applied. As β rises, the fracture behavior shifts towards a mixed-mode I/III, where anti-plane shear (mode III) becomes more influential in the crack propagation process. Recent studies^46–48 quantitatively reveal that increasing γ values lead to a decrease in the mode I stress intensity factor (FI) while enhancing the mode III component (FIII), especially near the central area of the specimen. This behavior indicates a gradual transition from crack opening primarily influenced by tension to deformation mainly governed by shear as the crack angle increases. Theoretical studies^49–51 suggest that a pure mode III fracture is expected to occur at an angle of γ = 90°. However, its occurrence in experiments depends on geometric factors: crack length-to-thickness ratio (a/t) and span-to-length ratio (S/R), both of which are instrumental in defining the local stress field at the crack tip region. A quantitative assessment of fracture behavior requires the calculation of the applied stress (σ), mode I (FI) and mode III (FIII) factors, the effective stress intensity factor (Feff), and the mixity parameter (M^e), as expressed in Eqs. (1)–(5)^52,53. The M^e provides a quantitative measure of the interaction between different fracture modes, with M^e being zero for a purely mode I fracture and one for a purely mode III fracture^54,55. Under mixed-mode conditions, M^e takes on intermediate values, indicating the joint effects of tensile and shear stresses. To ensure the outcomes are reliable and reproducible, each experimental condition was tested three times, which reduces variability and strengthens statistical reliability. Figure 7 shows the experimental fracture setup, including the specimen’s geometry, support configuration, and loading arrangement.

Fig. 6.

Details of the notch configurations (concept adopted and modified from⁵⁴.

Fig. 7.

Experimental test setup.

1

2

3

4

5

The geometric factors for Modes I and III, represented as and, respectively (Table 3), are sourced from existing literature^43,56. The parameters of interest are specified as follows: : signifies the depth of the notch (16 mm), : represents half the span (71.25 mm), : indicates the radius of the disc (75 mm), and denotes the critical load.

Table 3.

Geometrical factor^43,56.

Mode	I	III	I/III	I/III
XI	0.323	1	0.269	0.1
XIII	0	0.073	0.056	0.083

Results and discussions

The control GC (GC-F) confirmed a compressive strength of 43.07 MPa³⁰. The addition of coir fibers had varied effects: GC-CF20 experienced a slight rise in strength to 43.52 MPa, marking a 1.04% enhancement, whereas the use of longer fibers (40–60 mm) causing a decrease in strength to 41.58–39.95 MPa, attributed to poor dispersion, diminished workability, and heightened porosity³⁰. Flax fibers had a more significant effect, with GC-FF20 achieving 45.87 MPa, representing a 6.50% improvement. However, the advantages diminished with longer fibers, as seen in GC-FF40 (4.22%) and GC-FF60 (0.74%)³⁰. In general, shorter fibers boosted compressive performance, and flax outperformed coir at the same lengths, underscoring the significance of fiber type and optimized geometry for effective reinforcement^57,58.

Effect of coir and flax fiber length on mode I-FRTS of GC

Figure 8 illustrates the FRTS characteristics of GC when coir and flax fibers of different lengths are incorporated. The unreinforced control sample (GC-R) showed a baseline FRTS of 0.86 MPa.m^0.5. When coir fibers were added, the FRTS values improved to 0.95, 1.03, and 0.87 MPa.m^0.5 for GC-CF20, GC-CF40, and GC-CF60, respectively, indicating improvements of 9.70%, 18.96%, and 1.04% compared to the control (Fig. 8a). In a similar manner, GC enhanced with flax fibers (GC-FF20, GC-FF40, and GC-FF60) reached FRTS levels of 0.91, 0.99, and 0.85 MPa.m^0.5. These results signify enhancements of 5.67% and 14.93% for the first two, while the last one experienced a slight decrease of 1.19% (Fig. 8b). Both fiber types exhibited a consistent trend, with FRTS enhancing as fiber length increased, peaking at 40 mm, beyond which the performance started to decline⁵⁹. The enhanced toughness observed at this fiber length can be attributed to a well-balanced interaction between fiber dispersion, the bonding at the matrix-fiber interface, and efficient stress transfer. At 40 mm, fibers have enough embedded length to effectively bridge microcracks and dissipate fracture energy through mechanisms of pull-out and frictional resistance, without causing clustering or entanglement during mixing³⁰. This optimal microstructural arrangement improves crack-bridging efficiency, promotes even stress distribution, and slows down crack propagation, which consequently enhances the overall fracture performance³⁰.

Fig. 8.

Mode I FRTS of GC.

In contrast, employing 60 mm fibers yielded a noticeable decrease in mode I FRTS for systems reinforced with both coir and flax. This decline is mainly attributed to reduced workability and uneven fiber distribution, which cause fiber clumping, voids, and localized stress concentrations within the geopolymer matrix. Additionally, fibers that are too long tend to curl or fold instead of aligning with the intended crack path, which reduces their effectiveness in being pulled out and limits their ability to bridge cracks efficiently⁶⁰. In comparison, coir fiber-reinforced GC consistently surpassed its flax-reinforced equivalent in performance across all fiber lengths. This superior performance is mainly due to the inherent properties of coir fibers, particularly their rougher surface texture and greater elongation capacity, which promote better mechanical interlocking with the geopolymer matrix and increased energy absorption during fiber stretching and pull-out^30,61. On the other hand, flax fibers, due to their rigidity and smoother surfaces, are prone to early debonding and brittle failure, which diminishes their ability to bridge cracks effectively⁶². In contrast, coir fibers offer superior crack-arresting capabilities and enhance ductility and fracture toughness after cracking, making them a more effective reinforcement option for improving the fracture performance of GC.

Effect of coir and flax fiber length on mode I/III FRTS of GC

Figure 9 depicts the FRTS behavior in mixed-mode conditions for GC composites reinforced with coir and flax fibers of different lengths. The baseline composite (GC-R) demonstrated FRTS values of 0.88 MPa.m^0.5 and MPa.m^0.5 at mixed-mode angles (γ) of 20° and 50°, respectively. The gradual reduction in FRTS as γ increases indicates a growing impact of shear stresses, which naturally lower the apparent fracture toughness due to decreased crack-bridging effectiveness and increased interfacial sliding along the crack path. Adding coir fibers notably improved the FRTS of the GC. At an angle of γ = 20°, the FRTS values for the GC-CF20, GC-CF40, and GC-CF60 specimens were recorded as 1.00, 1.06, and 0.89 MPa.m^0.5, respectively, as shown in Fig. 9a. These figures indicate increases of 13.18%, 20.65%, and 0.75% when compared to GC-R. A comparable pattern, albeit slightly less pronounced, was noted for flax fibre reinforcement. The FRTS values for GC-FF20, GC-FF40, and GC-FF60 were 0.99, 1.02, and 0.88 MPa.m^0.5 (Fig. 9b), which correspond to enhancements of 12.19%, 15.92%, and − 0.62% relative to the GC-R specimen. These findings collectively suggest that fibers of medium length, around 40 mm, offer the most significant improvement in fracture toughness when subjected to predominantly tensile-dominated mixed-mode loading. The increase in FRTS at γ = 20° is likely due to the effective crack-bridging and pull-out actions facilitated by the natural fibers. These processes enable the dissipation of fracture energy through the stretching of fibers, debonding at interfaces, and resistance due to friction during pull-out. Coir fibers, characterized by their uneven and rough surface texture, form strong mechanical bonds with the geopolymer matrix, enhancing stress transfer across the crack interface. While flax fibers also aid in fracture toughness through similar processes, their relatively smoother surface and lower ductility reduce the amount of interfacial frictional energy dissipation, leading to a slightly lesser toughening effect³⁰.

Fig. 9.

Mixed-mode (I/III) FRTS of GC.

When γ is at 50°, all samples exhibited a significant decrease in FRTS, which aligns with the dominance of shear deformation over the typical crack opening. In these circumstances, the effectiveness of fiber bridging is reduced, as shear stresses lead to fiber shearing and interfacial slip rather than tensile pull-out. As illustrated in Fig. 9c, the FRTS measurements for the GC-CF20, GC-CF40, and GC-CF60 composites were 0.71, 0.73, and 0.67 MPa.m^0.5, respectively. These observations reveal increments of 6.82%, 9.30%, and 0.56% in comparison to GC-R. Analogously, flax fibre composites (GC-FF20, GC-FF40, and GC-FF60) attained respective values of 0.71, 0.72, and 0.66 MPa.m^0.5 (Fig. 9d), corresponding to variations of 5.43%, 8.14%, and − 1.01%. The exceptional performance of the 40 mm fibers under both mixed-mode conditions highlights an ideal combination of fiber length, aspect ratio, and distribution within the geopolymer matrix. Fibres that are 20 mm long have a restricted capacity to bridge due to insufficient embedding depth, whereas those measuring 60 mm tend to cluster, leading to an uneven spread and weaker bonds with the matrix³⁰, as well as localized stress concentrations that negate potential toughening advantages. Moreover, the ability of fibers to resist being pulled out is determined by the strength of the bond at the interface and the surface characteristics of the reinforcing fibers. Coir fibers exhibit greater resistance to pull-out compared to flax, due to their lignocellulosic makeup, uneven surface texture, and improved bonding with the geopolymeric matrix³⁰.

Effect of Coir and flax fiber length on mode III FRTS of GC

Figure 10 illustrates the mode III FRTS properties of GCs that are reinforced with coir and flax fibers of different lengths. The unreinforced reference sample (GC-R) showed a baseline FRTS of 0.44 MPa.m^0.5. Adding 20 mm coir fibers (GC-CF20) increased this value to 0.47 MPa.m^0.5 (Fig. 10a), which represents an enhancement of about 5.61% compared to the control. Using coir fibers of 40 mm length (GC-CF40) resulted in a further enhancement of FRTS, reaching 0.48 MPa.m^0.5, which signifies an overall rise of 9.70%. In contrast, the inclusion of 60 mm coir fibers (GC-CF60) led to a decrease in FRTS to 0.43 MPa.m^0.5, marking a 2.64% reduction compared to GC-R. In GC reinforced with flax fibers, a comparable behavior was recorded. The GC-FF20 sample, which includes 20 mm flax fibers, demonstrated a FRTS of 0.46 MPa.m^0.5, marking a 4.09% enhancement compared to the control. (Fig. 10b). Extending the fibre length to 40 mm (GC-FF40) resulted in a toughness of 0.48 MPa.m^0.5, signifying an 8.05% enhancement. Conversely, a slight decline in fracture toughness was detected for the composite reinforced with 60 mm flax fibres (GC-FF60), which achieved a toughness of 0.43 MPa.m^0.5, indicating a 1.58% decrease compared to GC-R. These findings reveal a strong reliance of fracture behavior on the length of the fibers, indicating that moderate fiber reinforcement (20–40 mm) promotes optimal interaction between fibers and the matrix. In contrast, longer fibers (60 mm) lead to negative outcomes due to clumping and poor stress distribution. The results clearly indicate the pivotal role of customizing fiber dimensions to balance the competing requirements of strong crack-bridging and homogeneous fiber dispersion, thereby enhancing the fracture toughness³⁰. A study by Bazarkhankyzy et al.³⁰ demonstrated that the incorporation of coir, flax, jute, and sisal fibers enhances crack resistance by enabling effective stress transfer across developing fractures. These natural fibers promote a discrete bridging mechanism that restrains crack propagation and reduces crack opening, thereby contributing to improved damage tolerance within the composite matrix. These behaviour is consistently reported by the other studies^63–65.

Fig. 10.

Mode III FRTS of GC.

Mixity parameter

In ENDB specimens, the mixity parameter provides a quantitative measure of the proportion of shear relative to opening at the crack tip. It serves to characterize the mixed-mode fracture state generated by the eccentric application of the load. The relationship between M^e and the effective FRTS of GC is clearly inverse: as Me increases, the effective FRTS decreases for all mixtures (Fig. 11). When fracture modes transition from being tensile to shear-dominated, the material’s ability to resist crack growth diminishes due to limited tolerance for shear-induced microcracks and early debonding between fibers and the matrix. Increased mixity diminishes energy loss from fiber pull-out and crack bridging, resulting in unstable crack growth and reduced overall fracture toughness. All GC specimens exhibited effective FRTS values surpassing 0.85 MPa.m^0.5 under both pure mode I loading (M_e = 0) and mainly tensile–shear conditions (M_e = 0.869), indicating excellent resistance to crack initiation and propagation in tension-dominated scenarios. Reducing the mixity parameter to M_e = 0.559 led to a decrease in effective FRTS, which fell within the range of 0.66–0.73 MPa.m^0.5. This change signified a transition from tensile to mixed-mode fracture behavior. The reduction in toughness indicates a growing dominance of shear stresses at the crack tip, which limits intrinsic energy-dissipation processes like fiber pull-out and crack bridging. A significant decline in fracture toughness was observed under conditions of fully shear-dominated loading (M_e = 1.0), where the effective FRTS values dropped to a range of 0.43 to 0.48 MPa.m^0.5. This decrease highlights the geopolymer matrix’s limited ability to withstand shear-induced damage, which mainly results from interfacial debonding and micro-shearing at the fibre-matrix interface. In pure mode III scenarios, deformation is concentrated along shear planes, preventing the activation of tensile fracture mechanisms that usually improve toughness through fibre–matrix interaction.

Fig. 11.

Mixity parameter of GC.

Failure pattern of GC

Figures 12 and 13 illustrate the fracture patterns observed in GC specimens when exposed to different stress conditions. Fractures consistently begin at the midpoint of the specimen and then extend along the entire crack front, aligning with the points of maximum stress intensity factors for modes I, III, and mixed-mode I/III. In pure mode I loading, fractures occur symmetrically when the critical load is reached. Both specimens without fibers and those reinforced with coir or flax fibers, regardless of whether the fibers are 20, 40, or 60 mm long, experience bifurcation along the plane of the initial notch, consistent with previous findings⁶⁵. The primary mechanism is characterized by consistent tensile stresses that act perpendicular to the crack plane, encouraging a planar, opening-mode fracture. Without the presence of fibers, cracks spread quickly due to the absence of bridging mechanisms³⁰, which exhibits a lower fracture toughness as depicts in Fig. 8. In fibrous materials, fibers enhance resistance by bridging cracks and dissipating energy; nevertheless, the tensile stresses applied still surpass the combined tensile strength of both the matrix and the reinforcement, leading to a fracture along the notch plane. The fact that fracture behavior seems largely unaffected by fiber length specifies that, in pure mode I scenarios, failure is mainly governed by the composite’s inherent tensile characteristics and the shape of the initial notch. Coir and flax fibers primarily contribute to improving the load-bearing capacity after cracking, rather than changing the main crack path, which enhances the fracture toughness, as depicts in Fig. 8.

Fig. 12.

Failure pattern of fiber-free and coir fibrous GC specimens.

Fig. 13.

Failure pattern of flax fibrous GC specimens.

When experienced to mixed-mode I/III loading, crack paths deviate from the initial crack plane, resulting in a twisted, non-planar shape. In specimens without fibers, the failure is mainly brittle, displaying a variety of fracture mechanisms. Similarly, pure mode III loading causes a helical fracture path around the middle of the specimen, characterized by an antisymmetric curvilinear surface in relation to the original notch. The significant asymmetry of the fracture surfaces seen under I/III and III conditions suggests that crack propagation begins at the specimen’s midspan and moves outward, creating an antisymmetric fracture profile. Hypotheses suggesting different locations for fracture initiation do not align with the morphologies observed in experiments. Fibrous specimens failed to maintain their structural integrity across all three fracture modes, ultimately leading to complete separation. This behavior is ascribed to the fibers’ limited effectiveness in bridging developing cracks and hinder fracture progression, thus being unable to avert catastrophic fragmentation³⁰. As a result, failure is marked by localized cracking instead of maintaining structural integrity. The fracture patterns observed are similar in both coir- and flax-reinforced samples, highlighting that while the fibers offer some reinforcement, they are not adequate to keep the specimens cohesive under various stress conditions³⁰.

In summary of “Results and discussions”, the inclusion of coir and flax fibers markedly improves the fracture toughness of GC subjected to mode I, mixed-mode (I/III), and mode III loading regimes. Among the evaluated fiber lengths, the 40 mm fibers consistently yielded the most pronounced enhancements in fracture toughness, attributable to their balanced distribution, adequate embedment, and effective crack-bridging and pull-out responses. Conversely, extending the fiber length to 60 mm diminished performance due to reduced workability, fiber agglomeration, and the development of localized stress concentrations. Across all loading configurations, coir-based composites demonstrated superior resistance compared to flax-reinforced systems, a behavior linked to the inherently rougher surface texture and greater deformability of coir fibers, which facilitate improved mechanical interlock and energy dissipation. As the fracture mode shifted toward shear-dominant conditions, all mixtures exhibited significant reductions in toughness, underscoring the limited capability of fibers to counteract shear-controlled crack advance. Fractographic evidence further indicated that crack trajectories were primarily dictated by the prevailing fracture mode, with fibers contributing to post-crack resistance yet insufficient to avert ultimate separation under severe mixed-mode and pure mode III loading. Although the study offers valuable insights, several limitations present opportunities for further investigation:

• A systematic quantification of the influences of fiber dispersion, mixture rheology, and void generation on FRTS is essential for formulating optimized mix designs that effectively mitigate clustering tendencies, particularly in composites incorporating longer fibers with higher dosages.

• Implementing targeted chemical or mechanical modification techniques for coir and flax fibers may strengthen fiber-matrix interfacial interactions and thus improve their load-transfer performance when subjected to shear-governed stress states.

• Hybridization strategies involving multiple natural fibers or combinations of natural and synthetic fibers have the potential to generate synergistic reinforcement mechanisms that enhance crack-bridging across diverse fracture modes. However, the durability of these systems remains insufficiently understood, as the long-term influences of moisture ingress, thermal fluctuations, and chemical environments on fiber-matrix interfacial integrity have not yet been comprehensively characterized an essential consideration for their reliable use in structural applications.

Microstructure study

SEM analysis of GC

Figure 14a illustrates a geopolymer matrix that is densely packed, characterized by a uniformly distributed collection of reaction products. This suggests an effective geopolymerisation process involving the FA, slag, and SF precursors. The formation of a continuous N–A–S–H gel structure showcases a significant level of bonding continuity and notable pore refinement, which enhances the matrix’s compactness and fracture toughness (Fig. 8). Nonetheless, the presence of crystalline residues like unreacted FA, quartz, and mullite indicates incomplete dissolution of aluminosilicates. This is a typical limitation of FA-based systems, which can lead to localized inconsistencies while still offering the potential for prolonged pozzolanic activity³⁰. The matrix predominantly consists of a cohesive amorphous gel phase with a few crystalline fragments, indicating a mature and well-balanced geopolymerisation process. Figure 14b SEM reveal the interactions at the interface between coir fibers and the geopolymer matrix, marked by partial fiber debonding and limited crack propagation. This controlled separation at the interface promotes effective crack-bridging and energy dissipation processes, thereby enhancing the composite’s fracture toughness and ductility. The formation of needle-like ettringite crystals within the matrix and at the fiber–matrix interface indicates secondary hydration or calcium-sulfate reactions linked to slag components. Although moderate formation of ettringite can improve the densification of the microstructure, excessive crystallization might affect dimensional stability and lead to microcracking. Conversely, the microstructure shown in Fig. 14c, which pertains to the flax fiber–reinforced system, exhibits similar debonding behavior but with a significantly greater level of microstructural heterogeneity. The significant buildup of ettringite around flax fibers suggests enhanced secondary reactions, which may impede proper matrix encapsulation and reduce the adhesion between the fiber and matrix. The matrix remains primarily composed of an amorphous N–A–S–H gel phase³⁰.

Fig. 14.

SEM analysis of GC with and without fiber.

In a comparison of all systems, the unreinforced geopolymer matrix exhibits a consistently solid and unified microstructure, while the fiber-reinforced versions create interfacial gaps that significantly affect the fracture behavior. Reinforcing with coir fiber fosters a relatively stable interfacial transition zone, which allows for controlled debonding and excellent crack-bridging capabilities, thus improving the composite’s ability to absorb energy. In contrast, flax fiber reinforcement creates more chemically active and varied interfacial areas, resulting in less stable bonding and a higher likelihood of microstructural variability. As a result, the characteristics of the natural fiber are crucial in determining the interfacial microstructure, the overall integrity of the matrix, and the primary toughening mechanisms that influence the mechanical performance of the GC system³⁰. Overall, the dense amorphous N–A–S–H gel in the unreinforced GC forms a continuous matrix that delays crack initiation but leads to a predominantly brittle fracture. Coir-reinforced specimens exhibit controlled fiber debonding and crack-bridging, enhancing energy absorption and toughness. In contrast, flax reinforcement introduces heterogeneity and ettringite-rich interfaces that weaken fiber–matrix bonding and promote unstable crack paths. Overall, variations in the interfacial transition zone critically govern crack propagation and fracture resistance across the systems.

XRD analysis of GC

The XRD examination of the GC produced from FA, GGBS, and SF uncovers a complex and well-balanced mineralogical composition. This composition highlights the interaction between the reactivity of the precursors, the availability of calcium, and the geopolymerisation kinetics (Fig. 15). Significant diffraction peaks at around 20.84° and 26.58° 2θ, which correspond to quartz, along with reflections at 16.46° and 33.08° 2θ linked to mullite, indicate the presence of crystalline residues derived from the FA precursor⁶⁶. The persistence of these aluminosilicate phases suggests that a portion of the FA is only partially dissolved during alkaline activation, a typical characteristic of FA-based geopolymeric systems due to the uneven distribution of their amorphous and crystalline components⁶⁷. Although these unreacted crystalline residues can enhance the packing efficiency and dimensional stability of the matrix through microfiller effects, retaining too much of these phases might impede the development of a continuous amorphous gel network, thus limiting the composite’s overall microstructural cohesion and long-term performance. Reflections attributed to calcite within similar 2θ intervals offer strong evidence of carbonation phenomena, primarily caused by the inclusion of slag. The addition of GGBS boosts the concentration of calcium ions in the reactive environment, which then reacts with atmospheric CO₂ to form calcite⁶⁸. This finding highlights the dual role of calcium in the geopolymer matrix: it not only aids in the creation of N–A–S–H gels that greatly improve early mechanical strength but also makes the system more prone to carbonation³⁰. The appearance of faint portlandite peaks, which is rare in systems without Portland cement, indicates either partial hydration or an incomplete reaction of calcium-rich slag particles. This may be due to variations in alkali activation or limited ionic mobility caused by diffusion constraints. The minor diffraction signals linked to hematite and magnetite are likely due to inherent impurities in the raw materials or the formation of secondary phases under alkaline activation conditions. While these iron-containing oxides do not take part directly in the geopolymerization process, they can subtly alter the surrounding chemical environment, affecting pH stability, charge equilibrium, and the development of the gel network³⁰.

Fig. 15.

XRD analysis of GC with and without fiber.

Overall, unreacted quartz and mullite improve packing but act as stiff stress concentrators that influence crack initiation. Carbonation-derived calcite increases density and promotes more tortuous, energy-absorbing crack paths. Residual portlandite and partially reacted slag create heterogeneous zones that favour crack nucleation. Overall, the mixed amorphous-crystalline structure balances toughness gains from densification with reduced fracture resistance from microstructural discontinuities.

TGA analysis of GC

Thermogravimetric analysis (TGA) of GC that includes FA, GGBS, and SF shows a series of distinct mass-loss occurrences as the temperature rises from room temperature to 1000 °C, each representing particular physicochemical changes within the binder matrix (Fig. 16). The initial decrease in mass, occurring between roughly 33 °C and 99 °C, is largely resulting from the evaporation of weakly bound and free water molecules occupying the pore spaces⁶⁹. This process, typical of cementitious materials, signifies the release of moisture that is not chemically bound and was introduced during the early mixing and curing phases^70,71. Compared to standard Portland cement systems, the degree of mass loss in geopolymer matrices is typically less, due to their naturally denser microstructure and reduced capillary porosity⁷². This characteristic is further enhanced by the addition of ultrafine SF, which enhances particle packing and supports the development of a more continuous gel network. While the evaporation of physically bound water has a negligible immediate impact on mechanical strength, the varying rates of thermal contraction can lead to microcracks, especially in slag-rich mixtures that dry more quickly⁷¹. A further stage of mass loss occurs between roughly 219 °C and 265 °C. This phase involves the dehydration of loosely attached interlayer water and the partial breakdown of weakly polymerized gel structures. It also includes the elimination of remaining moisture from amorphous hydrates and, in some cases, the evaporation of unreacted organic or activator components. During this period, the extent of mass loss is typically more significant in mixtures enriched with slag. This is due to the existence of calcium aluminosilicate hydrate gels, which hold a greater amount of chemically bound water compared to the N–A–S–H gels found in systems dominated by fly ash. From a mechanical perspective, this dehydration causes minor changes in the microstructure and an increase in pore size, leading to a slight decrease in stiffness without immediately affecting compressive strength. A significant reduction in mass is observed between roughly 370 °C and 528 °C, which aligns with the main breakdown of the geopolymer gel network. This process involves the dehydroxylation and structural breakdown of N–A–S–H phases, leading to the release of chemically bound water and the breaking of Si–O–Al and Si–O–Si bonds within the amorphous aluminosilicate structure³⁰. SF plays a stabilizing role at this phase, as its high silica content and reactivity promote improved polymerization.

Fig. 16.

TGA analysis of GC.

Overall, initial mass loss from free and weakly bound water (33–99 °C) induces thermal contraction and early microcracking. Dehydration of interlayer water (219–265 °C) enlarges pores and reduces stiffness, easing crack initiation and growth. Major mass loss from gel dehydroxylation (370–528 °C) severely weakens the aluminosilicate network, lowering fracture energy and increasing brittleness. Although SF enhances stability, progressive thermal degradation ultimately reduces crack-bridging capacity at high temperatures.

FTIR analysis of GC

The FTIR spectrum of GC synthesized from FA, GGBS, and SF displays a range of distinct absorption bands, each representing particular structural and chemical features of the aluminosilicate binder (Fig. 17). The 429 cm⁻¹ band is linked to the bending vibrations of Si–O bonds, signifying the creation of a stable silicate structure and verifying the successful development of a cohesive geopolymeric network^73,74. The connectivity of Si–O tetrahedra plays a vital role in mechanical properties. A significant peak at 709 cm⁻¹, associated with the symmetric stretching vibrations of Si–O–Si and Si–O–Al bonds, further confirms the effective polymerization of silicate and aluminate species into a three-dimensional cross-linked structure. This network enhances internal load transfer and contributes to the composite’s inherent durability and chemical resistance. The absorption detected at 875 cm⁻¹ is attributed to carbonate groups, which are likely formed due to the interaction between alkali-activated materials and atmospheric CO₂ during the curing process⁷⁵. Partial carbonation can lead to a denser microstructure, which may enhance surface hardness, while also affecting long-term alkalinity and durability. The main band at 1016 cm⁻¹, linked to the asymmetric stretching of Si–O–T (T=Si or Al) bonds, signifies the primary geopolymeric gel network formed from the combined pozzolanic effects of FA, slag, and SF^73,76. Changes in the location and strength of this band reflect the level of polymerization, where greater Si–O–Al connectivity is linked to improved mechanical strength and decreased porosity. Moreover, additional absorption bands at 1320 cm⁻¹ and 1390 cm⁻¹ confirm the existence of carbonate species, indicating that carbonation penetrates beyond the surface layers into the core matrix⁷⁵. These phenomena can affect dimensional stability and the behavior of long-term shrinkage. The bands at 1562 cm⁻¹ and 1624 cm⁻¹, which are linked to the bending vibrations of molecular water, indicate the presence of physically adsorbed or gel-bound water within the matrix⁷⁶. The existence of leftover moisture suggests that polycondensation is not fully complete, which might temporarily decrease stiffness while also promoting further curing and delayed strength enhancement. A slight absorption at 1978 cm⁻¹, attributed to overtone or combination vibrations of silicate bonds, indicates subtle structural variations within the aluminosilicate framework. This likely reflects the intricate spatial distribution of Si and Al species in this multi-source precursor system. FTIR results show a highly polymerized Si–O–Al network that enhances strength through improved load transfer and reduced porosity. Carbonate bands indicate partial carbonation, increasing surface densification and durability. Gel-bound water reflects ongoing polycondensation, supporting gradual strength gain and long-term stability. Overall, the FTIR results indicate a well-polymerized Si–O–Al network that enhances matrix cohesion and delays crack initiation. Carbonate bands suggest partial densification, leading to more tortuous crack paths and greater energy dissipation. Moisture-related peaks imply localized reductions in stiffness that can influence early microcracking. Together, these features improve crack resistance and contribute to a tougher, less brittle fracture response.

Fig. 17.

FTIR analysis of GC.

Insights and outlook of this research

Beyond the experimental observations, the present investigation offers several substantive contributions to the understanding of fiber-reinforced GC. First, the findings demonstrate that the combined influence of fiber aspect ratio and loading condition is more significant than has been previously acknowledged in geopolymer systems. The consistent performance advantage of the 40 mm fibers under all fracture modes indicates the presence of a geometric threshold at which natural fibers can develop effective bridging stresses without triggering agglomeration, a relationship that has not been clearly identified in earlier research. Second, the differing behaviours exhibited by coir and flax fibers underscore the importance of interfacial chemistry in governing fracture resistance, even when fiber dimensions are comparable. The contrasting interfacial responses characterized by stable debonding for coir and ettringite-influenced variability for flax suggest that fiber-matrix interactions must be evaluated with respect to both mechanical attributes and the evolving geopolymer gel chemistry. These results provide novel evidence that the development of N–A–S–H gels directly affects fiber adhesion, an aspect that remains insufficiently explored in natural-fiber geopolymer literature. Third, the integrated microstructural analyses reveal that crystalline remnants from fly ash and Ca-rich constituents from GGBS actively contribute to fracture behaviour by altering matrix packing, gel homogeneity, and energy-dissipation mechanisms. The combined interpretation of SEM, XRD, TGA, and FTIR data supports the view that the geopolymer matrix is a chemically dynamic medium in which partial carbonation, moisture retention, and heterogeneous gel formation collectively influence composite toughness.

Conclusions

This research showed a systematic assessment of how the aspect ratios of coir and flax fibers affect FRTS in GC synthesized with FA, GGBS, and SF. The conclusions reported herein are substantiated by the experimental evidence.

1. In Mode I, the FRTS of GC was enhanced by both coir and flax fibers up to a length of 40 mm, with GC-CF40 and GC-FF40 reaching values of 1.03 MPa.m^0.5 (+ 18.96%) and 0.99 MPa.m^0.5 (+ 14.93%), respectively. However, when the fibers were extended to 60 mm, performance declined (GC-CF60: 0.87 MPa.m^0.5; GC-FF60: 0.85 MPa.m^0.5), indicating that excessive fiber length results in clustering and reduced fracture toughness.

2. Under I/III loading, 40 mm fibers provided the greatest FRTS enhancement. At γ = 20°, GC-CF40 and GC-FF40 reached 1.06 (+ 20.65%) and 1.02 MPa.m^0.5 (+ 15.92%), respectively. At γ = 50°, toughness decreased due to shear-induced fiber slip, but GC-CF40 and GC-FF40 still showed improvements of 0.73 (+ 9.30%) and 0.72 MPa.m^0.5 (+ 8.14%), confirming the effectiveness of intermediate-length fibers.

3. Under Mode III, 40 mm fibers maximized FRTS, with GC-CF40 and GC-FF40 reaching 0.48 MPa.m^0.5 (+ 9.70% and + 8.05%). Extending fibers to 60 mm reduced toughness below the control (GC-CF60: 0.43 MPa.m^0.5, − 2.64%; GC-FF60: 0.43 MPa.m^0.5, − 1.58%) due to agglomeration and reduced stress transfer.

4. Under mode I (tensile) loading, all specimens fractured along the original notch, with fibers enhancing post-crack energy dissipation but not altering crack paths, and fiber length (20–60 mm) having little effect. Under shear-dominated or mixed-mode loading, fractures were non-planar and helical, with fibers providing limited crack bridging, leading to complete segmentation and localized failure regardless of fiber type or length.

5. SEM analysis demonstrated that coir fibers form stable interfacial zones that promote controlled debonding and effective crack bridging, enhancing fracture toughness, whereas flax fibers exhibit more variable interfaces due to ettringite formation, resulting in reduced adhesion and toughness.

6. XRD, TGA, and FTIR analyses revealed that the geopolymer matrix consists of a dense N–A–S–H gel with residual crystalline phases (FA, quartz, mullite) and calcium-rich phases from slag, which improve packing, dimensional stability, thermal resilience, and mechanical performance, while partial carbonation and retained moisture influence ongoing curing and microstructural densification.

Author contributions

A.B.: Conceptualization, methodology, investigation, data curation, Y.A.: Supervision, conceptualization, funding acquisition. Z.I.: Supervision, conceptualization. M.K.: Writing—original draft, writing—original draft, funding acquisition. H.S.A.: Writing—original draft, writing—original draft, funding acquisition. S.A.: Writing—original draft, writing—original draft. E.S.F.: Writing—original draft, writing—original draft, funding acquisition. G.M: Conceptualization, writing—original draft, writing—review and editing.

Funding

No funding is available for this research work.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article.

Declarations

Competing interests

The authors declare no competing interests.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the authors used Chat GTP AI in order to enhance readability and language quality. After using this tool, the authors reviewed and edited the content as needed and take responsibility for the content of the publication.