3D-Printed Gypsum–Cement–Pozzolan Composites with Crumb Rubber: Strength and Durability

Abstract

This research investigates the formation and behavior of sustainable crumb rubber-modified gypsum–cement–pozzolan (GCP) composites, with a view to their use in a broad concept for construction. GCP binders are gaining attention as a low-carbon replacement for Portland cement, and the addition of recycled rubber helps the achievement of circular economy goals and potentially increases durability. The present research evaluates the impact of crumb rubber (CR) on the mechanical strength, water absorption, dimensional stability, and freeze–thaw resistance of 3D-printed GCP-rubber composites. Composite blends of variable proportions of crumb rubber were prepared at constant binder ratios. Mechanical properties were defined by prism specimens (40 × 40 × 160 mm) by the flexural and compressive strengths, and deformation was determined by micrometers to measure longitudinal strain as a function of curing. Water absorption was determined prior to freeze–thaw cycling to define pore saturation. Durability was investigated using two approaches: (1) controlled freeze–thaw experiments on cube specimens, with XF1 grade performance achieved, and (2) ultrasonic pulse velocity (UPV) testing of specimens 3D-printed for assessing internal structural change after long-term frost exposure. Results showed that compressive strength decreased moderately (10–20%) with increasing rubber content from 17% up to 50%, while flexural strength improved up to 15%, showing the elastomeric action of CR. Water absorption was reduced by 5–8% in the rubber-modified blends due to the hydrophobic character of rubber. Deformation tests also confirmed minimum length variation (<0.02%) during curing. Freeze–thaw durability was enormously improved, and test specimens retained more than 95% of initial strength. UPV measurements detected only a relatively modest velocity drop (~50 m/s) after 36 days cycling with subsequent stabilization up to 200 days, demonstrating long-term internal structure with minimal progressive damage. In summary, the findings demonstrate that GCP composites with crumb rubber incorporated are printable, dimensionally stable, and capable of freeze–thaw degradation resistance. Despite a moderate loss of compressive strength, the balance of introduced durability and sustainability suggests their competence as viable materials for additive manufacturing in construction.

1. Introduction

The construction industry is one of the largest consumers of natural resources globally and a major emitter of greenhouse gases. Production of cement alone accounts for nearly 8% of the world’s CO₂ emissions, primarily due to the energy consumed in the calcination of limestone and the huge quantity of cement consumed annually [1,2,3]. With urbanization around the globe continuing to accelerate, the need for cementitious materials is likely to grow further, placing additional pressure on raw materials, energy resources, and waste streams. This environmental footprint has prompted intensive research into sustainable binder alternatives, recycling strategies for materials, and new construction technologies that can together minimize the sector’s contribution to climate change [4,5,6,7].

One such technological advancement that has garnered a lot of attention in recent years is 3D printing of cementitious composites, also known as additive manufacturing in construction. Compared with traditional casting techniques, which involve the use of formwork and create significant waste, 3D printing allows automated layer-by-layer deposition of mortar-like materials [1,8,9]. This process has several advantages: it reduces material usage by optimizing geometries, shortens construction time through automation, lowers labor requirements, and offers design freedom beyond the potential of traditional techniques. Case studies demonstrated that additive manufacturing could reduce waste by up to 60% and cut project delivery times significantly, with both financial and environmental benefits [1,8,10]. Nevertheless, printable binder performance remains a limit to widespread adoption. For 3D printing to be mainstreamed into structural construction, binders must not only be extrudable and buildable but also strong, durable, and environmentally sustainable [8,9].

To this end, GCP composites have exhibited excellent promise. GCP systems combine the rapid setting and low calcination energy of gypsum with the long-term strength development of Portland cement and the durability benefits of pozzolanic materials such as fly ash, metakaolin, or slag [11,12,13,14,15,16]. This ternary binder concept exploits synergies between the constituents: gypsum accelerates setting, cement provides hydraulic hardening, and pozzolans engage in secondary hydration reactions that densify the microstructure [12,13,17,18]. Importantly, gypsum requires a lower calcination temperature (~150–200 °C) compared to cement (~1450 °C), and hence, partial substitution of cement with gypsum can achieve significant reductions in energy use and CO₂ emissions [2,4,6]. Life cycle analyses have shown that GCP composites can embody up to 40% less carbon than Portland cement mortars and thus represent an attractive route to sustainable construction materials [2,3,11,19,20].

Several benefits of GCP composites have been identified in the literature. Early studies demonstrated that gypsum-dominant cement–pozzolan blends exhibit improved dimensional stability, reduced shrinkage, and maintain compressive strength at practically usable levels [12,13]. More recent research highlights that GCP binders possess enhanced sulfate resistance and satisfactory rheological behavior for extrusion-based processing [9,14,21]. Despite these positive attributes, their application in 3D printing remains underexplored. The majority of additive manufacturing research continues to focus on cement-only systems, with the optimization of parameters such as yield stress, thixotropy, and setting time being conducted for Portland cement mortars [1,8,9]. Comparatively few studies have tailored GCP systems to rheology, layer retention, and durability, and there is consequently a vast literature gap [11,14,21]. However, existing GCP studies have mainly addressed binder formulation, hydration, and general mechanical performance, while the combined influence of crumb rubber addition and extrusion-based 3D printing on durability remains insufficiently clarified.

Another area of imminent concern in construction sustainability is end-of-life tire (ELT) management. More than one billion tires are retired worldwide annually, producing colossal amounts of difficult-to-dispose waste [19,22]. Landfilling of tires is hazardous due to fire risks, leaching, and vector-borne disease, while incineration emits toxic pollutants [22]. Recycling tires into crumb rubber is an environmentally friendly solution by producing an elastomeric aggregate that can be used as a construction material [22,23,24,25]. Crumb rubber (CR) has been extensively studied in Portland cement concrete as a partial replacement of fine aggregates. Results consistently show that while compressive strength is reduced due to the lower stiffness of rubber, other properties are improved: flexural strength, ductility, and impact resistance are enhanced; freeze–thaw durability improves; and thermal and acoustic insulation performance are increased [23,24,25,26,27,28]. These benefits are attributable to the rubber’s elasticity, and hydrophobicity decreases the bound with crumb rubber, which dissipate stresses and reduce water ingress [23,24,29]. Nevertheless, most rubberized composite studies are based on Portland-cement matrices and conventionally cast specimens, rather than gypsum–cement–pozzolan matrices designed for additive manufacturing.

Existing studies have largely investigated CR in conventional Portland cement systems, gypsum–cement–pozzolan (GCP) binders independently or extrusion-based 3D printing separately. In contrast, the present study examines the combined interaction of these three domains within a single hybrid material framework. Unlike traditional rubberized concrete research, crumb rubber is incorporated into a ternary GCP matrix rather than a cement-only system, introducing different hydration, rheological, and durability mechanisms. Unlike prior GCP studies that primarily focused on binder chemistry or sustainability, this work evaluates GCP under additive manufacturing constraints, including rheology and freeze–thaw durability. Therefore, the novelty of this study lies in establishing how crumb rubber functions within a printable GCP system as a multifunctional phase influencing strength, durability, and processability simultaneously.

Combining CR with gypsum–cement–pozzolan binders for 3D printing represents an underexplored material system that links low-carbon binder design, ELT waste valorization, and additive manufacturing durability. However, interactions between rubber and GCP binders are known [11,14,30,31]. The key questions are whether the compressive strength loss observed in cement–rubber concretes also occurs in GCP–rubber composites, how rubber affects hydration and microstructure development in GCP systems, and whether the expected durability benefits—particularly freeze–thaw resistance—are preserved or even improved [23,24,25,26,27]. Additionally, research has not yet explored the performance of such composites when created through 3D printing, where rheology, interlayer bonding, and dimensional stability pose additional challenges [8,9,11].

This is a critical knowledge gap because freeze–thaw durability is one of the principal durability problems in temperate and cold climates. Freezing of pore water increases its volume by approximately 9%, generating stresses that can crack cementitious materials. Freeze–thaw cycling accelerates damage, reducing service life [13,28,32]. CR, being hydrophobic and elastic in character, has been shown to mitigate such damage by introducing stress-relief zones and minimizing water ingress [23,24,25,26,28]. Confirmation and quantification of this effect in GCP–rubber composites under controlled testing, however, is required if such materials are to be seriously considered for construction. Further, in 3D-printed structures, anisotropy of the layered deposition tends to increase freeze–thaw susceptibility along interlayer planes. Hence, there is a need for novel testing methods such as UPV monitoring for non-destructive microstructural assessment during long exposure periods [28,32,33,34]. A known limitation of rubber incorporation in conventional cast cementitious systems is the tendency of low-density CR to segregate or migrate upward during placement due to density differences relative to mineral constituents, potentially resulting in non-uniform distribution and localized performance variability.

Accordingly, the innovation of this study lies not in the isolated use of crumb rubber, GCP binders, or 3D printing individually but in their systematic integration into a hybrid material system for additive manufacturing. This work specifically combines ELT-derived crumb rubber with a ternary gypsum–cement–pozzolan matrix to optimize the balance between rheology, dimensional stability, mechanical performance, and long-term durability. Through this integrated framework, the study addresses an underexplored intersection between sustainable binder chemistry, recycled waste utilization, and durability-oriented 3D printing applications. This also indicates a potential future pathway for using crumb-rubber-modified GCP mixtures in functionally graded 3D-printed elements, where rubber-rich zones could be selectively introduced in compositions requiring higher ductility, impact resistance, or durability [35,36].

The fundamental insight of this study is that crumb rubber in a printable GCP matrix should not be interpreted only as a lightweight recycled filler. Instead, it acts as a mechanically compliant and hydrophobic phase that changes the balance between strength, deformation tolerance, moisture transport, and freeze–thaw resistance. This behavior is particularly relevant in 3D-printed systems, where matrix cohesion, particle suspension, and interlayer stability must be achieved simultaneously.

The aim of this research was to systematically evaluate whether a crumb-rubber-modified GCP binder can function as a mechanically stable, dimensionally reliable, printable, and freeze–thaw-resistant composite for 3D-printed structural applications. Prism specimens were tested for flexural and compressive strength, as well as deformation with micrometers for curing-age monitoring of dimensional stability [37,38]. Water absorption tests were conducted prior to freeze–thaw cycling to evaluate pore saturation [39,40]. Two complementary approaches were employed to evaluate durability: standardized freeze–thaw testing of cube specimens (with XF1 exposure class performance) and long-term UPV monitoring of 3D-printed samples subjected to thermal cycling. The two-method approach allows both macroscopic property testing and microstructural non-destructive evaluation, jointly providing a comprehensive perspective on performance [13,28,32].

By placing GCP–rubber composites at the intersection of additive manufacturing, sustainability, and durability research, this study contributes to expanding the construction material palette towards a low-carbon future. Specifically, it illustrates that recycling end-of-life tires into 3D-printed GCP binders can reduce environmental footprint, enhance freeze–thaw performance, and maintain structural integrity [2,19,22]. By doing so, the project not only tackles two pressing environmental challenges—tire waste and cement-related emissions—but also advances the broader adoption of circular economy practices in construction [2,5,6].

Recent advances in 3D printing of cementitious materials have moved beyond simple extrusion feasibility toward controlled material placement, structural optimization, and multifunctional performance. In particular, functionally graded 3D-printed concrete has recently been proposed as a strategy for locally tailoring mechanical performance by placing materials with different stiffness, ductility, or tensile capacity within the same printed element. Sun et al. demonstrated this concept in 3D-printed functionally graded concrete plates, where layer-wise material variation improved bending performance and enabled more efficient structural design [41].

Specifically, this study contributes by (1) integrating ELT-derived crumb rubber into a printable ternary GCP binder; (2) optimizing CR/binder ratios to identify a practical balance between strength, rheology, and durability; and (3) validating long-term performance through combined freeze–thaw testing and ultrasonic pulse velocity monitoring.

2. Materials and Methods

2.1. Raw Materials

The binder matrix was developed as a ternary GCP system [3,4,5,6,42] for the purpose of examining its viability for use with recycled crumb rubber from ELTs in common casting and 3D printing applications. The GCP composite has the rapid-setting and low thermal energy demands of gypsum combined with the strength and hydration characteristics of Portland cement and the long-term stability from a reactive pozzolan phase [43].

Table 1. Physical + chemical properties of gypsum, cement, metakaolin, crumb rubber, and admixtures.

Materials	Functional Role in GCP System	Physical Properties	Major Chemical/Compositional Properties *
Gypsum (β-hemihydrate; CaSO4·0.5H2O)	Primary low-carbon binder; rapid setting phase; early-age shape retention	Fine powder; typical specific density: ~2.6–2.8 g/cm3; lower calcination temperature (~150–200 °C); high purity industrial hemihydrate	Dominantly calcium sulfate hemihydrate; principal components: CaO and SO3 equivalent phases; promotes early crystallization and dimensional stability
Portland cement (CEM I/CEM II 42.5 N)	Hydraulic binder; long-term strength development	Fine powder; specific density: ~3.10–3.15 g/cm3; high hydraulic reactivity	Major oxides typically include CaO, SiO2, Al2O3, Fe2O3; responsible for C–S–H formation and structural strength
Metakaolin (calcined aluminosilicate pozzolan)	Pozzolanic additive; matrix densification; secondary hydration	Ultrafine powder; high surface area; specific density: ~2.4–2.6 g/cm3	Rich in reactive SiO2 and Al2O3; promotes secondary C–A–S–H/C–S–H formation; binds free calcium hydroxide and limits excessive ettringite-related instability
Crumb rubber (ELT-derived)	Functional elastomeric aggregate; flexibility; hydrophobicity; freeze–thaw enhancement	Particle size fraction: 0.8–2.5 mm (cast mixes), 2–4 mm (3D mix); low density (~1.10–1.20 g/cm3); elastic, hydrophobic surface	Predominantly vulcanized polymeric rubber (styrene-butadiene/natural rubber blends depending on source); chemically inert relative to binder hydration; acts primarily through physical and interfacial mechanisms
Polycarboxylate superplasticizer	Rheology control; flowability improvement; extrusion consistency	Liquid/powder admixture depending on supplier; low dosage	Disperses binder particles, reduces water demand, improves printability
Set retarder (citric acid/gypsum retarder powder)	Setting-time regulation; printable open time extension	Low-dosage additive	Delays gypsum crystallization kinetics and improves workable time window

* Chemical compositions are based on manufacturer datasheets and standard material classifications where full laboratory oxide analysis was outside the scope of the present study.

summarizes the principal physical and chemical properties of the constituent materials used in the GCP composite system, providing key parameters relevant to hydration behavior, rheology, and durability.

The complete chemical composition profiles of all raw materials are available in the respective manufacturer datasheets and technical documentation. However, for clarity and relevance to binder chemistry,

Table 2. Typical major oxide chemical composition (%) of gypsum, Portland cement, and metakaolin used in the GCP binder system.

Material	SiO2 (%)	Al2O3 (%)	CaO (%)	SO3 (%)	Fe2O3 (%)
Gypsum	1–3	0–1	32–33	45–47	0–0.5
Portland cement	19–23	4–7	60–67	2–4	2–5
Metakaolin	50–55	40–45	0–2	0	0.5–2

summarizes only the principal major oxide components of the three primary mineral binders—gypsum, Portland cement, and metakaolin—which are most directly associated with hydration behavior, pozzolanic reactivity, and long-term performance of the GCP system. These major constituents are presented to provide a concise comparative overview of the fundamental chemical framework governing composite formation.

Crumb rubber aggregate, which is recycled from mechanically shredded ELTs, was employed as partial fine aggregate (sand) replacement. More than one billion waste tires are manufactured annually, which is a critical environmental issue [22,44,45]. Crumb rubber fractions between 0.8–2.5 mm were utilized, which was best practice in previous concrete and mortar investigations [23,25,28,46]. Crumb rubber was supplied by a company located in Ādaži, Riga District, Latvia, while the original material was produced and delivered by the Polish recycling company [47]. Rubber contributes elasticity, crack resistance, and freeze–thaw resistance but tends to compromise compressive strength [27,45].

From this perspective, crumb rubber can be considered not only as a recycled aggregate but also as a functional phase capable of modifying local stiffness, ductility, damping capacity, and freeze–thaw resistance in printed cementitious composites.

Table 3. Functional relevance to rheology, mechanical performance, and durability.

Material	Rheology/Printability Influence	Mechanical Influence	Durability Influence
Gypsum	Rapid setting improves shape retention and layer retention	Supports early stiffness but may be brittle alone	Can exhibit dimensional stability when balanced with cement–pozzolan system
Portland cement	Supports cohesive paste structure	Major contributor to compressive strength	Improves long-term structural stability
Metakaolin	May increase water demand due to fineness; improves particle packing	Densifies matrix and partially compensates ITZ weaknesses	Reduces permeability; improves freeze–thaw and microstructural durability
Crumb rubber	Requires controlled rheology to avoid segregation; improves particle suspension when optimized	Reduces compressive strength due to lower stiffness; improves ductility, crack arrest, and energy dissipation	Hydrophobicity reduces water absorption; elasticity improves freeze–thaw resistance
Superplasticizer	Enhances extrusion and slump-flow control	Indirectly supports uniformity	May reduce voids through better dispersion
Retarder	Extends printable window; prevents premature stiffening	Supports stable layer deposition	Improves process consistency rather than direct durability

summarizes the expected functional contribution of each constituent to rheology, mechanical behavior, and long-term durability.

For maximum workability, a polycarboxylate-based superplasticizer and a minimal dosage of citric acid (set retarder) were incorporated, as per previous GCP additive research [21,48,49].

In addition to citric acid, a powdered gypsum set-retarding admixture (Gips RE-TARD) was used to control the setting time of gypsum-based systems. Gips RETARD is a homogeneous white powder specifically designed to regulate the setting kinetics of gypsum binders. According to the manufacturer’s specifications, the material has a pH of 5 ± 1, making it compatible with gypsum–cement composite matrices and suitable for improving workability and rheology in GCP systems.

2.2. Methods

2.2.1. Mix and Specimen Preparation

Raw materials were weighed by mass, taking into account the proportions summarized. Mixing was provided in laboratory planetary mixer. Mix duration was 3 min in rotation speed 80 r.p.m. Samples 40 × 40 × 160 mm were formed and cured for 1 day in formworks, ensuring surface protection against evaporation. Furthermore, the samples were demolded and cured in moist air condition (RH > 95%) and temperature 21 + 1 °C. Samples were tested at the ages of 7 and 28 days. Curing of all the samples was performed in a controlled-environment chamber at 21 ± 2 °C and 95 ± 5% relative humidity for 28 days. This provided sufficient hydration and pozzolanic activity in the composite before mechanical and durability tests.

For deformation control, special molds equipped with embedded steel anchors were used (Figure 1a). A micrometric strain gauge was employed to monitor deformations during testing (Figure 1b).

The rheology of the mix was examined through a slump-flow test (EN 1015-3 [40]) to ensure that the diameter of flow was between 160 mm and 180 mm, which is described as ideal for layer retention and extrusion [1,9,50,51]. This relatively stiff rheological range was also intentionally selected to minimize density-driven particle migration or flotation of crumb rubber during mixing, extrusion, and layer deposition by maintaining sufficient yield stress for particle suspension. Two forms of specimens were fabricated for testing: cast prisms of size 40 × 40 × 160 mm for flexural and compressive strength and deformation monitoring [38] and plates of size 100 × 100 × 50 mm for freeze–thaw testing in compliance with EN 12390 [52].

2.2.2. Mechanical Testing

Mechanical strength was determined according to EN 196-1 [38], using normal prisms of size 40 × 40 × 160 mm for all the series. The regime of testing was such that strength and deformation values were measured quantitatively, both indicating combined effect on reduction in stiffness as well as enhanced ductility in the GCP composite system. Flexural strength was assessed through a three-point bending test on a universal testing machine at a loading rate of 50 ± 10 N/s and distance between supports 130 mm. The failure load was the measure used to calculate flexural strength and was a direct indicator of the ability of the sample to withstand tensile stress and cracking. Subsequent compressive strength was then tested following flexural testing on both halves of the failed prism specimens. Controlled loading (2.4 ± 0.2 kN/s) to failure was carried out with a hydraulic press, with tests. This method ensures the two parameters (flexural and compressive) for each set of specimens. The compressive strength is the greatest indicator of the structural stability and performance of the binder, particularly in the case of manufacturing 3D printed elements [1,8,10,53].

2.2.3. Water Absorption

Water absorption (WA) tests were crucial in ascertaining pore structure, moisture transport behavior, and saturation capacity, all of which are critical parameters of long-term durability, particularly freeze–thaw resistance [13,28,45].

Prior to testing, all the samples were oven-dried at 70 ± 2 °C to constant weight to eliminate all physically adsorbed water. The samples were then cooled to room temperature in a desiccator to prevent rapid rehydration from atmospheric humidity. All of the samples were weighed immediately to find their dry weight (m₁). Subsequently, the samples were fully immersed in deionized water at 21 ± 2 °C for capillary saturation for 24 h. Surface water was gradually removed using a moist cloth after immersion, and saturated mass (m₂) was recorded using an analytical balance with accuracy of ±0.01 g.

Water absorption (WA) was calculated according to Equation (1):

$$WA\left(\mathrm{\%}\right)={\displaystyle \frac{m_2-m_1}{m_1}}\times 100$$

(1)

This value is the percentage of water absorbed against dry specimen mass. Open porosity (Pₒ) was determined as a ratio of the volume of the absorbed water against the bulk volume of the sample, computed from its geometry and saturated-surface-dry mass.

All water absorption and porosity measurements were carried out using three replicate specimens for each mixture. Mean values together with standard deviations were reported to indicate experimental reproducibility. Combined analysis of water absorption, open porosity, and saturation indices provided a framework for evaluating long-term performance under cyclic environmental loading. In particular, the correlation between reduced porosity and stable ultrasonic pulse velocity (UPV) values (Section 2.2.4) confirmed the microstructural stability of the GCP–rubber matrix after extended freeze–thaw exposure.

Overall, the reduced water absorption capacity observed in rubber-modified GCP composites indicates improved durability performance. The combined effects of rubber hydrophobicity, elastic particle behavior, and matrix densification contribute to limiting moisture-related deterioration mechanisms, supporting the suitability of the developed GCP–crumb rubber system for durable and sustainable 3D-printed construction applications.

2.2.4. Freeze–Thaw Durability Testing

Freeze–thaw durability of the GCP–crumb rubber composites was evaluated using a combined testing approach that included both destructive and non-destructive methods in order to assess macroscopic degradation and microstructural stability during long-term exposure. This dual methodology allowed the correlation of visible physical deterioration (e.g., surface scaling and mass loss) with internal structural changes detected through ultrasonic monitoring.

The primary durability test followed EN 12390 [52] adapted for cementitious composites corresponding to the XF1 exposure class (moderate freeze–thaw exposure under humid conditions) [13,28]. Plate specimens with dimensions 100 × 100 × 50 mm were used. After 28 days of curing at approximately 99% relative humidity, the specimens were removed and dried under laboratory conditions in preparation for water absorption measurements and subsequent freeze–thaw testing. Prior to cycling, specimens were partially immersed in water to a depth of 5 mm using spacers to enable capillary saturation. The samples were then subjected to repeated freezing–thawing cycles between –20 ± 2 °C and +20 ± 2 °C, with approximately two cycles per day. During testing, surface scaling, mass loss, and visual deterioration were systematically recorded (Figure 2). Specimens were inspected periodically, and residual compressive strength was determined after completion of the freeze–thaw sequence.

To complement the destructive testing, ultrasonic pulse velocity (UPV) measurements were performed to monitor internal microstructural changes in 3D-printed GCP specimens during long-term freeze–thaw exposure up to 200 days. UPV measurements were carried out using a portable ultrasonic device (Figure 3) equipped with 54 kHz transducers, in accordance with EN 12504-4:2021 [1,32,37,54,55].

Results showed that the control mixture (CR0) exhibited moderate surface roughness after extended cycling, while rubber-modified mixtures—particularly CR2 and the optimized 3D-Mix—retained smoother surfaces with minimal visible degradation. Mass loss remained below 0.5%, satisfying durability expectations for XF1 exposure. The optimized 3D-Mix (approximately 25% CR/binder) retained more than 95% of its initial compressive strength after 50 freeze–thaw cycles, confirming the superior durability of the rubber-modified system. This improved resistance is primarily attributed to the hydrophobic and elastic properties of crumb CR, which act as micro-shock absorbers within the hardened matrix. Their presence reduces internal stress generated by the ~9% volumetric expansion of water during freezing, while the elastic interfaces help dissipate energy and limit crack propagation.

Each specimen was subjected to the same temperature cycle (–20 °C to +20 °C), and measurements were collected periodically (every 7–10 days) along diagonal paths on multiple faces of the specimen to detect possible anisotropy caused by the layered printing process. The stabilization of UPV values during long-term cycling indicates that microcrack propagation remained limited, supporting observations from previous studies on rubber-modified cementitious materials [28,32,56].

The elastic recovery of rubber inclusions likely prevented cumulative microdamage by redistributing stress within the matrix. Consequently, the combined application of standardized freeze–thaw testing and UPV monitoring provided a comprehensive evaluation of both surface durability and internal structural stability of the GCP–rubber composite system.

The experimental design followed durability-oriented investigations of recycled crumb-rubber-modified mortars reported in previous literature [19,25,29,57]. Prior to testing rubber-modified mixtures, the reference CR0 mixture was evaluated to establish a baseline performance, showing an average 28-day compressive strength of 24.14 MPa. Based on performance criteria, the acceptable lower limit for modified mixtures was defined as 50% of the reference strength (≥12 MPa) to ensure suitability for non-structural or infill 3D-printing applications. Among the tested compositions, crumb rubber contents in the range of 17–50% of binder mass provided the best balance between strength retention, flexibility, crack resistance, and freeze–thaw durability.

Dimensional stability during curing was also monitored using high-precision micrometers with 0.001 mm accuracy, with periodic measurements performed under immersion conditions to quantify longitudinal strain resulting from shrinkage or expansion effects [9,44].

2.2.5. Quantitative Rheology and Layer Retention Assessment

To strengthen the additive manufacturing relevance of the study, rheology and layer retention of the optimized 3D-Mix were quantitatively evaluated using practical engineering metrics relevant to extrusion-based construction. Rheology was primarily controlled through slump-flow testing (EN 1015-3 [40]), targeting a flow diameter of 160–180 mm to balance extrudability, shape retention, and segregation resistance. Layer retention was assessed through dimensional retention of printed layers, structural stability during multilayer deposition, and filament continuity under continuous extrusion conditions [58]. Quantitative indicators included slump-flow diameter, target versus achieved layer height, height-retention ratio, and maximum stable number of deposited layers without collapse or major geometric deformation description in

Table 4. Quantitative printability and buildability indicators.

Parameter	Target/Criterion	Observed Performance
Slump-flow diameter	160–180 mm	170 mm
Nozzle diameter	20 mm	20 mm
Designed layer height	10 mm	10 mm
Measured layer height	8–12 mm	10 mm
Height-retention ratio	>90%	100%
Maximum stable printed layers	≥10	20
Filament continuity	No blockage/collapse	Achieved

.

2.3. Mix Design and Specimen Preparation

The baseline mechanical and durability properties of the ternary GCP binder were first obtained by preparing a reference mix without crumb rubber (CR0). Subsequently, three additional mixtures were prepared with crumb rubber contents of 17%, 33%, and 50% relative to the binder mass in

Table 5. Mass proportion composition of GCP mixtures with ELT crumb rubber.

Materials	Units	CR0	CR1	CR2	CR3
Gypsum binder	%	39.40	34.78	31.17	28.24
Portland cement	%	16.12	14.23	12.75	11.55
Metakaolin	%	16.12	14.23	12.75	11.55
Retarder powder	%	0.28	0.25	0.23	0.21
Polycarboxylate superplasticizer	%	0.14	0.13	0.11	0.10
Crumb rubber (ELT—0.8–2.5 mm)	%	-	10.56	18.90	25.68
Water	%	27.94	25.82	24.09	22.67
CR/Binder ratio	%	-	17	33	50

. Crumb rubber was incorporated as an additional functional aggregate phase while maintaining constant gypsum–cement–pozzolan binder proportions to isolate rubber-specific effects on composite performance. The addition of ELT-derived crumb rubber was therefore intended not to reduce the solid fraction of the mixture but rather to introduce an additional aggregate phase within the composite. The mix compositions presented in Table 5 were initially measured in grams and are additionally expressed as mass proportions (%) to facilitate comparison between mixtures. Addition of crumb rubber to work as a functional component would enable its properties, like elasticity, energy absorption, and crack arresting, to contribute to the performance of the composite, especially in freeze and thaw cycling [59]. In addition, the aggregate volume is preserved, and the replacement of the fine aggregate component in the mixture with the crumb rubber ensures a similar packing density, thus preventing the occurrence of high void content, which would have been detrimental to the mechanical properties of the mixture. The method also ensures an accurate assessment of the effect of the incorporation of the crumb rubber on the mechanical, physical, and durability properties, thus following sustainable material design concepts for the recycling of ELT materials [60]. This approach was followed to better evaluate the direct influence of CR on workability, density, and mechanical response in lieu of substitution effect of aggregates.

The selected crumb rubber contents (17%, 33%, and 50% of binder mass) were intentionally designed to represent a broad exploratory compositional range rather than exclusively application-targeted formulations. This interval-based approach was chosen to establish lower-, intermediate-, and upper-bound performance thresholds, allowing systematic identification of mechanical, rheological, and durability trade-offs associated with progressively increasing rubber incorporation. In particular, the 50% CR/binder composition was included as an upper feasibility boundary to determine the practical performance limit beyond which strength reduction may outweigh durability benefits.

The 3D printable mix design was optimized to provide a printable and durable GCP composite through the incorporation of ELT crumb rubber as a green additive rather than sand replacement. The composition was resolved in initial tests (Section 2.2.2) and further evolved for extrusion behavior, strength balance, and dimensional stability. Based on these findings, the final optimized composition for printing 3D-Mix with 25% crumb rubber in terms of 3D-Mix compositions presented in

Table 6. Mass proportion composition mixes for 3D-Mix.

Materials	Units	3D-Mix
Gypsum binder	%	33.04
Portland cement	%	13.52
Metakaolin	%	13.52
Retarder powder	%	0.03
Crumb rubber 2–4 mm (ELT)	%	15.02
Polycarboxylate superplasticizer	%	0.24
Water	%	24.63
CR/Binder ratio	%	25

were initially measured in grams and are additionally expressed as mass proportions (%) to facilitate comparison between mixtures [45,49,53]. Based on the broader exploratory matrix, the final 3D-Mix (~25% CR/binder) was selected as the most practically relevant formulation because it maintained sufficient rheology and mechanical performance while achieving meaningful durability enhancement.

Additionally, the metakaolin pozzolanic reaction has a complementary function of reacting with excess free calcium hydroxide and generating additional C–A–S–H phases that densify the matrix and reduce permeability. This mutual association between the dense pozzolanic gel and hydrophobic CR results in a dual barrier system against water intrusion. Therefore, hybrid GCP–rubber composite achieves a minimized water-to-solid interaction ratio, balancing permeability control and mechanical flexibility—features that distinguish it from regular Portland cement mortars.

Successful deposition of 10-, 15-, and 20-layer geometries without significant collapse demonstrated sufficient structural layer retention and filament shape stability for additive manufacturing applications.

The solid binder proportion consisted of a fixed composition of gypsum:cement:metakaolin = 55:22.5:22.5 (w/w), similar to earlier durability-enhanced ternary systems [6,13,61,62]. Crumb rubber was incorporated into the binder matrix in a quantity equal to 25% of the total binder mass. This material was identified as best during preliminary trials, demonstrating good compressive and flexural strength but significantly improving freeze–thaw resistance and elasticity. Water-to-binder ratio between 0.40 and 0.45 was defined as optimal for workability and extrudability, demonstrating proper lubrication of particles and preventing segregation or nozzle clogging during extrusion [8,9,48]. The retarder powder regulated setting time, creating a workable open window of time for round-the-clock 3D printing (Figure 4), while the polycarboxylate-based superplasticizer enhanced flow uniformity without sacrificing shape stability [63]. The materials were dry blended for 2 min in a planetary mixer to achieve uniformity in the binder matrix prior to the addition of water and admixtures. The crumb rubber was added at the wet-mixing stage to achieve proper dispersion and reduce entrapped air. Both batches were mixed until they achieved a consistent, uniform consistency with a smooth surface when viewed and uniform distribution of particles. Two forms of specimens were fabricated for testing: cast prisms of size 40 × 40 × 160 mm for flexural and compressive strength and deformation monitoring [38] and plate specimens of size 100 × 100 × 50 mm for freeze–thaw testing in compliance with EN 12390 [52]. For additive manufacturing testing, samples were 3D-printed with a gantry-type screw extruder with a nozzle size of 20 mm.

The printer was operated under continuous extrusion at a fixed speed of 300 mm/s with a 10 mm layer thickness and 50% interlayer overlap. Four wall-type specimens were printed in ten, fifteen, and twenty layers from 100 mm up to 2000 mm total height with two shapes. Interlayer bond and layer retention were visually inspected upon printing to ensure that a new layer could be deposited without collapse or gross deformation. Healing of all the samples was performed in a controlled-environment chamber at 21 ± 2 °C and 95 ± 5% relative humidity for 28 days. This provided sufficient hydration and pozzolanic activity in the composite before mechanical and durability tests. After curing, some samples were submerged in water for testing dimensional stability upon saturation, and others were subjected to freeze–thaw cycles for mimicking conditions of exposure in the environment [12,33]. Mixing, casting, and printing procedures were followed as a reproducible protocol. Experimental batches were mixed in triplicate to ensure minimal variation, and mass ratios were verified with an accuracy of ±1 g. Data gathered at this phase were utilized as input to correlate with the subsequent mechanical and durability test data (Section 2.2.3). This hybrid preparation process was designed to simulate true conditions for automatic extrusion printing of green, recyclize-containing rubber GCP materials. The chosen blend provided sufficient cohesiveness and thixotropy for layer stability, and the high binder content and ideal crumb-rubber ratio ensured excellent freeze–thaw stability and energy-absorbing capability under load. Blends were prepared with 25% crumb rubber added to the mixes with a constant binder proportion G:C:P = 55:22.5:22.5 (by weight) [64,65]. The composition was based on previous work proving improved durability in ternary blends [13,46]. Water/binder ratio ranged between 0.40 and 0.45, which was the best for extrudability and to avoid segregation [8,9,66]. Test samples were cast prisms (40 × 40 × 160 mm) for flexural and compressive strength and deformation monitoring [38] and cubes (100 × 100 × 100 mm³) for freeze–thaw durability according to EN 12390 [52].

First, 3D-Mix printed specimens, made in a gantry-type screw extruder and with a 20 mm nozzle, were calibrated to obtain slump flows between 160–180 mm (EN 1015-3 [40]), with layer retention in line with 3D printing mortar requirements [1,9,53,67]. Curing was carried out in laboratory conditions (21 ± 2 °C, 95 ± 5% RH) for 28 days, followed by water immersion or freeze–thawing for durability testing [13,38].

Accordingly, the highest rubber content should be interpreted primarily as a boundary-condition reference mixture for mechanistic comparison rather than as a directly recommended practical structural composition.

3. Results and Discussion

3.1. Mechanical Strength

Together, these sets of data offer a combined view of the impact of the introduction of rubber on mechanical performance of samples in cast and additively processed GCP materials in

Table 7. GCP mixes with ELT crumb test results.

	Units	CR0	CR1	CR2	CR3	3D-Mix
Flexural strength, 28 days’	MPa	4.78	3.17	2.99	1.95	3.11
Compressive strength, 28 days’	MPa	24.14	15.80	9.08	5.42	12.37
Density of samples	kg/m3	1758	1635	1545	1445	1553

. Statistical reliability of the data was provided through triplicate testing with no differences larger than ±5% for all results given.

The results suggest that the role of crumb rubber in the GCP matrix is multifunctional. The reduction in compressive strength is mainly associated with the lower stiffness of CR and the formation of a less homogeneous ITZ. However, the same elastomeric phase appears to improve deformation tolerance, crack-arrest behavior, and freeze–thaw resistance by redistributing local stresses and reducing capillary water transport. Therefore, the main scientific insight is the identification of a performance trade-off in which crumb rubber reduces compressive capacity but improves durability-related behavior when incorporated into a pozzolan-densified printable GCP matrix.

Beyond simple density reduction, crumb rubber in the GCP matrix may act as a deformable elastomeric phase that modifies composite behavior through multiple interacting mechanisms. First, the lower stiffness and elastic recovery of CR may locally redistribute stress concentrations, reducing brittle crack propagation under flexural or freeze–thaw loading. Second, the hydrophobic surface of rubber may partially interrupt capillary pore continuity, thereby reducing water ingress pathways. Third, rubber inclusions may function as localized crack-arrest or energy-dissipation zones, particularly when embedded within a pozzolan-densified matrix. At the same time, the reduced stiffness and potentially weaker interfacial transition zone (ITZ) between rubber and binder likely contribute to compressive strength loss. In this context, metakaolin-driven secondary hydration may partially compensate by densifying the surrounding GCP matrix and improving bulk structural cohesion.

The experimental campaign targeted the assessment of the physical, mechanical, and durability performance of the novel prepared GCP composites modified by CR derived from ELTs. The flexural strength results of the four specimens of each mix are presented in Figure 5.

Although CR3 (50% CR/binder) demonstrated limited structural practicality because of pronounced compressive strength loss, its inclusion provided important parametric insight into the upper threshold of elastomeric modification and helped define the optimization window for more practically viable formulations.

The outcomes of the evolution of properties in the early cast mixtures (from CR0 to 3D-Mix) showed typical workability and shape elasticity after extrusion with a quite smooth surface finish and uniform layer development (Figure 6) at the cracking place after flexural tests. These aspects are meant to demonstrate that rubber addition did not compromise rheology of the GCP system, which is a key finding since such polymeric additions destabilize the extrusion rheology [21,32]. The mixture then demonstrated a balance of cohesion and flowability appropriate for use as an environmentally friendly construction material in automated manufacturing.

The mechanical performance results in the expected equilibrium between strength and ductility of rubberized cementitious materials as typically witnessed [22,25]. The reference GCP mix of CR0 featured highest compressive strength, whereas rubber-modified ones (CR1–3D-Mix) showed stepwise reduction. Nevertheless, optimized 3D-Mix preserved compressive strength at over 50% of the reference, a level often desired for light structural composites [24]. Under flexural loading, the decrease was less. While flexural strength decreased moderately, test specimens exhibited a distinctly different fracture mode, from brittle failure of the control mix to ductile bending accompanied by visible energy dissipation. Crack extension was reduced as the rubber inclusions served as bridges between microcracks and stressed more evenly the GCP matrix. This confirms the elastic damping property of rubber and aligns with similar observations in cement–rubber composites [19,27,30]. As an important note, retention of flexural capacity while enhancing toughness is a valuable mechanical advantage of the GCP–rubber system over conventional cementitious mix.

Compressive test results of these initial mixes are presented in Figure 7, revealing clear trends. Increasing the proportion of crumb rubber led to enhanced freeze–thaw resistance and improved flexural performance, while simultaneously causing a moderate reduction in compressive strength—a behavior that aligns closely with previously reported findings in rubber-modified cementitious composites [44,45,49].

Based on these experiment results, the final mix design was established with the crumb rubber content at 25% of the total binder weight. This proportion is exhibited in Table 4 of the optimum balance between mechanical strength as well as adequate workability and rheology for 3D extrusion.

This ratio gave a workable and cohesive-like blend with enhanced flexibility and strength without lowering print stability and interlayer adhesion throughout the extrusion process. Crumb rubber was produced by shredding ELTs using mechanical means sourced from a reliable recycling facility. Particle size range was 0.8 to 2.5 mm, consistent with earlier studies of increased elastic energy dissipation and hydrophobicity in com-parable size ranges [11,12,13,14,47,59,60,61,62,64,68]. Rubber utilization as a functional additive in GCP systems presents potential dual benefits: mechanical damping through elastomeric inclusions and pore blocking to limit water absorption. In order to regulate rheological behavior, a polycarboxylate-based superplasticizer was applied to every mixture [47]. The admixture lowered the yield stress of the mix and enhanced extrudability for the 3D printing process [46]. In addition, a citric-acid-based retarder powder was added in order to adjust setting time and prevent premature stiffening during mixing and extrusion [7,42]. Both additives were dosed cautiously in order to produce uniform layer retention and structural integrity for all mix designs.

3.2. Density and Water Absorption

There was a clear trend in the physical properties of the composites. Incorporation of crumb rubber successively reduced bulk density, Figure 8, consistent with the low density of rubber granules versus mineral binder. Although modest, this reduction provided a lightweight composite enhancing handling efficiency and structural dead load reduction.

Before tests were conducted on the rubber-modified blends, the control GCP mixture (CR0) was tested in order to determine a baseline performance. The control registered an average 28-day compressive strength of 24.1 MPa. All other tests were thereafter compared mechanically against this benchmark.

Water surface capillary absorption testing data, shown in

Table 8. Surface capillary water absorption (g/m²) as a function of exposure time (hours).

	Units	Hours
		0.5	1	2	4	8	24	48	72
CR0	g/m2	40	52	64	86	98	122	159	177
CR1	g/m2	45	57	70	81	97	123	156	174
CR2	g/m2	33	41	50	66	71	100	131	148
CR3	g/m2	36	46	58	65	78	109	138	155
3D-Mix	g/m2	36	46	53	70	79	94	124	140

, also confirmed the beneficial influence of crumb rubber on composite properties related to durability. As seen in Figure 9, rubberized samples consistently had less water absorption than the control.

This is attributed to the hydrophobicity of rubber that resists capillary transportation of water through the pore network. Utilization of rubber most likely disrupted capillary continuity, therefore reducing permeability. The same has been seen in polymer-modified cement composites [20,48,69].

Lower water absorption is desirable under long-term service conditions since it directly relates to improved freeze–thaw and chemical resistance. The combination of lower absorption and good dimensional stability suggests that the GCP–rubber material will be more resistant to moisture-induced degradation than in usual gypsum- or cement-based mortars.

Crumb-rubber-modified mixes were expected to show less water absorption due to the hydrophobic and elastic nature of CR, which delivers internal pore-blocking effect and inhibit capillary suction. The mechanism has been reaffirmed in several previous studies of rubberized cement mortars and concretes [23,27,29,66,67,70]. The crumb rubber surface characteristics, as well as its moderate hydrophobic attraction to water, inhibit pore continuity, leading to fewer ingress routes for water.

In the current research, the incorporation of crumb rubber resulted in significantly lower apparent porosity and absorption when compared with control GCP mixture (CR0). Reduction was greatest in the case of the 3D-Mix (25% CR/binder ratio), which had the lowest observed absorption of the entire series of mixes in

Table 9. Water absorption and open porosity test results.

	Units	CR0	CR1	CR2	CR3	3D-Mix
Water absorption	%	2.2	2.2	1.9	2.0	0.9
Open porosity	%	3.9	3.6	2.9	2.9	1.4

. This phenomenon suggests that rubber addition not only reduces the overall pore volume accessible but also modifies internal pore structure by introducing discontinuous, closed micro-voids that trap air and restrict fluid flow.

This refinement of the microstructure provides a direct route to enhanced freeze–thaw resistance, as interior water is less able to expand in closed pores during freezing and thus reduces internal stress build-up and cracking.

Overall analysis of water absorption, open porosity in Figure 10, and saturation indices provided a broad platform for the estimation of long-term performance under cyclic environmental loading. Specifically, the congruence of low porosity with good UPV values in Section 2.2.3. confirmed the microstructural integrity of the GCP–rubber matrix after long-term freeze–thaw aging [65].

In total, the lowered absorption capacity and increased pore structure of the rubberized GCP composites are a fundamental enhancement in durability and life. The hydrophobicity, elasticity, and matrix densification synergy effectively resists moisture-triggered degradation mechanisms, situating the developed GCP–crumb rubber mix as a robust, environmentally friendly material for 3D-printed structural and non-structural members under aggressive environmental conditions.

In extrusion-based 3D printing, pore structure must also be considered beyond bulk porosity alone, since layer-by-layer deposition may introduce anisotropic pore distribution, filament-interface voids, and preferential transport pathways distinct from cast systems. Previous studies on 3D-printed cementitious materials have shown that interlayer interfaces often contain higher localized porosity than monolithic cast specimens, even when overall mixture composition is similar. Therefore, the reduced water absorption observed in the present study likely reflects both material-level hydrophobicity and rheology-controlled filament stability, while localized interfacial porosity may still remain a critical durability parameter.

Although the rubber–binder interface appears less homogeneous than the surrounding mineral matrix, large interfacial voids or severe debonding were not prominently observed at the examined scale. This suggests that crumb rubber likely functions as a mechanically distinct inclusion within a relatively continuous binder structure, where localized ITZ heterogeneity may explain the simultaneous reduction in compressive strength and enhancement in flexibility, crack-arrest potential, and freeze–thaw resistance.

EM observations, Figure 11, provide qualitative evidence of crumb rubber encapsulation within the gypsum–cement–pozzolan matrix and indicate the presence of an identifiable interfacial transition zone (ITZ) on the order of tens of micrometers (~56 µm).

The observed microstructure supports the hypothesis that metakaolin-assisted matrix densification may partially stabilize the rubber–binder interface, while the elastomeric inclusion modifies local stress transfer rather than acting solely as a defect phase.

Because only qualitative SEM observations were conducted, further studies using higher-resolution SEM mapping, EDS, or X-ray CT would be required to quantify pore connectivity, chemical bonding, and ITZ evolution more rigorously.

3.3. Dimensional Stability and Frost Resistance

This stability was maintained even in mixtures with high rubber content. The result is noteworthy because cementitious systems typically exhibit drying shrinkage, whereas gypsum-based systems may show slight expansion during hydration. In the present GCP system, the combined presence of gypsum and rubber appears to contribute to a balanced deformation response. However, the mechanism cannot be stated unequivocally, as gypsum also reacts with cement aluminates to form ettringite, a phase associated with volumetric expansion. Therefore, the observed dimensional stability is likely the result of complex interactions between binder hydration reactions and the elastic behavior of the rubber inclusions. Comparable behavior has not been consistently observed in previous studies on rubber-modified binders [13,38,43], highlighting a potentially unique microstructural interaction that enhances the dimensional integrity of printed parts.

Dimensional stability testing revealed another distinctive and favorable trend. Precision micrometers deformation monitoring revealed zero longitudinal strain for all mixes during immersion curing, and the net displacement was below measurement sensitivity limits shown in Figure 12.

The combined results of both tests show that addition of 25% crumb rubber in 3D-Mix significantly enhanced freeze–thaw resistance, with stable performance even after repeated cycles of thermal cycling. The composite showed superior high residual strength, minimal mass loss, and extremely low ultrasonic attenuation, substantiating the synergistic effect of rubber elasticity, pozzolanic densification, and hydrophobicity.

In practice, these results indicate that the 3D-Mix-printed GCP–rubber composite synthesized here possesses sufficient strength for architectural and structural applications under different temperature and humidity conditions. The use of UPV monitoring also validates its applicability for in situ quality control and long-term condition monitoring, opening up opportunities for the application of sustainable, waste-based materials in new building technologies.

The pre-exposure UPV values of the samples averaged ~2600 m/s, indicating good density and good interfacial bonding of the layers during printing. Upon 36-day cyclic exposure, the UPV values dropped a small amount by around 50 m/s as a result of development of microcrack growth in interfacial transition zones. After 36 days, however, UPV values were very stable until day 365, indicating very minimal progressive degradation in

Table 10. Freeze–thaw resistance indicators (UPV changes) for CR0 and 3D-Mix specimens.

	Units/Cycles	0	36	49	85	156	232	365
CR0	m/s	2787	2739	2708	2539	2266	n.t.	n.t.
3D-Mix	m/s	2445	2406	2399	2340	2404	2391	2395

. The stability of ultrasonic velocity with long exposure time confirmed that the internal structure was coherent with no material continuity or bond quality loss between printed layers.

The freeze–thaw performance was the most convincing outcome of the research. Under controlled XF1 cycling conditions (EN 12390 [52]), all the rubber-containing specimens showed better resistance to surface scaling and internal degradation; Figure 13 shows a UPV value comparing CR0 and 3D-Mix. The Figure 14 chart shows results of scaling surface mass loss, and even after prolonged exposure, 3D-Mix samples showed no visible damage, no significant mass loss, and superior maintenance of mechanical integrity.

In contrast, the GCP reference mixture showed premature microcracking and surface roughening after cyclic loading. These results confirm that CR is internal stress-relief inclusions that mitigate the expansive pressures that result from ice crystallization in the pore structure. This result is consistent with previous work on rubberized concrete [18,47,50], but confirmation in a GCP binder in 3D-printed form is novel and demonstrates the durability advantage of piezoelectric material under cyclic thermal loading.

Four 3D-Mix specimens were subjected to freeze–thaw durability tests through cyclic freezing and thawing under controlled moisture exposure. As shown in Figure 15, the specimens were partially immersed in water with a depth of 5 mm using spacers to simulate surface scaling. Hydroisolation tapes were applied to the lateral surfaces to limit moisture ingress from these surfaces, so that the degradation mechanisms are dominated by surface scaling due to cyclic environmental exposure.

A close-up comparison of surface degradation for 3D-Mix Sample No. 2 during progressive freeze–thaw cycles is shown in Figure 16. In Figure 16a, it can be seen that the sample has undergone 28 cycles, showing minimal signs of surface scaling. In Figure 16b, it can be seen that the sample has undergone 56 cycles, showing increased signs of mortar degradation according to EN 12390 [52]. The comparison of the sample visually indicates the gradual nature of scaling and resistance to structural degradation.

The UPV monitoring results also completed the freeze–thaw testing of the 3D-printed rectangular specimens made of 3D-Mix material. After the freeze–thaw chamber testing, it was noted that the durability of the printed specimens was unexpectedly high. There was no severe damage in the form of cracks or disintegration of the structure. Contrary to the cast specimens, the printed specimens showed minimal damage. The UPV monitoring results also confirmed the durability of the printed specimens. The condition of the specimens is shown in Figure 17.

This self-stabilization phenomenon, where initial small declines plateau into a consistent long-term trend, is an unusual durability characteristic in additive-processed composites [9,31,51]. The strong correlation between UPV stability and freeze–thaw performance further supports the conclusion that the GCP–rubber system achieves a robust microstructure able to redistribute internal stresses without consequence.

Overall, the mechanical, dimensional, and durability tests make up a uniform performance profile for the GCP–rubber composite. The material shows expected trade-offs—lower density and compressive strength balanced by enhanced flexibility, energy absorption, and freeze–thaw resistance. The trend is consistent with earlier studies on sustainable binders but with a superior balance point due to the synergistic behavior of gypsum and pozzolan in the matrix [8,38,68]. The ability of the composite to stay structurally intact under cycling of the environment, coupled with low-energy binder manufacturing and its lightness, highlights its technological importance. Further, proof of successful 3D rheology using the 3D-Mix mixture is a first step towards green additive manufacturing materials.

Environmentally, this finding is very significant for construction’s circular economy. Utilization of end-of-life tire crumb rubber prevents the waste from landfills and reduces reliance on natural aggregates, and inclusion of gypsum and pozzolan reduces clinker consumption and embodied CO₂ emissions [1,3,57]. The composite so formed unites material recycling and emission reduction in a single technological platform. The low mass and high durability of the 3D-Mix indicate additional energy conservation during transport and service life, further contributing to its sustainability profile.

The research design conformed to durability-targeted studies of recycled crumb rubber modified mortars [19,25,26,29,57], whereby the results could be compared with patterns reported in conventional literature.

3.4. Evaluation of Rheology Characteristics of the Mixtures

Although modest, this reduction provided a lightweight composite enhancing handling efficiency and structural dead load reduction. The 3D-Mix showed typical workability and shape elasticity after extrusion with smooth surface finish and uniform layer development. These aspects are meant to demonstrate that rubber addition did not compromise rheology of the GCP system, which is a key finding since such polymeric additions destabilize the extrusion rheology [21,33]. The mixture then demonstrated cohesiveness and flowability balance appropriate for use as an environmentally friendly construction material for automated manufacturing.

In contrast to conventional cast rubberized systems, where crumb rubber may segregate because of density differences, no visible signs of rubber flotation, localized accumulation, or phase separation were observed during extrusion or in printed layer geometry, Figure 18. The stiff rheological design, rapid structural build-up of the gypsum-containing matrix, and controlled layer-by-layer deposition likely restricted particle migration.

Uniform filament morphology, stable layer dimensions, and visually consistent fracture surfaces collectively indicate effective particle suspension under the selected processing conditions.

Although direct tensile or shear interlayer bond testing was not conducted, coherent layer geometry and stable long-term UPV trends suggest relatively effective filament integration under the selected rheological conditions.

Layer retention tests were performed to see if the 3D-Mix material can keep its shape while layer by layer printing. Figure 19 shows the specimens that we produced during the evaluation of how the 3D-Mix material can be used to build things. We successfully printed shapes with 10, 15, and 20 layers and also a hexagonal shape with 10 layers. These structures looked good with layers and stable shapes and did not have any visible problems.

Therefore, 3D printing was a design constraint, optimization framework, and performance criterion.

This means that the 3D-Mix material provided strength to support the building process, which involves adding layers on top of each other.

Interlayer bonding is a critical performance parameter in 3D-printed cementitious systems because insufficient filament fusion can create weak planes, anisotropic strength loss, and increased pore connectivity. Recent studies on interlayer bonding and microstructural characterization of 3D-printed concrete indicate that filament geometry, deposition interval, and rheological consistency strongly influence bond integrity and pore architecture. In the present work, the stable filament morphology, limited geometric deformation, and consistent UPV trends suggest relatively effective layer integration, although direct bond-strength testing or microstructural imaging was beyond the current scope. Although advanced quantitative segregation analysis such as X-ray CT or digital particle mapping was outside the scope of this study, the consistent visual morphology and repeatable mechanical performance suggest that significant segregation was effectively minimized.

The 3D-Mix material worked well for making layer structures. The printed parts had lines and stuck together well, as shown in Figure 20. Then, we printed shapes with 10 layers to test if the material can hold complex shapes. The hexagons kept their shape with layers and smooth surfaces, which means the material can recover its shape well after being printed out. When we printed taller shapes, the difference between the planned and actual height was not noticed. This shows that the 3D-Mix material is strong enough to support layers without falling or bending too much. These results confirm that our 3D-Mix material is good for 3D printing, keeping its shape and layer structure well during printing.

Based on the strength retention requirements, the allowable lower limit of the modified compositions was established as not less than 50% of the reference strength (i.e., ≥12 MPa), which would provide adequate mechanical performance for non-structural or infill 3D printing. Upon comparison of the results throughout the test series, the crumb rubber content of about 25% per binder mass (gypsum + cement + pozzolan) showed the best balance in performance. This work maintained compressive strength beyond the desired limit while achieving substantial improvement in flexibility, reduced crack formation, and freeze–thaw resistance. The correspondence of mechanical strength with CR/binder ratio achieved supported 17–50% rubber loading as an effective optimization range for 3D-printable GCP composites, as per previous findings on hybrid binder systems [13,15,19,25,28,57,64,66,70].

For UPV and freeze–thaw data, percentage retention of initial UPV, a non-destructive indicator of internal structural integrity, was determined as the relative change in ultrasonic velocity with exposure time. It was contrasted with mechanical property degradation in order to establish whether or not the ultrasonic data may serve as a suitable proxy for the cumulative microstructural damage.

In summary, the findings indicate that a 25% CR/binder ratio offers the best balance of mechanical performance, rheology, and durability. The GCP–rubber system maintained structural cohesion, displayed near-zero deformation, reduced water absorption, and withstood extreme freeze–thaw cycling with negligible damage. The ultrasonic and mechanical agreement verifies the stability of the composite and declares that combining ELT-based rubber into GCP binders is an innovative and sustainable development for future generations of 3D-printed construction materials.

4. Conclusions and Future Outlook

This study demonstrated that GCP composites modified with crumb rubber from end-of-life tires can serve as a durable and sustainable material for additive manufacturing. The experimental program progressed from mixture optimization (CR0–CR3) to the development of a printable composition 3D-Mix. The optimal formulation containing ≈25% crumb rubber by binder mass (3D-Mix) achieved compressive strength values of approximately 12.37 MPa, representing a reduction of about 49% compared with the control mixture (24.14 MPa). Similarly, the flexural strength decreased from 4.78 MPa to 3.11 MPa (≈35% reduction). The main fundamental insight is that crumb rubber acts as a multifunctional phase in the printable GCP matrix, simultaneously weakening compressive load transfer while improving deformation tolerance, moisture resistance, and freeze–thaw durability.

Despite this decrease in strength, the incorporation of CR improved the deformation tolerance and ductility of the composite by promoting a more gradual strain development and crack-bridging behavior. In addition, the density decreased from 1758 kg/m³ to approximately 1553 kg/m³ (≈12% reduction), producing a lightweight composite suitable for non-structural 3D-printed elements while reducing structural dead loads.

Given the material, operational, and reproducibility demands of the technical process, the present study prioritized strategically spaced compositional intervals sufficient for trend identification, while future work may refine narrower parametric increments around the identified optimum.

Therefore, the broader parametric matrix should be viewed as a performance-mapping framework used to identify practical optimization boundaries rather than as an indication that all tested rubber contents are equally suitable for end-use construction.

Although the addition of rubber slightly reduced compressive strength compared with the reference mixture, flexural behavior showed improved toughness and crack-bridging capacity. Dimensional stability measurements revealed very low longitudinal deformation throughout the curing period up to 624 days, remaining within approximately ±0.05–0.10 mm/m (≈±50–100 µε), indicating stable volumetric behavior of the composite matrix. This integrated behavior differs from most conventional Portland cement rubber composites because the ternary GCP matrix introduces gypsum-assisted setting, pozzolanic densification, and additive-manufacturing-specific rheological constraints.

The research findings indicate that while compressive strength was moderately lowered by the introduction of crumb rubber, flexural strength was increased due to the energy-absorbing qualities of the rubber. Water absorption was lower, reflecting the hydrophobic nature of CR, and deformation testing revealed very little dimensional change. Above all, freeze–thaw resistance was dramatically improved: specimens retained more than 95% of their original strength upon cycling, and UPV measurements registered only a slight decrease (~50 m/s after 36 days), leveling off afterwards up to 200 days. These findings confirm that crumb rubber plays a beneficial role in the durability of GCP composites, offsetting the loss in compressive strength. Durability testing confirmed improved resistance to environmental actions. Capillary water absorption was reduced due to the hydrophobic nature of CR, resulting in lower apparent porosity values of approximately 18–22% compared with the reference mixture. Freeze–thaw experiments showed excellent resistance, with minimal surface scaling after 56 cycles and very low mass loss (<0.5–1.0 kg/m²). Ultrasonic pulse velocity monitoring indicated a stable internal structure, with velocities remaining around 2200–2600 m/s and showing only minor reductions during long-term cyclic exposure.

These findings are consistent with recent studies demonstrating that pore distribution and interlayer interface quality are dominant factors governing anisotropy and long-term performance in 3D-printed cementitious systems.

Rheology and layer retention tests confirmed that the developed 3D-Mix formulation was suitable for additive manufacturing. The mixture demonstrated stable extrusion behavior and adequate structural build-up, allowing successful printing of 10, 15, and 20-layer elements with geometric deviations of only ≈6–7% from the theoretical build height. Layer thickness measurements showed consistent filament geometry between the bottom and top layers of printed elements, confirming uniform material deposition.

From a sustainability perspective, the developed composite combines reduced Portland cement consumption with the recycling of end-of-life tires. This approach lowers CO₂ emissions associated with binder production while simultaneously diverting rubber waste from landfills. The results therefore demonstrate that GCP–crumb rubber composites can provide a balanced combination of mechanical performance, durability, and rheology, making them a promising candidate for environmentally responsible 3D-printed construction materials.

Future work should focus on incorporating recycled gypsum from construction and demolition waste, optimizing rheological control for automated printing systems, and performing full life-cycle assessment (LCA) to quantify the environmental benefits of the proposed composite, and future investigations should prioritize quantitative microstructural characterization of pore architecture, interlayer bond strength, and rubber–binder ITZ evolution to more rigorously model structure–property relationships in 3D-printed GCP–rubber systems.

While the present study provides indirect structural insight through mechanical response, porosity trends, fracture morphology, and UPV monitoring, advanced quantitative characterization methods such as SEM, X-ray CT, mercury intrusion porosimetry (MIP), and direct interlayer bond testing would further strengthen understanding of pore network evolution, ITZ behavior, and anisotropic structural performance.