1. Introduction
Gypsum-based materials (GBMs) have emerged as a cornerstone of the construction industry, owing to their extensive availability, cost efficiency, and superior fire resistance properties [
1,
2,
3]. Their adaptability as building materials has facilitated widespread adoption in diverse construction applications, such as walls, ceilings, and partition systems, highlighting their pivotal role in contemporary construction practices [
4,
5]. Nevertheless, traditional gypsum products are constrained by inherent drawbacks, such as elevated density, which escalates transportation expenses and installation labor demands, as well as inadequate mechanical strength, thereby limiting their suitability for structural applications [
6,
7,
8,
9].
To address these challenges, researchers have investigated a range of strategies to enhance gypsum composites, with particular emphasis on lightweight aggregates (LWAs) and fiber reinforcement [
10,
11,
12,
13]. LWAs, including expanded clay, are extensively employed in the construction industry, owing to their thermal insulation, acoustic damping, and structural load-reducing capabilities [
14,
15,
16]. Nevertheless, the incorporation of LWAs often compromises the mechanical performance of composites [
17,
18,
19,
20].
The incorporation of lightweight aggregates (LWAs) often leads to a decline in mechanical strength, whereas fiber reinforcement can introduce aesthetic inconsistencies or increased material costs [
21]. A promising approach to overcoming these limitations lies in the synergistic combination of these two strategies. Hollow glass microspheres (HGMs), a type of lightweight aggregate, are widely recognized for their low density and superior thermal insulation properties, making them applicable across diverse fields [
22,
23,
24]. HGMs have shown considerable potential in reducing the weight of construction materials while maintaining structural integrity [
25,
26]. For instance, Yu et al. [
27] utilized the low density and dielectric characteristics of HGMs to develop an advanced wave-absorbing material, representing a notable breakthrough in microwave absorption technology. Similarly, natural fibers, such as sisal fibers (SFs), have emerged as a viable alternative due to their biodegradability, cost efficiency, and widespread availability. Deng et al. [
28] demonstrated that SFs can improve the physical properties of frost-prone soils, with enhanced structural integrity observed in SF-reinforced soils. Additionally, Jin et al. [
29] explored the role of SFs in augmenting the toughness of cement-based materials. Furthermore, SFs have been proven to significantly enhance the mechanical performance of gypsum composites, particularly in improving flexural strength and toughness [
30,
31].
Theoretically, the incorporation of both HGMs and SFs has the potential to enhance the performance of NGBC. Specifically, HGMs, as lightweight aggregates, can effectively reduce the density of NGBC, thereby rendering it lighter, while SFs, as reinforcing fibers, can optimize its mechanical properties. However, prior research has predominantly focused on the self-weight reduction in gypsum materials through lightweight aggregates and the mechanical reinforcement provided by fibers [
32,
33], with limited attention given to their impact on the aesthetic properties of gypsum materials. Moreover, there is a notable scarcity of studies exploring the combined use of lightweight aggregates and reinforcing fibers. This study aims to address this research gap and serve as a foundational reference for future investigations. Despite the limited historical integration of HGMs and SFs in gypsum composites, their unique characteristics position them as promising candidates for innovative applications in this domain.
In this study, lightweight, high-strength, and high-whiteness natural gypsum-based composites (NGBCs) were fabricated through the synergistic integration of hollow glass microspheres (HGMs) and sisal fibers (SFs). The influence of HGMs on the density and mechanical performance of the gypsum composites was systematically investigated. Furthermore, the combined effects of HGMs and SFs on the properties of NGBCs were comprehensively evaluated. Additionally, the underlying microstructural mechanisms driving these property enhancements were meticulously examined. The findings of this study are anticipated to advance the development of efficient, sustainable, and high-performance materials for contemporary construction applications.
2. Materials and Methods
2.1. Materials
Natural gypsum (CaSO
4∙0.5H
2O), with a density of 2.60 g/cm
3, was sourced from Ningxia Jutuo Industry Co., Ltd., Wuzhong, China. HGMs were procured from 3M Company, Sao Paulo, MN, USA, and their microscopic morphology, as depicted in
Figure 1b, reveals well-defined spherical structures. The HGMs display highly uniform spherical geometries with smooth surfaces, indicative of their structural integrity and consistency. The fundamental properties of HGMs, including their particle size distribution, are summarized in
Table 1. Melamine (CQJ-SQ01), utilized as a water-reducing agent, was obtained from Shanghai Chenqi Chemical Technology Co., Shanghai, China. The SFs, with lengths ranging from 1.0 to 1.5 cm, were sourced from Hengtai Lai Hemp Products Factory, Xuchang, China. These fibers exhibit a slightly yellowish hue and possess a rough, burr-like surface characteristic of natural plant fibers. The microscopic morphology of the SFs, as illustrated in
Figure 1c–e, confirms their natural plant fiber origin, characterized by a distinctly rough surface texture.
2.2. Preparation of Natural Gypsum-Based Composites
In accordance with the Chinese standard GB/T 17669.3-1999 (‘Gypsum plasters—Determination of mechanical properties’) [
34], 1000 g of natural gypsum, along with a specified quantity of HGMs or SFs and a water-reducing agent, were prepared at a controlled room temperature of 25 ± 2 °C. Five distinct particle sizes of HGMs (20, 40, 60, 80, and 100 µm) were employed as lightweight aggregates and integrated into the natural gypsum matrix to systematically evaluate their influence on the physical properties of the composite and their interfacial interactions with the gypsum matrix. The mixtures underwent an initial stirring process using a mortar mixer for 60 s, followed by a high-speed stirring phase for an additional 30 s. Following the stirring process, the mixtures were carefully poured into molds measuring 40 × 40 × 160 mm. The molds were allowed to set for 2 h under ambient conditions. The experimental specimens were subsequently demolded and prepared for further analysis. A subset of specimens was oven-dried at 40 ± 2 °C until achieving complete desiccation. A schematic representation of the experimental protocol is provided in
Figure 2.
2.3. Materials Characterization
The experimental procedure involved positioning the specimens on two supporting rollers, with the molding surface oriented laterally. A progressively increasing load was applied until fracture initiation, and the mean flexural strength was subsequently determined. For compressive strength evaluation, six half-section specimens obtained from the flexural tests were examined using a compression fixture (Zhongyike Instrument Factory, Shaoxing, China). The specimens were loaded continuously until failure, and the mean compressive strength was recorded.
For the XRD analysis, the samples were first dried at a controlled temperature to remove any moisture that could affect the results. The dried samples were then ground to a fine powder using a mortar and pestle to ensure a uniform particle size, which is crucial for obtaining consistent XRD patterns. The powdered samples were carefully loaded into sample holders, ensuring no contamination or preferred orientation that could skew the diffraction results. The sample holder was then placed in the XRD instrument for analysis. The phase composition of the natural gypsum specimens was characterized using X-ray diffraction (XRD) (Rigaku D/max-3A; Rigaku Corporation, Tokyo, Japan), with diffraction angles ranging from 2θ = 3° to 90°. XRD data analysis was performed using MDI Jade 6 software. For the SEM analysis, the samples were first cleaned thoroughly to remove any surface contaminants. They were then dried and mounted onto SEM stubs using double-sided carbon tape. To ensure a clear image with high conductivity, the samples were sputter-coated with a thin layer of gold. This process involved placing the samples in a vacuum chamber where a gold target was bombarded with ions, causing gold atoms to sputter onto the sample surface. After coating, the samples were ready for imaging under the SEM. The micromorphology of the specimens was examined via scanning electron microscopy (SEM; EVO 10/LS10; Carl Zeiss, Oberkochen, Germany). Whiteness measurements were conducted using a WSB-1 whiteness meter (Hangzhou Chroma Technology Co., Hangzhou, China), with the blue whiteness index R457 as the metric.
3. Results and Discussion
3.1. Effect of HGM Content on Physical and Mechanical Properties of Natural Gypsum-Based Composites
The densities of NGBCs with varying HGM content and particle size are illustrated in
Figure 3. The density of NGBCs exhibited a progressive decline with increasing HGM content. At a constant HGM content, an increase in the particle size of HGM correlates with a reduction in the density of NGBC. This suggests that larger HGM particles exert a more pronounced influence on lowering the density of NGBC. This trend is primarily attributable to the significantly lower density of HGMs relative to natural gypsum. The integration of HGMs into the gypsum matrix led to the formation of microscopic cavities, a direct consequence of their inherent hollow structure. These cavities contributed to a reduction in the overall composite density. With increasing HGM content, both the number and volume of cavities expanded, progressively displacing solid gypsum with low-density glass microspheres. Notably, at a particle size of 100 µm and an HGM content of 50 vol.%, the density of NGBCs reached a minimum value of 0.94 ± 0.05 g/cm
3. A maximum density reduction of 40% was achieved under these conditions, compared to the control specimen. This significant reduction underscores the efficacy of HGMs in reducing the self-weight of NGBCs.
The 2 h mechanical properties of NGBCs with varying HGM content and particle size are illustrated in
Figure 4. The integration of HGMs significantly influenced the 2 h mechanical properties of NGBCs, resulting in a gradual decline in both flexural and compressive strengths with increasing HGM content and particle size. Notably, the detrimental effect of HGMs on mechanical properties becomes more pronounced with larger particle sizes. Specifically, the 2 h flexural strength of NGBCs reached a minimum value of 1.12 MPa at a particle size of 100 µm and an HGM content of 50 vol.%. A comparable trend was observed for the 2 h compressive strength, which attained a minimum value of 1.88 MPa at a particle size of 60 µm and an equivalent HGM content of 50 vol.%. Both the integration of HGMs and their particle size exerted a direct influence on the overall structural integrity of NGBCs. Evidently, increasing both the particle size and content of HGMs results in a corresponding deterioration of the mechanical properties of NGBCs.
The 28 d mechanical properties of NGBCs with differing HGM content and particle size are illustrated in
Figure 5. A comparison of the 2 h and 28 day mechanical properties indicates a more stable and consistent trend in the latter. Moreover, fully cured NGBCs demonstrated enhanced structural performance, reflecting a more mature material state. Nevertheless, the overall trend remains consistent, with the 28 day mechanical properties of the composites exhibiting a gradual decline as HGM content increases. At a constant HGM content, the mechanical properties of NGBC exhibited a gradual decline with increasing HGM particle size. This trend suggests that larger HGM particles have a more detrimental impact on the mechanical performance of NGBC. The incorporation of HGMs weakened the densification and inter-crystalline bonding strength of the material, thereby reducing the 28 day flexural and compressive properties of NGBCs as HGM content increased. A maximum flexural strength of 5.61 MPa was achieved at a particle size of 40 µm and an HGM content of 20 vol.%, suggesting that moderate levels of glass microspheres do not significantly impair the flexural properties of the composites. Conversely, the microspheres were found to improve the dispersion homogeneity of the composites, owing to their reinforcing backbone effect. The maximum 28 day compressive strength of 8.71 MPa was attained in specimens with a 20 µm particle size and an HGM content of 10 vol.%. This phenomenon can be attributed to the uniform distribution of smaller microspheres within the gypsum matrix, which minimizes the formation of large-scale pore structures at lower HGM content, thereby preserving high compressive strength.
The 28 day flexural and compressive strengths of the NGBCs exhibited a significant reduction as the HGM content increased from 30 vol.% to 40 vol.%. Additionally, it was noted that elevating the HGM content beyond a specific threshold markedly reduced the fluidity of the gypsum slurry, leading to challenges in achieving uniform mold filling during the casting process. This reduction in fluidity was especially evident at 40 vol.% HGM content, resulting in the formation of numerous microscopic pores. These pores have been demonstrated to compromise the structural integrity of the composite, increasing stress concentration points and thereby reducing overall strength. Moreover, the inclusion of HGMs in the mixture significantly influenced the gypsum hydration process. When the HGM content exceeded 30 vol.%, a ‘skeleton effect’ (characterized by the formation of a rigid network of lightweight aggregates within the gypsum matrix, which physically isolates and restricts moisture diffusion and gypsum particle contact) was observed, hindering the growth and interconnection of gypsum crystals. During hydration, gypsum crystals typically require close bonding to form a dense, mechanically robust structure. However, beyond a critical concentration of the HGMs, effective inter-crystal bonding was obstructed, leading to the development of microcracks and voids within the NGBCs. Consequently, this further diminished the flexural and compressive properties of the NGBCs.
Figure 6 demonstrates a direct correlation between the whiteness of the NGBCs and the content of the HGMs at a particle size of 80 µm. The figure clearly indicates that the whiteness of NGBCs varies significantly with varying HGM content. A notable increase in the whiteness of the composites was observed with increasing HGM content. At HGM contents of 0, 10, 20, 30, 40, and 50 vol.%, the whiteness values of NGBCs were measured as 65.01 ± 0.01, 67.56 ± 0.01, 69.84 ± 0.01, 71.25 ± 0.01, 73.02 ± 0.02, and 75.15 ± 0.01, respectively. The maximum whiteness value of 75.15 was achieved at an HGM content of 50 vol.%. The enhancement in whiteness can be attributed to the light-scattering properties of the microspheres, which significantly improve the surface appearance and aesthetic quality of NGBCs.
3.2. Effect of HGM Content on Microscopic Morphology of Natural Gypsum-Based Composites
The XRD pattern of NGBCs containing HGMs (with a particle size of 80 µm and a content of 30 vol.%) is presented in
Figure 7. The NGBC specimen displayed a predominant gypsum phase, with its primary characteristic peaks at 11.66°, 20.76°, and 29.15° remaining largely unchanged. This finding indicates that HGMs did not participate in significant chemical reactions with the gypsum matrix. Instead, HGMs were uniformly dispersed within the gypsum phase as a lightweight aggregate, confirming their role as an inert filler in natural gypsum. Notably, the characteristic peaks at 33.82°, 43.39°, 47.90°, and 56.82° exhibited a discernible reduction in intensity, implying that the inclusion of HGMs altered the hydration process of natural gypsum, leading to a decrease in gypsum dihydrate formation. Consequently, this reduction in peak intensity elucidates the minimal changes observed in the mechanical properties of natural gypsum specimens with HGMs. However, the mechanical performance of the natural gypsum specimens was found to decline due to the presence of microspheres.
The microscopic morphology of the NGBCs with HGM (80 µm of particle size and 30 vol.% of content) is demonstrated in
Figure 8. As shown in
Figure 8a, HGMs with a particle size of 80 µm and a content of 30 vol.% were uniformly distributed throughout the NGBC matrix. Furthermore, the inherent mechanical properties of HGMs are superior, leading to only a marginal reduction in the mechanical properties of NGBCs at 30 vol.% HGM content compared to the control specimen. As depicted in
Figure 8b,c, the interface between HGMs and the gypsum matrix is characterized by close proximity, with natural gypsum crystals demonstrating effective growth and adhesion onto the HGM surfaces. This observation indicates enhanced interfacial bonding, which contributes significantly to the structural stability of the NGBC specimens. However, at HGM contents exceeding 30 vol.%, the microspheres occupy a larger volume fraction within the gypsum matrix. This results in diminished space and bonding between gypsum crystals, thereby weakening the inter-crystalline linkage effect. While higher HGM content does lead to reduced density, it may also result in a more interconnected pore structure, which could potentially influence crack propagation and durability. Consequently, this compromises the overall mechanical strength of the NGBCs.
Compared to the control specimen, the incorporation of HGMs resulted in a moderate reduction in the mechanical properties of the composites. However, the addition of HGMs introduced cavity structures, akin to pores, within the NGBC, significantly reducing its overall density. To ensure the material’s suitability for molding during construction, its fluidity was rigorously evaluated. The optimal particle size and content of HGMs were identified as 80 µm and 30 vol.%, respectively.
3.3. Effect of SF Content on Physical and Mechanical Properties of Natural Gypsum-Based Composites
The experimental program was designed to identify the optimal formulation of hollow glass microspheres (HGMs) and sisal fibers (SFs) to achieve lightweight and high-strength properties in NGBCs. However, it was observed that HGM contents exceeding 30 vol.% significantly reduced the workability of the NGBC slurry, rendering it unsuitable for the preparation of qualified specimens. Subsequent analysis of the mechanical properties of NGBCs guided the selection of HGMs with a particle size of 80 μm and a content range of 10–30 vol.%, combined with an SF content adjustment (0.1–0.5 vol.%), to determine the optimal HGM-SF mixing ratio, as detailed in
Table 2.
The densities of NGBCs with differing HGM and SF content are illustrated in
Figure 9. The density of NGBCs exhibited a gradual decline with increasing HGM content. This reduction is primarily attributed to the integration of HGMs, which possess a lower density than natural gypsum, into the composite matrix, leading to a net reduction in the overall bulk density of the material. In contrast, the inclusion of SFs had a negligible effect on the density of the composites. The addition of SFs did not significantly alter the mass or volume of the material as their minimal contribution primarily enhanced toughness and tensile strength without inducing substantial volumetric changes. These findings highlight the dominant influence of HGM content on the quality, density, and dimensions of NGBCs. The ability to tailor the density of the composite by adjusting HGM content provides a practical strategy for designing lightweight materials while maintaining other critical mechanical properties.
The 2 h mechanical properties of the NGBCs with varying HGM and SF content are illustrated in
Figure 10. As SF content increased, the 2 h mechanical properties of NGBCs across three HGM content levels initially improved, followed by a gradual decline. The peak 2 h flexural and compressive strengths were achieved at an SF content of 0.3 vol.%. Specifically, the highest 2 h flexural and compressive strengths of 3.7 MPa and 4.78 MPa, respectively, were recorded at an HGM content of 10 vol.% and an SF content of 0.3 vol.%. Compared to the control specimen, the 2 h flexural and compressive strengths increased by 119% and 24%, respectively. The lowest 2 h flexural strength was observed at an HGM content of 30 vol.% and an SF content of 0.2 vol.%. At an SF content of 0.2 vol.%, the 2 h flexural strength of the composite displayed a fluctuating trend, characterized by an initial increase followed by a decrease. The lowest 2 h flexural strength of 1.67 MPa, representing a 4% reduction, was recorded at an HGM content of 30 vol.% and an SF content of 0.2 vol.%. The lowest 2 h compressive strength of 3.3 MPa, corresponding to an 11% reduction, was observed at an HGM content of 30 vol.% and an SF content of 0.5 vol.%.
The 28 d mechanical properties of the NGBCs with varying HGM and SF content are illustrated in
Figure 11. The 28 day mechanical properties of NGBCs incorporating HGMs and SFs displayed trends analogous to those observed in
Figure 10. As SF content increased, the 28 day mechanical properties of NGBCs with 10, 20, and 30 vol.% HGMs initially improved, followed by a gradual decline. The peak 28 day flexural strength of 6.15 MPa was achieved at an HGM content of 10 vol.% and an SF content of 0.3 vol.%. In contrast, the peak 28 day compressive strength of 9.78 MPa was recorded at an HGM content of 20 vol.% and an SF content of 0.3 vol.%.
Based on the experimental data, the combination of HGMs and SFs effectively reduced the weight of NGBCs while simultaneously enhancing their mechanical properties, achieving a balance of lightweight and high-strength characteristics. Furthermore, SFs exhibited a significantly greater enhancement effect on the 28 day flexural strength of NGBCs compared to compressive strength, primarily due to the fiber-induced toughening mechanism within the composite and the inherent high tensile strength and flexibility of SFs, which effectively dissipate stress at crack tips under bending loads, thereby inhibiting crack propagation and enhancing the flexural performance of the composites. However, the improvement in 28 day compressive strength was limited as compressive stress is predominantly transmitted through the particulate phase, which is less influenced by fiber distribution and orientation. Additionally, fibers are susceptible to bending or deformation under compressive loads, thereby diminishing their contribution to compressive performance. Consequently, SFs contribute significantly more to the 28 day flexural properties of NGBCs than to their compressive properties.
Analysis of the density and mechanical property data for NGBCs identified an optimal formulation of 20 vol.% HGMs and 0.3 vol.% SFs. This formulation achieved a significant reduction in density while maintaining mechanical properties within the desired range. This ratio represents an optimal balance, enabling the simultaneous achievement of lightweight and high-strength characteristics.
As demonstrated in
Figure 12, the whiteness of the NGBCs with 20 vol.% HGMs (80 um) and varying SF content is evident. The whiteness of NGBCs incorporating SFs and HGMs exhibited a progressive increase with higher SF content. Specifically, at SF contents of 0.1, 0.2, 0.3, 0.4, and 0.5 vol.%, the whiteness values of NGBCs were measured as 65.01 ± 0.01, 68.25 ± 0.01, 71.53 ± 0.01, 73.56 ± 0.01, 75.48 ± 0.01, and 77.62 ± 0.01, respectively. The whiteness of the composites increased significantly with higher SF content, as the synergistic effect of SFs and HGMs enhanced light scattering and surface finish, thereby improving the aesthetic properties of NGBCs. Thus, the simultaneous integration of SFs and HGMs effectively enhances the whiteness of NGBCs, contributing to improvements in both mechanical performance and visual appeal. Additionally, the whiteness of NGBCs was found to increase proportionally with higher HGM and SF content.
3.4. Effect of SF Content on Microscopic Morphology of Natural Gypsum-Based Composites
As illustrated in
Figure 13, the XRD images of the NGBCs with complex doped HGMs (80 um, 20 vol%) and SFs (1–1.5 cm, 0.3 vol%) are presented. The specimen exhibited a predominant gypsum phase, with its primary characteristic peaks at 11.66°, 20.76°, and 29.15° remaining largely unchanged. The complex-doped HGMs and SFs did not chemically react with the gypsum matrix; instead, they were physically embedded within the gypsum phase, serving as modifying agents that exerted purely physical effects on the composite structure.
The microscopic morphology of the NGBCs with HGMs (80 um, 20 vol.%) and SFs (0.3 vol.%) is shown in
Figure 14. The HGMs and SFs were effectively integrated with natural gypsum crystals, forming a hierarchical composite structure within the matrix. The HGMs exhibited excellent dispersibility, ensuring uniform distribution throughout the natural gypsum matrix. This homogeneous dispersion enhanced the mechanical properties of the material while simultaneously reducing its overall density. Consequently, the structural stability of the material was significantly improved. However, a small fraction of HGMs were observed to fracture during the gypsum slurry mixing process. The number of fractured microspheres was minimal, and their size had a negligible impact on the lightweight performance of the NGBCs. As a result, the fractured microspheres did not significantly affect the lightweight characteristics of the composites.
In contrast, the dispersion efficiency of SFs was marginally inferior to that of HGMs, with localized clusters of two to three fibers being observed. The dispersibility of SFs was significantly governed by their intrinsic physical properties. The abundance of fine burrs on the fiber surface promoted stronger interfacial bonding with the gypsum matrix, thereby enhancing flexural strength. However, these surface features also facilitated the formation of physical linkages between fibers, which impeded their uniform dispersion within the matrix and increased the likelihood of localized aggregation. As a result, when the SF content exceeded 0.3 vol.%, the strength enhancement of NGBCs demonstrated a gradual reduction. The reinforcement efficacy of SFs is contingent upon the equilibrium between the ‘proportion of effective load-bearing fibers’ and the ‘volume of defects introduced’. Beyond 0.3 vol.%, defect-related phenomena such as agglomeration, porosity, and hydration inhibition became predominant, ultimately leading to a decline in mechanical strength.