Shanxi Province, as one of the regions with the largest accumulation of coal gangue in China, faces an urgent demand for efficient resource utilisation of bulk solid waste to promote regional sustainable development. The rapidly developing 3D printing construction technology, with its precise control over material distribution and the advantage of forming complex components without molds, provides an innovative path for the large-scale consumption of solid waste. Existing research focuses on the application of coal gangue as fine aggregate substitute in printing mortar, while studies regarding its utilisation as coarse aggregate in 3D printed concrete remain unreported. It is vital to reveal the micro/mesoscale mechanisms governing the printability and fracture mechanics behaviour of such concrete material, in order to develop low-carbon coal gangue-based concrete materials and advance their engineering applications. This study innovatively uses coal gangue as a coarse aggregate and combines it with 3D printing technology to produce a novel 3D printed coal gangue concrete (3DP-CC), offering a data foundation and experimental reference for the resourceful and intelligent application of coal gangue.

In the experiment, 3DP-CC specimens were prepared with coal gangue contents of 0%, 10%, 20%, and 40% by mass. Firstly, the printing properties were evaluated, including flowability, buildability, extrudability, and printable time. Flowability was measured using a concrete slump cone in accordance with the JG/T 248 standard. Buildability was assessed through cylindrical slump and layer stacking tests. Extrudability was determined by evaluating the continuity and uniformity of the extruded filaments, with measurements of their width and thickness. Printable time was recorded as the duration over which the mixture could be extruded smoothly and consistently. Secondly, the internal microstructures of the 3DP-CC were examined using high-resolution X-ray CT scanning technology, allowing for the analysis of pore and aggregate characteristics. Finally, uniaxial compression and three-point bending tests were performed to determine the compressive and flexural strengths. The anisotropy index λ was introduced to quantify the directional variation in mechanical strength, followed by an analysis of the underlying mechanisms by which printing direction and coal gangue content influence mechanical properties. Failure modes were examined to highlight the critical roles of weak interlayer and interstrip interfaces, thereby providing further insight into the microscopic mechanisms governing the mechanical anisotropy of 3DP-CC.

The test results indicate that both the compressive and flexural strengths of 3DP-CC exhibit a trend of first increasing and then decreasing with rising coal gangue content, accompanied by notable mechanical anisotropy. At coal gangue contents of 10% and 20%, strength improved significantly while anisotropy was reduced. In contrast, a content of 40% led to a decrease in strength and an increase in anisotropy. These findings suggest that an appropriate amount of coal gangue coarse aggregate can provide skeletal support, thereby enhancing the mechanical performance of concrete and mitigating anisotropic effects. However, when the coal gangue content is too high, the low strength and high brittleness of the gangue itself, combined with an increase in medium and large pores, result in diminished overall strength and greater anisotropy.

Coal gangue content also has a pronounced effect on the pore structure of 3DP-CC, with total porosity initially decreasing and then increasing as the gangue content rises. At coal gangue contents of 10% and 20%, the total porosity of the specimens is 1.8% and 2.1%, respectively—both lower than the 3.9% observed in the control 3DPM. However, the total porosity increases to 3.4% in the 3DP-CC-40 specimens. From a microstructural perspective, the morphology of coal gangue aggregates plays a significant role in this change. The aggregates exhibit rough surfaces and angular shapes, which lead to loose packing and reduced efficiency in filling medium and large pores. Consequently, the number of pores with equivalent diameters greater than 1 mm increases in 3DP-CC-40 specimens. Furthermore, the water absorption capacity of coal gangue affects the internal pore structure of the concrete. As the aggregates absorb water, the moisture content in the surrounding matrix decreases, resulting in increased matrix porosity.

In terms of failure modes, the compressive behaviour of 3DP-CC is strongly influenced by the loading direction. When loading is applied on the x-face, the failure pattern on the y-side face shows cracks initiating at both edges, propagating inward, and then extending vertically downward. These cracks often penetrate the interlayer interfaces, leading to a relatively brittle failure mode. In contrast, loading on the y-face or z-face produces a distinctive "π"-shaped crack pattern on the x-face, accompanied by spalling damage on both sides, indicating that the weak interlayer and interstrip interfaces play a critical role in crack development. The failure mode under three-point bending is also dependent on the loading direction. When the x-face serves as the loading surface, all specimens—regardless of coal gangue content—exhibit through-thickness, straight cracks and undergo brittle fracture into two parts. When the y-face is loaded, through-thickness cracks are absent in specimens with 0% coal gangue but present in those with higher contents. For loading on the z-face, only specimens with 40% coal gangue exhibit through-thickness cracks, while others show no visible cracking on the loading surface. These observations suggest that specimens loaded on the x-face exhibit greater brittleness, with cracks more readily propagating along interstrip interfaces.

The coal gangue content significantly influences the performance of 3D printed concrete (3DP-CC). Optimal mechanical properties—both compressive and flexural strength—are achieved at a coal gangue content of 10~20%, attributed to the skeletal support provided by the gangue particles and the associated reduction in porosity. However, at a content of 40%, both strengths significantly decrease due to the inherent weakness and brittleness of coal gangue, its tendency to induce cracking, and the increase in macroporosity. The pore structure exhibits a non-linear response to coal gangue content: total porosity decreases when the content is below 20%, but rises sharply at 40% as a result of insufficient fine aggregate filling, leading to the formation of numerous large pores. 3D printing introduces directional porosity, with the upper layers exhibiting slightly higher porosity. In addition, pronounced mechanical anisotropy is observed, as quantified by the anisotropy index. Increasing the coal gangue content tends to reduce the anisotropy of compressive strength. In contrast, the anisotropy of flexural strength exhibits a non-monotonic trend: it initially decreases and then increases, reaching a maximum at 40% gangue content. Failure is predominantly governed by weak interlayer and interstrip interfaces, along which cracks consistently propagate under both compressive and flexural loading. Coal gangue aggregates influence crack trajectories—cracks either bypass or penetrate the aggregates depending on local strength fluctuations. These findings offer valuable insights for the design of sustainable 3D printed concrete, emphasizing the importance of optimizing coal gangue content and microstructures to balance strength, anisotropy, and resource utilisation.