TY - JOUR
AU - Kim, Mintae
AB - 1. Introduction With the rapid growth of population, the solution to the contradiction between development and environment has become the focus of the world. In this context, the investigation of sustainable development technology becomes a global consensus. Especially, 3D printing based on concrete material is recently a novel and promising process technology in the building industry. Nowadays, 3D printing is capable of creating almost any complicated structures and has been confirmed as a promising and green manufacturing technology for various fields including aerospace, robotics, architecture and biomedical research, etc [1–3]. In comparison to others, 3D printing based on concrete materials is an innovative and promising process technology in the building industry [4–6], which possesses the superiorities of flexible construction [7, 8], efficient and automated construction [9], and accurate construction [10]. As a consequence, 3D printing concrete technology has attracted considerable attentions in recent years [11, 12]. At present, the raw materials of 3D printing cement are mostly dependent on the large-scale mining. With the rapid development of Chinese economy, natural sand has become the largest, indispensable and irreplaceable raw materials because of an annual consumption of about 20 billion tons [13]. However, the extensive consumption of natural sand will not only destroy the ecological environment of the earth, but also inevitably result in enormous energy consumptions and excessive greenhouse gases emissions [14–16]. In general, the manufacture of 1 ton of natural sand consumes nearly 2 tons of water and produces a large amount of carbon dioxide [17]. Therefore, utilizing recyclable or waste materials to replace natural sand not only promotes the recycle of solid waste, but also substantially reduces the carbon dioxide emissions during the production process, thus contributing to an environmental sustainability development [18–21]. During the past decades, various solid waste materials including geopolymer pastes, recycled glass, mining tailings and so on have been continuously utilized to replace the natural sand for producing the 3D printing cementitious materials [22–24]. However, most materials inevitably have an impact on the inherent mechanical strength and printability of concrete material. Moreover, the small amount of heavy metals existed in some mining tailings may restrict their application in the civilian field. So, there is an urgent requirement to further expand the currently limited range of materials, and improve the mechanical strength and printability of concrete materials during 3D printing process. As a kind of common solid waste materials, the mass stack of limestone powders will seriously pollute the atmosphere and water, which brings severe threats to the safety of human beings [25–27]. Therefore, the Chinese government strongly encourages the explorations of the limestone powder resource utilization technology. Among them, the limestone powders are capable to been employed as an alternative material to concrete mixtures, which is a promising resource utilization method. Despite some explorations have been carried out in relation to limestone powders in the field of concrete preparation, the application of limestone powders in 3D printing cementitious materials is insufficient and the optimal replacement ratio is yet to be explored. In this paper, a novel concrete material that replaced natural sand with fine limestone powders was proposed for 3D concrete printing applications. The concrete materials with different replacement ratios of limestone powders to natural sand were investigated. Experimental measurements were performed to evaluate and compare the printable properties of above materials, including microstructures characteristics, flowability, buildability, shrinkability, layer-interface properties, mechanical properties and interlayer bonding strength, etc. And then the reasonable limestone powder replacement ratio was determined. This research explores an eco-friendly 3D printing concrete material, which not only significantly improves its mechanical strength and printability, but also effectively develops the current solid waste, thus contributing to an environmental sustainability development. 2. Materials and methods 2.1. Raw materials and procedures 2.1.1 Raw materials. The raw materials in this study were natural sand, Portland cement, limestone powder, fly ash and superplasticizer. Among them, the specific surface area of natural sand is 0.101 m2/g. And the chemical compositions of cement, limestone powders as well as fly ash have been measured and analyzed by the X-ray fluorescence (XRF) analysis, as displayed in Table 1. In order to improve the buildability of the novel concrete materials, a powdered polycarboxylate superplasticizer (BASF Company) was employed to meet the fresh behavior requirements. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 1. The composition of concrete materials. https://doi.org/10.1371/journal.pone.0292951.t001 Particle size distribution testing can be employed to help understand the dispersion degree of particles, including particle size, shape, density, and the gaps among particles, etc, which is of great importance for the subsequent processing and performance of the materials. Specially, 3D printing concrete materials are required to possess smaller and smoother particle grading to flow easily out of the printer and ensure the flowability and printability [28, 29]. A proper grading can provide a stable composting structure, which contributes to the formation of a dense filling system and decreases the manufacturing defects or cracks, thus improving the interlayer bonding strength of the structures. In this paper, the particle size distribution of limestone powder was measured by laser particle size analyzer (Topsizer 2000). The cumulative curve of particle volume percentage of limestone powder is shown in Fig 1. The measurement result displayed that the limestone powders had much smaller diameters than the natural sand. And the result was recalculated as a percentage of the particles number. It displayed that the proportions of limestone powders less than 1μm and greater than 200 μm were negligible, and large amounts of particles were distributed over the size of 10–100 μm. Specially, its average particle size was 77.5 μm, which made it easy to pass through the printer nozzle, and the filaments being printed were smoother and suitable for the 3D printing. Above analysis demonstrated that appropriate additions of limestone powders were effectively filled into the interspace between the natural sand particles, thus improving the compactness of the mixtures. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 1. Particle size distribution of the limestone powder. https://doi.org/10.1371/journal.pone.0292951.g001 2.1.2 Fabrication procedures. The fabrication processes, characterization and main measurement methods of the 3D printing concrete materials are shown in Fig 2, and the detailed procedures are as follows: A planetary mixer was employed to prepare the concrete materials. Firstly, the limestone, natural sand, fly ash and Portland cement were gradually introduced into a mortar mixer and slowly stirred at 59 rpm for 1 min [30]. Afterward, the 90% of the predetermined amounts of deionized water and superplasticizer were slowly poured into the above solutions and stirred at quick speed (198 rpm) for 1 min to obtain a relatively homogeneous mixture. Then, the remaining deionized water was immersed into above mixture while stirring for another 5 min. Finally, the fresh mixture was put into the charging barrel and shaken evenly. After which, the material was deposited in the 3D printer for printing the samples, followed by the curing at 25°C for 48~42 h in a relative humidity of 95%. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 2. Schematic illustration of fabrication process and main measurement methods for 3D printing concrete materials. https://doi.org/10.1371/journal.pone.0292951.g002 Table 2 displays the composition of concrete materials for fabrication process. The mass ratios of limestone powders are 0%, 10%, 20%, 30%, 40%, 50%, 60% and 100% for samples T1, T2, T3, T4, T5, T6, T7 and T8, respectively. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 2. The composition of concrete materials used for 3D printable (kg/m3). https://doi.org/10.1371/journal.pone.0292951.t002 2.2. Estimation of printability 2.2.1 Flowability test. To achieve the flowability evaluation of the concrete materials, a series of experiments have been performed, including the slump test and the jumping table test. The slump test is most widely used in laboratories and fields [31]. As shown in Fig 3A, a slump cone was designed as the diameter of 100 mm and 50 mm at the base and top, and the height of 150 mm. Firstly, the freshly mixed cement was placed into above slump cone in three times, of which each time was about one-third of its volume. And each time was evenly rammed for about 25 times along the spiral from the edge to the center with a vibrator. In addition, the rammer was required to run through the entire depth. Then, a trowel was employed to scrape off the excess mixture and trowel the mouth of the cone. Finally, the cone was immediately and smoothly removed within 5s to 10s, so that the concrete was not affected by the lateral and torsional force. The flowability was evaluated by the decrease in height of concrete materials. The slump was considered as the height from the top of cone to the highest point of the concrete within 15±5s. Importantly, the tests were repeated every 10 min for each sample for a total of 90 min and each sample was repeated three times. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 3. (a) Slump test and (b, c, d) Jumping table test method for flowability measurement. https://doi.org/10.1371/journal.pone.0292951.g003 The jumping table test was carried out according to the Chinese national standard (GB/T 2419–2005) [32], as shown in Fig 3B. Firstly, the concrete materials were prepared according to the mixing ratio, during which it should be uniformly vibrated or compacted to ensure the compactness of the sample. Then, a cone mould was evenly filled with the freshly mixed concrete materials, and the surface of slurry was trowelled by a knife to set it level with the upper part of the mold. Afterwards, the cone was vertically quickly removed and an iron ball of the specific weight fell freely above the table. As a result, the impact of the iron ball allowed the concrete material to freely spread out. At this moment, the height of the iron ball was recorded, and the ball jumped for 25 times within 15±1s. Finally, the flowability was characterized by the diameter of slurry measured in two perpendicular directions x and y (Fig 3C and 3D). Similarly, the tests were repeated every 10 min for each sample for a total of 90 min and each sample was repeated three times. 2.2.2 Buildability test. Buildability is one of the critical parameters to estimate its printability, which is essentially the capability to maintain its extruded shape and resist deformation under external load for the concrete. In general, concrete materials were stored in a barrel and evenly extruded from the nozzle by screw rotation to generate multilayered structures [31, 33]. In the experiment, the width of nozzle used was 30 mm, the total height of the model was 150 mm. The sample had a total of 10 layers and each layer was formed at a rate of 100 mm/s. The entire printing process was finished within 5 min. Then, two parameters including vertical strain and aspect ratio of the printing samples were investigated, as shown in Fig 4. They could be calculated by the Eqs (1) and (2), respectively [30]: (1) where, H0 and ΔH were the initial height and height variations of the printed samples. (2) where H and w were the real-time height and width of the printed samples. Without loss of generality, the geometric factor Ahw was adopted to estimate the stability of the structures. Meanwhile, εH and Ahw were used to synthetically analyze the buildability of concrete materials. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 4. Buildability test method and evaluation parameters of concrete materials. https://doi.org/10.1371/journal.pone.0292951.g004 2.2.3 Shrinkage test. The fresh mixture was required to possess good water retention performance and no damage, separation, or blockage. The base was made of plastic plate, and the clean cling film was placed on the plate to ensure its free shrinkage. On this basis, the concrete strips extruded from the printer were cut into 5 pieces with a length of 100 mm. Initially, measurements were carried out every 2 min with a high-precision digital caliper, and then taken every 10 min after 10 min. The shrinkage test was carried out according to the Chinese national standard (GB/T 50082–2009) [30]. The three points of the cross-section were selected during the measurement, and the average value was calculated for expression after measuring five segments. Owing to the extrusion of the upper print body, the phenomenon of test point deviation may occurs. At the moment, adjacent complete points within 5mm should be selected for testing, and the test section was required to be reselected if adjacent points cannot meet the requirements. 2.2.4 Layer-interface properties and microstructure analysis. In order to explore the differences between the layer-interface structures of 3D printing concrete materials with different limestone powders substituting ratios, the samples cured for 28 days were observed and studied by using a magnifying glass. Firstly, samples cured for 28 days were cut into small pieces near the interlayers and polished. Then, a magnifying glass with 100x was employed to observe their interlayer structures and compare the differences between the various samples. Besides, the microstructures of the optimum designed concrete materials were characterized by scanning electron microscopy (SEM, Gemini 500, Carl Zeiss Jena, Germany). The printed samples were cut into small flakes with planarizing surface and vacuum dried at 55°C for 12h. After that, the samples were treated with gold plating and the SEM tests were carried out. 2.2.5 Mechanical properties test. Mechanical characteristics always play an important role in the printability. The samples were cured in a standard curing room for a certain period of time after being printed layer by layer, and then the samples were shaped in the dimension of 70mm×70mm×70mm. The printing direction is along the length direction, as shown in Fig 5. The vertical pressure along the Z-axis was applied at a loading speed of 2kN/min. The uniaxial compressive strengths of different samples were measured after 3, 14 and 28 days according to EN 196–1 (2005) [34]. Each set of samples were repeated three times with the average as the final results. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 5. (a) Schematic diagram and (b) physical picture of the mechanical properties test for samples. https://doi.org/10.1371/journal.pone.0292951.g005 2.2.6 Interlayer bonding behavior. Interlayer bonding strength is closely related to the mechanical properties of printing concrete materials [35]. To investigate the effect of limestone powder on the interlayer bonding behavior, the 3D printed specimens have been cured for 7 days after printing into the dimension of 100 mm × 100 mm × 100 mm. The experiments were carried out according to the Chinese national testing standard GB/T 50081–2019. And the interlayer interface was conducted by the splitting tensile strength tests with the loading direction in Fig 6. Similarly, each set of samples were repeated three times with the average as the final results. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 6. Schematic diagram and physical picture of the structural build up behavior test for samples. https://doi.org/10.1371/journal.pone.0292951.g006 2.1. Raw materials and procedures 2.1.1 Raw materials. The raw materials in this study were natural sand, Portland cement, limestone powder, fly ash and superplasticizer. Among them, the specific surface area of natural sand is 0.101 m2/g. And the chemical compositions of cement, limestone powders as well as fly ash have been measured and analyzed by the X-ray fluorescence (XRF) analysis, as displayed in Table 1. In order to improve the buildability of the novel concrete materials, a powdered polycarboxylate superplasticizer (BASF Company) was employed to meet the fresh behavior requirements. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 1. The composition of concrete materials. https://doi.org/10.1371/journal.pone.0292951.t001 Particle size distribution testing can be employed to help understand the dispersion degree of particles, including particle size, shape, density, and the gaps among particles, etc, which is of great importance for the subsequent processing and performance of the materials. Specially, 3D printing concrete materials are required to possess smaller and smoother particle grading to flow easily out of the printer and ensure the flowability and printability [28, 29]. A proper grading can provide a stable composting structure, which contributes to the formation of a dense filling system and decreases the manufacturing defects or cracks, thus improving the interlayer bonding strength of the structures. In this paper, the particle size distribution of limestone powder was measured by laser particle size analyzer (Topsizer 2000). The cumulative curve of particle volume percentage of limestone powder is shown in Fig 1. The measurement result displayed that the limestone powders had much smaller diameters than the natural sand. And the result was recalculated as a percentage of the particles number. It displayed that the proportions of limestone powders less than 1μm and greater than 200 μm were negligible, and large amounts of particles were distributed over the size of 10–100 μm. Specially, its average particle size was 77.5 μm, which made it easy to pass through the printer nozzle, and the filaments being printed were smoother and suitable for the 3D printing. Above analysis demonstrated that appropriate additions of limestone powders were effectively filled into the interspace between the natural sand particles, thus improving the compactness of the mixtures. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 1. Particle size distribution of the limestone powder. https://doi.org/10.1371/journal.pone.0292951.g001 2.1.2 Fabrication procedures. The fabrication processes, characterization and main measurement methods of the 3D printing concrete materials are shown in Fig 2, and the detailed procedures are as follows: A planetary mixer was employed to prepare the concrete materials. Firstly, the limestone, natural sand, fly ash and Portland cement were gradually introduced into a mortar mixer and slowly stirred at 59 rpm for 1 min [30]. Afterward, the 90% of the predetermined amounts of deionized water and superplasticizer were slowly poured into the above solutions and stirred at quick speed (198 rpm) for 1 min to obtain a relatively homogeneous mixture. Then, the remaining deionized water was immersed into above mixture while stirring for another 5 min. Finally, the fresh mixture was put into the charging barrel and shaken evenly. After which, the material was deposited in the 3D printer for printing the samples, followed by the curing at 25°C for 48~42 h in a relative humidity of 95%. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 2. Schematic illustration of fabrication process and main measurement methods for 3D printing concrete materials. https://doi.org/10.1371/journal.pone.0292951.g002 Table 2 displays the composition of concrete materials for fabrication process. The mass ratios of limestone powders are 0%, 10%, 20%, 30%, 40%, 50%, 60% and 100% for samples T1, T2, T3, T4, T5, T6, T7 and T8, respectively. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 2. The composition of concrete materials used for 3D printable (kg/m3). https://doi.org/10.1371/journal.pone.0292951.t002 2.1.1 Raw materials. The raw materials in this study were natural sand, Portland cement, limestone powder, fly ash and superplasticizer. Among them, the specific surface area of natural sand is 0.101 m2/g. And the chemical compositions of cement, limestone powders as well as fly ash have been measured and analyzed by the X-ray fluorescence (XRF) analysis, as displayed in Table 1. In order to improve the buildability of the novel concrete materials, a powdered polycarboxylate superplasticizer (BASF Company) was employed to meet the fresh behavior requirements. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 1. The composition of concrete materials. https://doi.org/10.1371/journal.pone.0292951.t001 Particle size distribution testing can be employed to help understand the dispersion degree of particles, including particle size, shape, density, and the gaps among particles, etc, which is of great importance for the subsequent processing and performance of the materials. Specially, 3D printing concrete materials are required to possess smaller and smoother particle grading to flow easily out of the printer and ensure the flowability and printability [28, 29]. A proper grading can provide a stable composting structure, which contributes to the formation of a dense filling system and decreases the manufacturing defects or cracks, thus improving the interlayer bonding strength of the structures. In this paper, the particle size distribution of limestone powder was measured by laser particle size analyzer (Topsizer 2000). The cumulative curve of particle volume percentage of limestone powder is shown in Fig 1. The measurement result displayed that the limestone powders had much smaller diameters than the natural sand. And the result was recalculated as a percentage of the particles number. It displayed that the proportions of limestone powders less than 1μm and greater than 200 μm were negligible, and large amounts of particles were distributed over the size of 10–100 μm. Specially, its average particle size was 77.5 μm, which made it easy to pass through the printer nozzle, and the filaments being printed were smoother and suitable for the 3D printing. Above analysis demonstrated that appropriate additions of limestone powders were effectively filled into the interspace between the natural sand particles, thus improving the compactness of the mixtures. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 1. Particle size distribution of the limestone powder. https://doi.org/10.1371/journal.pone.0292951.g001 2.1.2 Fabrication procedures. The fabrication processes, characterization and main measurement methods of the 3D printing concrete materials are shown in Fig 2, and the detailed procedures are as follows: A planetary mixer was employed to prepare the concrete materials. Firstly, the limestone, natural sand, fly ash and Portland cement were gradually introduced into a mortar mixer and slowly stirred at 59 rpm for 1 min [30]. Afterward, the 90% of the predetermined amounts of deionized water and superplasticizer were slowly poured into the above solutions and stirred at quick speed (198 rpm) for 1 min to obtain a relatively homogeneous mixture. Then, the remaining deionized water was immersed into above mixture while stirring for another 5 min. Finally, the fresh mixture was put into the charging barrel and shaken evenly. After which, the material was deposited in the 3D printer for printing the samples, followed by the curing at 25°C for 48~42 h in a relative humidity of 95%. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 2. Schematic illustration of fabrication process and main measurement methods for 3D printing concrete materials. https://doi.org/10.1371/journal.pone.0292951.g002 Table 2 displays the composition of concrete materials for fabrication process. The mass ratios of limestone powders are 0%, 10%, 20%, 30%, 40%, 50%, 60% and 100% for samples T1, T2, T3, T4, T5, T6, T7 and T8, respectively. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 2. The composition of concrete materials used for 3D printable (kg/m3). https://doi.org/10.1371/journal.pone.0292951.t002 2.2. Estimation of printability 2.2.1 Flowability test. To achieve the flowability evaluation of the concrete materials, a series of experiments have been performed, including the slump test and the jumping table test. The slump test is most widely used in laboratories and fields [31]. As shown in Fig 3A, a slump cone was designed as the diameter of 100 mm and 50 mm at the base and top, and the height of 150 mm. Firstly, the freshly mixed cement was placed into above slump cone in three times, of which each time was about one-third of its volume. And each time was evenly rammed for about 25 times along the spiral from the edge to the center with a vibrator. In addition, the rammer was required to run through the entire depth. Then, a trowel was employed to scrape off the excess mixture and trowel the mouth of the cone. Finally, the cone was immediately and smoothly removed within 5s to 10s, so that the concrete was not affected by the lateral and torsional force. The flowability was evaluated by the decrease in height of concrete materials. The slump was considered as the height from the top of cone to the highest point of the concrete within 15±5s. Importantly, the tests were repeated every 10 min for each sample for a total of 90 min and each sample was repeated three times. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 3. (a) Slump test and (b, c, d) Jumping table test method for flowability measurement. https://doi.org/10.1371/journal.pone.0292951.g003 The jumping table test was carried out according to the Chinese national standard (GB/T 2419–2005) [32], as shown in Fig 3B. Firstly, the concrete materials were prepared according to the mixing ratio, during which it should be uniformly vibrated or compacted to ensure the compactness of the sample. Then, a cone mould was evenly filled with the freshly mixed concrete materials, and the surface of slurry was trowelled by a knife to set it level with the upper part of the mold. Afterwards, the cone was vertically quickly removed and an iron ball of the specific weight fell freely above the table. As a result, the impact of the iron ball allowed the concrete material to freely spread out. At this moment, the height of the iron ball was recorded, and the ball jumped for 25 times within 15±1s. Finally, the flowability was characterized by the diameter of slurry measured in two perpendicular directions x and y (Fig 3C and 3D). Similarly, the tests were repeated every 10 min for each sample for a total of 90 min and each sample was repeated three times. 2.2.2 Buildability test. Buildability is one of the critical parameters to estimate its printability, which is essentially the capability to maintain its extruded shape and resist deformation under external load for the concrete. In general, concrete materials were stored in a barrel and evenly extruded from the nozzle by screw rotation to generate multilayered structures [31, 33]. In the experiment, the width of nozzle used was 30 mm, the total height of the model was 150 mm. The sample had a total of 10 layers and each layer was formed at a rate of 100 mm/s. The entire printing process was finished within 5 min. Then, two parameters including vertical strain and aspect ratio of the printing samples were investigated, as shown in Fig 4. They could be calculated by the Eqs (1) and (2), respectively [30]: (1) where, H0 and ΔH were the initial height and height variations of the printed samples. (2) where H and w were the real-time height and width of the printed samples. Without loss of generality, the geometric factor Ahw was adopted to estimate the stability of the structures. Meanwhile, εH and Ahw were used to synthetically analyze the buildability of concrete materials. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 4. Buildability test method and evaluation parameters of concrete materials. https://doi.org/10.1371/journal.pone.0292951.g004 2.2.3 Shrinkage test. The fresh mixture was required to possess good water retention performance and no damage, separation, or blockage. The base was made of plastic plate, and the clean cling film was placed on the plate to ensure its free shrinkage. On this basis, the concrete strips extruded from the printer were cut into 5 pieces with a length of 100 mm. Initially, measurements were carried out every 2 min with a high-precision digital caliper, and then taken every 10 min after 10 min. The shrinkage test was carried out according to the Chinese national standard (GB/T 50082–2009) [30]. The three points of the cross-section were selected during the measurement, and the average value was calculated for expression after measuring five segments. Owing to the extrusion of the upper print body, the phenomenon of test point deviation may occurs. At the moment, adjacent complete points within 5mm should be selected for testing, and the test section was required to be reselected if adjacent points cannot meet the requirements. 2.2.4 Layer-interface properties and microstructure analysis. In order to explore the differences between the layer-interface structures of 3D printing concrete materials with different limestone powders substituting ratios, the samples cured for 28 days were observed and studied by using a magnifying glass. Firstly, samples cured for 28 days were cut into small pieces near the interlayers and polished. Then, a magnifying glass with 100x was employed to observe their interlayer structures and compare the differences between the various samples. Besides, the microstructures of the optimum designed concrete materials were characterized by scanning electron microscopy (SEM, Gemini 500, Carl Zeiss Jena, Germany). The printed samples were cut into small flakes with planarizing surface and vacuum dried at 55°C for 12h. After that, the samples were treated with gold plating and the SEM tests were carried out. 2.2.5 Mechanical properties test. Mechanical characteristics always play an important role in the printability. The samples were cured in a standard curing room for a certain period of time after being printed layer by layer, and then the samples were shaped in the dimension of 70mm×70mm×70mm. The printing direction is along the length direction, as shown in Fig 5. The vertical pressure along the Z-axis was applied at a loading speed of 2kN/min. The uniaxial compressive strengths of different samples were measured after 3, 14 and 28 days according to EN 196–1 (2005) [34]. Each set of samples were repeated three times with the average as the final results. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 5. (a) Schematic diagram and (b) physical picture of the mechanical properties test for samples. https://doi.org/10.1371/journal.pone.0292951.g005 2.2.6 Interlayer bonding behavior. Interlayer bonding strength is closely related to the mechanical properties of printing concrete materials [35]. To investigate the effect of limestone powder on the interlayer bonding behavior, the 3D printed specimens have been cured for 7 days after printing into the dimension of 100 mm × 100 mm × 100 mm. The experiments were carried out according to the Chinese national testing standard GB/T 50081–2019. And the interlayer interface was conducted by the splitting tensile strength tests with the loading direction in Fig 6. Similarly, each set of samples were repeated three times with the average as the final results. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 6. Schematic diagram and physical picture of the structural build up behavior test for samples. https://doi.org/10.1371/journal.pone.0292951.g006 2.2.1 Flowability test. To achieve the flowability evaluation of the concrete materials, a series of experiments have been performed, including the slump test and the jumping table test. The slump test is most widely used in laboratories and fields [31]. As shown in Fig 3A, a slump cone was designed as the diameter of 100 mm and 50 mm at the base and top, and the height of 150 mm. Firstly, the freshly mixed cement was placed into above slump cone in three times, of which each time was about one-third of its volume. And each time was evenly rammed for about 25 times along the spiral from the edge to the center with a vibrator. In addition, the rammer was required to run through the entire depth. Then, a trowel was employed to scrape off the excess mixture and trowel the mouth of the cone. Finally, the cone was immediately and smoothly removed within 5s to 10s, so that the concrete was not affected by the lateral and torsional force. The flowability was evaluated by the decrease in height of concrete materials. The slump was considered as the height from the top of cone to the highest point of the concrete within 15±5s. Importantly, the tests were repeated every 10 min for each sample for a total of 90 min and each sample was repeated three times. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 3. (a) Slump test and (b, c, d) Jumping table test method for flowability measurement. https://doi.org/10.1371/journal.pone.0292951.g003 The jumping table test was carried out according to the Chinese national standard (GB/T 2419–2005) [32], as shown in Fig 3B. Firstly, the concrete materials were prepared according to the mixing ratio, during which it should be uniformly vibrated or compacted to ensure the compactness of the sample. Then, a cone mould was evenly filled with the freshly mixed concrete materials, and the surface of slurry was trowelled by a knife to set it level with the upper part of the mold. Afterwards, the cone was vertically quickly removed and an iron ball of the specific weight fell freely above the table. As a result, the impact of the iron ball allowed the concrete material to freely spread out. At this moment, the height of the iron ball was recorded, and the ball jumped for 25 times within 15±1s. Finally, the flowability was characterized by the diameter of slurry measured in two perpendicular directions x and y (Fig 3C and 3D). Similarly, the tests were repeated every 10 min for each sample for a total of 90 min and each sample was repeated three times. 2.2.2 Buildability test. Buildability is one of the critical parameters to estimate its printability, which is essentially the capability to maintain its extruded shape and resist deformation under external load for the concrete. In general, concrete materials were stored in a barrel and evenly extruded from the nozzle by screw rotation to generate multilayered structures [31, 33]. In the experiment, the width of nozzle used was 30 mm, the total height of the model was 150 mm. The sample had a total of 10 layers and each layer was formed at a rate of 100 mm/s. The entire printing process was finished within 5 min. Then, two parameters including vertical strain and aspect ratio of the printing samples were investigated, as shown in Fig 4. They could be calculated by the Eqs (1) and (2), respectively [30]: (1) where, H0 and ΔH were the initial height and height variations of the printed samples. (2) where H and w were the real-time height and width of the printed samples. Without loss of generality, the geometric factor Ahw was adopted to estimate the stability of the structures. Meanwhile, εH and Ahw were used to synthetically analyze the buildability of concrete materials. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 4. Buildability test method and evaluation parameters of concrete materials. https://doi.org/10.1371/journal.pone.0292951.g004 2.2.3 Shrinkage test. The fresh mixture was required to possess good water retention performance and no damage, separation, or blockage. The base was made of plastic plate, and the clean cling film was placed on the plate to ensure its free shrinkage. On this basis, the concrete strips extruded from the printer were cut into 5 pieces with a length of 100 mm. Initially, measurements were carried out every 2 min with a high-precision digital caliper, and then taken every 10 min after 10 min. The shrinkage test was carried out according to the Chinese national standard (GB/T 50082–2009) [30]. The three points of the cross-section were selected during the measurement, and the average value was calculated for expression after measuring five segments. Owing to the extrusion of the upper print body, the phenomenon of test point deviation may occurs. At the moment, adjacent complete points within 5mm should be selected for testing, and the test section was required to be reselected if adjacent points cannot meet the requirements. 2.2.4 Layer-interface properties and microstructure analysis. In order to explore the differences between the layer-interface structures of 3D printing concrete materials with different limestone powders substituting ratios, the samples cured for 28 days were observed and studied by using a magnifying glass. Firstly, samples cured for 28 days were cut into small pieces near the interlayers and polished. Then, a magnifying glass with 100x was employed to observe their interlayer structures and compare the differences between the various samples. Besides, the microstructures of the optimum designed concrete materials were characterized by scanning electron microscopy (SEM, Gemini 500, Carl Zeiss Jena, Germany). The printed samples were cut into small flakes with planarizing surface and vacuum dried at 55°C for 12h. After that, the samples were treated with gold plating and the SEM tests were carried out. 2.2.5 Mechanical properties test. Mechanical characteristics always play an important role in the printability. The samples were cured in a standard curing room for a certain period of time after being printed layer by layer, and then the samples were shaped in the dimension of 70mm×70mm×70mm. The printing direction is along the length direction, as shown in Fig 5. The vertical pressure along the Z-axis was applied at a loading speed of 2kN/min. The uniaxial compressive strengths of different samples were measured after 3, 14 and 28 days according to EN 196–1 (2005) [34]. Each set of samples were repeated three times with the average as the final results. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 5. (a) Schematic diagram and (b) physical picture of the mechanical properties test for samples. https://doi.org/10.1371/journal.pone.0292951.g005 2.2.6 Interlayer bonding behavior. Interlayer bonding strength is closely related to the mechanical properties of printing concrete materials [35]. To investigate the effect of limestone powder on the interlayer bonding behavior, the 3D printed specimens have been cured for 7 days after printing into the dimension of 100 mm × 100 mm × 100 mm. The experiments were carried out according to the Chinese national testing standard GB/T 50081–2019. And the interlayer interface was conducted by the splitting tensile strength tests with the loading direction in Fig 6. Similarly, each set of samples were repeated three times with the average as the final results. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 6. Schematic diagram and physical picture of the structural build up behavior test for samples. https://doi.org/10.1371/journal.pone.0292951.g006 3. Results and discussion 3.1. Flowability evaluation Fig 7(A) and 7(B) display the slump and jumping table test results of concrete materials with different limestone powders replacement ratios from 10min to 90min, respectively. A larger slump value and spreading diameter result in an increased flowability and vice versa. Remarkably, with the increase of limestone powders replacement ratios, the flowability at the specific time improves. At the time of 20 min, the slump in heights of T1-T7 is 3.67cm, 4.40cm, 6.15cm, 7.34cm, 8.18cm, 9.23cm and 9.35cm, respectively. In general, it could be considered to display better flowability when the slump of the mixture slurry falls into the range of 3.5cm to 8cm. Therefore, substituting natural sand with 40% limestone powders (sample T5) increases the fluidity by 1.23% compared to the reference sample T1. While their corresponding spreading diameters in the jumping table test are 18.32cm, 18.50cm, 18.79cm, 19.96cm, 20.26cm, 21.05cm, and 21.66cm of T1-T7 at the rest time of 20min. In general, it could be considered to display better flowability when the jumping table test of the mixture slurry falls into the range of 17cm to 20.5cm. So the spreading diameter of sample T5 is improved by 10.59% compared to sample T1, which is consistent with the slump test analysis. Additionally, the fluidity displays a decrease tendency with the elevation of rest time. It is due to the fact that a longer rest time is capable of improving the hydration degree of binding materials, which leads to a decrease in the free water of slurry and an increase in static yield stress, thus resulting in the reduced flowability of the materials. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 7. (a) Slump and (b) spreading diameter dependence of rest time for concrete materials with various limestone powders replacement ratios. https://doi.org/10.1371/journal.pone.0292951.g007 3.2. Buildability evaluation In order to quantitatively analyze the buildability of concrete materials, two important geometric factors including vertical strain εH and aspect ratio Ahw for different limestone powders replacement ratios were measured, as illustrated in Fig 8(A). According to the experimental results, it can be observed that the samples T2-T6 display good buildability because the layers of model are well stacked and stable. The sample T1 cannot be extruded due to the fact that the concrete materials are too thick, while the samples T7 and T8 cannot be shaped because the corresponding concrete materials are too thin. Moreover, with the increase of limestone powders contents, the vertical strain εH significantly increases. Besides, the aspect ratio Ahw is highly dependent on the replacement ratios of limestone powders, which indicates that the structural or geometric stability of the printed structures are obviously varying with the dosage of limestone. Although all of the samples T2-T6 are capable of being shaped, it is obvious that the lower the vertical strain and the larger the aspect ratio, the better the buildability of the samples owing to the higher quasi-static shape retention. Among them, the aspect ratios of all samples are close, but the vertical strains of T2, T3, T4, T6 are quite different as well as the cracks appear on their surfaces compared to sample T5. Therefore, it is accepted that the sample T5 displays optimal buildability, which causes its preferable selection for the 3D printing. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 8. (a) Vertical strain and aspect ratio versus limestone powders replacement ratio. (b) Shrinkage ratio as functions of the time for different concrete samples T1 to T7. (c) Extruded filaments of samples T1 to T8 after 180 min. https://doi.org/10.1371/journal.pone.0292951.g008 3.3. Shrinkage evaluation Shrinkage results can provide supplementary for the evaluation of the stability of concrete materials. In general, the smaller shrinkage ratios at the specific time, the greater the deformation resistance of structure, indicating a better stability of the concrete materials. As shown in Fig 8(B), different shrinkage ratios of the concrete materials are investigated. It is obvious that the shrinkage properties are highly dependent on the limestone powders replacements ratio. When the replacements ratios are varied from 0 to 40%, the shrinkage properties significantly decrease. It may be due to the fact that the limestone powders are filled into the tiny voids between the original particles, thus improving the compactness of the mixtures and decreasing its shrinkage properties. Besides, the shrinkage properties increase slightly as the limestone powders further increase. And the shrinkage ratios increase from 0.505 to 0.82 when the limestone powders increase from 40% to 60%, revealing a reduction in the stability performance. Specially, the shrinkage ratio of filament T8 cannot be measured because it breaks during the test. Hence, the additions of excessive limestone powders can result in unreasonable grading of solid particles in the mixture, which makes a large number of fine powders filled with water and compressed after being dried, thus increasing the shrinkage properties of materials. From the foregoing, replacing natural sand with proper limestone powders is capable to improve the compactness of concrete materials, which is of great importance to reinforce the deformation resistance and stability of the printed structures. 3.4. Layer-interface properties evaluation and microstructure analysis The extrusion-based 3D printing technology is stacked layer by layer during the actual printing, which inevitably leads to a weak interface between the layers, thus seriously reducing the mechanical properties of the printed samples. It is well known that the binding strengths between the layers are closely related to the characteristics of the layer interfaces [30]. As shown in Fig 9, with the increase of limestone powders contents, the workability of slurry is improved as well as the surface on the filament is smoother, and the layers are easier to deform and match with the upper layer. This is the reason that the porosity of macroscopic interface gradually becomes lower and narrower for samples T1-T5. The gradual elimination of the voids contributes to the improvement of interlayer bonding strength, which can enhance the mechanical properties of the samples. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 9. Layer-interface properties of different concrete materials. https://doi.org/10.1371/journal.pone.0292951.g009 Nevertheless, with the further increase of limestone powders replacement ratios (greater than 40%), the macroscopic voids begin to increase, and the natural sand particles that supported the framework are gradually decreased. The augment in the voids may be due to the increase of capillary pressure between fine particles as the amount of limestone powder contents increase, resulting in a plastic shrinkage of the materials and more microcrack on their surface. Additionally, some researchers consider that the increase of limestone powder contents leads to an increase in the viscosity of the slurry, and the bubbles in the slurry are unable to escape, thereby increasing their macroscopic voids. As shown in Fig 9, the increasing of limestone powder contents inevitably results in a decrease in the bonding strengths for samples T6-T8. In order to further evaluate the properties of the sample T5, its microstructure is characterized and compared to the reference sample T1, as shown in Fig 10. It is found that the microstructure of sample T5 is denser and its interfacial transition zone is less obvious compared to the reference sample T1. So, it can be summarized that the concrete material T5 possesses the denser and more optimized structures. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 10. SEM images of the concrete (a) sample T1 (reference group) and (b) sample T5. https://doi.org/10.1371/journal.pone.0292951.g010 3.5. Compressive strength evaluation Fig 11 shows the compressive strength of samples with various limestone powders replacement ratios after 3, 14 and 28 days. The average strengths of samples T1-T8 after 3 days are 24.45MPa, 24.49MPa, 27.96MPa, 32.59MPa, 32.62MPa, 25.75MPa, 22.09MPa and 6.32MPa, respectively. Clearly, the compressive strength significantly increases as the limestone powders contents increase from 0 to 40%. Analogously, the fine limestone powders contents have a positive effect on the compressive properties of the concrete materials when the replacement ratios are no more than 40% from the results after 14 and 28 days. The compressive strengths of the sample T5 with limestone powders contents of 40% at the range of 14 and 28 days are 42.17MPa and 50.39MPa, respectively, which are 30.3% and 21.7% higher than the sample T1. Hence, the limestone powder acts as natural fine sand in 3D printing concrete, which plays a filling role in the mesostructure and changes the particle grading of the materials. When the less limestone powders are added, a large number of tiny voids between natural sand particles cannot be completely filled, resulting in a lower mechanical strength. With increasing of limestone powder contents, the proper additions of particles are filled into the voids between original particles, and the fresh density and hardened density are improved [11]. So, a stable spatial skeleton structure is generated by the mutual contact of coarse particles, which improves the mechanical strength of the hardened samples. Besides, this may also be due to the fact that the stable mesostructure decreases the shrinkage and cracking of the specimens during the curing process and then results in a higher mechanical strength for the concrete materials to the external compressive load. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 11. Compressive strength of samples with various limestone powders contents after 3, 14 and 28 days. https://doi.org/10.1371/journal.pone.0292951.g011 Whereas, the mechanical properties of printing materials are reduced to a certain extent as the limestone powder contents are greater than 40%. When the limestone powders replacement ratios increase from 60% to 100%, their compressive strength after 3 days decreases by about 9.7% and 74.2% compared with sample T1. With the further increase of limestone powder and decrease of natural sand contents, a small number of natural sand particles are suspended in the limestone powder, which makes them insufficient to contact each other to require a stable skeleton structure and limits the improvement of mechanical strength. More importantly, the excessive limestone powder contents could not only weaken the framework effect, but also hinder the hydration of the cement, which blocks the sustainable development of the mechanical strengths [19]. Therefore, the preferable proportions of limestone powders, fly ash, Portland cement in combination with natural sand are conducive to strength developments of the concrete materials. Specifically, the variations depend on the factors including the rheology, buildability and grading distribution of the concrete materials, etc. 3.6. Interlayer bonding strength evaluation Interlayer bonding strength is a crucial parameter that assessing the structural strength of printed samples and closely connected with the buildability of the concrete materials. Fig 12 presents the variations of splitting tensile strength with the limestone powders replacement ratios. It can be seen that the splitting tensile strengths display an increasing tendency with the augment of limestone contents based on the stronger adhesion forces. However, the higher replacement ratios do not always perform better. The average tensile strengths of samples T5, T6, T7 and T8 are 5.32MPa, 5.17MPa, 5.03MPa and 4.15MPa, respectively. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 12. Comparison of the splitting tensile strength for different concrete samples. https://doi.org/10.1371/journal.pone.0292951.g012 It may be attributed to the weaken interfaces and adhesion reduction problems between the adjacent layers caused by the retention of tiny bubbles as well as the increase of microcracks, which causes the loss of tensile strengths. Therefore, it is recommended to replace the natural sand with 40% limestone powders contents in the manufacturing process so as to enhance the structural strength and buildability of the printed structures. In conclusion, the sample T5 exhibits preferable comprehensive characteristics, including printability, buildability, mechanical properties and so on, demonstrating that the concrete material with the limestone powders replacement ratio of 40% has a broad promising in the field of 3D printing. 3.1. Flowability evaluation Fig 7(A) and 7(B) display the slump and jumping table test results of concrete materials with different limestone powders replacement ratios from 10min to 90min, respectively. A larger slump value and spreading diameter result in an increased flowability and vice versa. Remarkably, with the increase of limestone powders replacement ratios, the flowability at the specific time improves. At the time of 20 min, the slump in heights of T1-T7 is 3.67cm, 4.40cm, 6.15cm, 7.34cm, 8.18cm, 9.23cm and 9.35cm, respectively. In general, it could be considered to display better flowability when the slump of the mixture slurry falls into the range of 3.5cm to 8cm. Therefore, substituting natural sand with 40% limestone powders (sample T5) increases the fluidity by 1.23% compared to the reference sample T1. While their corresponding spreading diameters in the jumping table test are 18.32cm, 18.50cm, 18.79cm, 19.96cm, 20.26cm, 21.05cm, and 21.66cm of T1-T7 at the rest time of 20min. In general, it could be considered to display better flowability when the jumping table test of the mixture slurry falls into the range of 17cm to 20.5cm. So the spreading diameter of sample T5 is improved by 10.59% compared to sample T1, which is consistent with the slump test analysis. Additionally, the fluidity displays a decrease tendency with the elevation of rest time. It is due to the fact that a longer rest time is capable of improving the hydration degree of binding materials, which leads to a decrease in the free water of slurry and an increase in static yield stress, thus resulting in the reduced flowability of the materials. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 7. (a) Slump and (b) spreading diameter dependence of rest time for concrete materials with various limestone powders replacement ratios. https://doi.org/10.1371/journal.pone.0292951.g007 3.2. Buildability evaluation In order to quantitatively analyze the buildability of concrete materials, two important geometric factors including vertical strain εH and aspect ratio Ahw for different limestone powders replacement ratios were measured, as illustrated in Fig 8(A). According to the experimental results, it can be observed that the samples T2-T6 display good buildability because the layers of model are well stacked and stable. The sample T1 cannot be extruded due to the fact that the concrete materials are too thick, while the samples T7 and T8 cannot be shaped because the corresponding concrete materials are too thin. Moreover, with the increase of limestone powders contents, the vertical strain εH significantly increases. Besides, the aspect ratio Ahw is highly dependent on the replacement ratios of limestone powders, which indicates that the structural or geometric stability of the printed structures are obviously varying with the dosage of limestone. Although all of the samples T2-T6 are capable of being shaped, it is obvious that the lower the vertical strain and the larger the aspect ratio, the better the buildability of the samples owing to the higher quasi-static shape retention. Among them, the aspect ratios of all samples are close, but the vertical strains of T2, T3, T4, T6 are quite different as well as the cracks appear on their surfaces compared to sample T5. Therefore, it is accepted that the sample T5 displays optimal buildability, which causes its preferable selection for the 3D printing. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 8. (a) Vertical strain and aspect ratio versus limestone powders replacement ratio. (b) Shrinkage ratio as functions of the time for different concrete samples T1 to T7. (c) Extruded filaments of samples T1 to T8 after 180 min. https://doi.org/10.1371/journal.pone.0292951.g008 3.3. Shrinkage evaluation Shrinkage results can provide supplementary for the evaluation of the stability of concrete materials. In general, the smaller shrinkage ratios at the specific time, the greater the deformation resistance of structure, indicating a better stability of the concrete materials. As shown in Fig 8(B), different shrinkage ratios of the concrete materials are investigated. It is obvious that the shrinkage properties are highly dependent on the limestone powders replacements ratio. When the replacements ratios are varied from 0 to 40%, the shrinkage properties significantly decrease. It may be due to the fact that the limestone powders are filled into the tiny voids between the original particles, thus improving the compactness of the mixtures and decreasing its shrinkage properties. Besides, the shrinkage properties increase slightly as the limestone powders further increase. And the shrinkage ratios increase from 0.505 to 0.82 when the limestone powders increase from 40% to 60%, revealing a reduction in the stability performance. Specially, the shrinkage ratio of filament T8 cannot be measured because it breaks during the test. Hence, the additions of excessive limestone powders can result in unreasonable grading of solid particles in the mixture, which makes a large number of fine powders filled with water and compressed after being dried, thus increasing the shrinkage properties of materials. From the foregoing, replacing natural sand with proper limestone powders is capable to improve the compactness of concrete materials, which is of great importance to reinforce the deformation resistance and stability of the printed structures. 3.4. Layer-interface properties evaluation and microstructure analysis The extrusion-based 3D printing technology is stacked layer by layer during the actual printing, which inevitably leads to a weak interface between the layers, thus seriously reducing the mechanical properties of the printed samples. It is well known that the binding strengths between the layers are closely related to the characteristics of the layer interfaces [30]. As shown in Fig 9, with the increase of limestone powders contents, the workability of slurry is improved as well as the surface on the filament is smoother, and the layers are easier to deform and match with the upper layer. This is the reason that the porosity of macroscopic interface gradually becomes lower and narrower for samples T1-T5. The gradual elimination of the voids contributes to the improvement of interlayer bonding strength, which can enhance the mechanical properties of the samples. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 9. Layer-interface properties of different concrete materials. https://doi.org/10.1371/journal.pone.0292951.g009 Nevertheless, with the further increase of limestone powders replacement ratios (greater than 40%), the macroscopic voids begin to increase, and the natural sand particles that supported the framework are gradually decreased. The augment in the voids may be due to the increase of capillary pressure between fine particles as the amount of limestone powder contents increase, resulting in a plastic shrinkage of the materials and more microcrack on their surface. Additionally, some researchers consider that the increase of limestone powder contents leads to an increase in the viscosity of the slurry, and the bubbles in the slurry are unable to escape, thereby increasing their macroscopic voids. As shown in Fig 9, the increasing of limestone powder contents inevitably results in a decrease in the bonding strengths for samples T6-T8. In order to further evaluate the properties of the sample T5, its microstructure is characterized and compared to the reference sample T1, as shown in Fig 10. It is found that the microstructure of sample T5 is denser and its interfacial transition zone is less obvious compared to the reference sample T1. So, it can be summarized that the concrete material T5 possesses the denser and more optimized structures. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 10. SEM images of the concrete (a) sample T1 (reference group) and (b) sample T5. https://doi.org/10.1371/journal.pone.0292951.g010 3.5. Compressive strength evaluation Fig 11 shows the compressive strength of samples with various limestone powders replacement ratios after 3, 14 and 28 days. The average strengths of samples T1-T8 after 3 days are 24.45MPa, 24.49MPa, 27.96MPa, 32.59MPa, 32.62MPa, 25.75MPa, 22.09MPa and 6.32MPa, respectively. Clearly, the compressive strength significantly increases as the limestone powders contents increase from 0 to 40%. Analogously, the fine limestone powders contents have a positive effect on the compressive properties of the concrete materials when the replacement ratios are no more than 40% from the results after 14 and 28 days. The compressive strengths of the sample T5 with limestone powders contents of 40% at the range of 14 and 28 days are 42.17MPa and 50.39MPa, respectively, which are 30.3% and 21.7% higher than the sample T1. Hence, the limestone powder acts as natural fine sand in 3D printing concrete, which plays a filling role in the mesostructure and changes the particle grading of the materials. When the less limestone powders are added, a large number of tiny voids between natural sand particles cannot be completely filled, resulting in a lower mechanical strength. With increasing of limestone powder contents, the proper additions of particles are filled into the voids between original particles, and the fresh density and hardened density are improved [11]. So, a stable spatial skeleton structure is generated by the mutual contact of coarse particles, which improves the mechanical strength of the hardened samples. Besides, this may also be due to the fact that the stable mesostructure decreases the shrinkage and cracking of the specimens during the curing process and then results in a higher mechanical strength for the concrete materials to the external compressive load. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 11. Compressive strength of samples with various limestone powders contents after 3, 14 and 28 days. https://doi.org/10.1371/journal.pone.0292951.g011 Whereas, the mechanical properties of printing materials are reduced to a certain extent as the limestone powder contents are greater than 40%. When the limestone powders replacement ratios increase from 60% to 100%, their compressive strength after 3 days decreases by about 9.7% and 74.2% compared with sample T1. With the further increase of limestone powder and decrease of natural sand contents, a small number of natural sand particles are suspended in the limestone powder, which makes them insufficient to contact each other to require a stable skeleton structure and limits the improvement of mechanical strength. More importantly, the excessive limestone powder contents could not only weaken the framework effect, but also hinder the hydration of the cement, which blocks the sustainable development of the mechanical strengths [19]. Therefore, the preferable proportions of limestone powders, fly ash, Portland cement in combination with natural sand are conducive to strength developments of the concrete materials. Specifically, the variations depend on the factors including the rheology, buildability and grading distribution of the concrete materials, etc. 3.6. Interlayer bonding strength evaluation Interlayer bonding strength is a crucial parameter that assessing the structural strength of printed samples and closely connected with the buildability of the concrete materials. Fig 12 presents the variations of splitting tensile strength with the limestone powders replacement ratios. It can be seen that the splitting tensile strengths display an increasing tendency with the augment of limestone contents based on the stronger adhesion forces. However, the higher replacement ratios do not always perform better. The average tensile strengths of samples T5, T6, T7 and T8 are 5.32MPa, 5.17MPa, 5.03MPa and 4.15MPa, respectively. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 12. Comparison of the splitting tensile strength for different concrete samples. https://doi.org/10.1371/journal.pone.0292951.g012 It may be attributed to the weaken interfaces and adhesion reduction problems between the adjacent layers caused by the retention of tiny bubbles as well as the increase of microcracks, which causes the loss of tensile strengths. Therefore, it is recommended to replace the natural sand with 40% limestone powders contents in the manufacturing process so as to enhance the structural strength and buildability of the printed structures. In conclusion, the sample T5 exhibits preferable comprehensive characteristics, including printability, buildability, mechanical properties and so on, demonstrating that the concrete material with the limestone powders replacement ratio of 40% has a broad promising in the field of 3D printing. 4. Conclusions In this paper, the concrete materials with different replacement ratios of limestone powders to natural sand were investigated and compared to evaluate the printability in the field of 3D printing. The experimental measurements were performed including microstructures characteristics, flowability, buildability, shrinkability, layer-interface properties, mechanical properties and interlayer bonding strength. Besides, an effective method is proposed to characterize the printable properties of concrete materials and then the reasonable stone powder replacement ratio is determined. By comparison, the sample T5 (substituting natural sand with 40% limestone powders) could effectively improve the grading of the concrete, displaying an optimal printability and buildability. Besides, its mechanical property significantly improved by approximately 45%. The produced concrete not only greatly reduced the 3D printing costs and expanded the alternative range of materials, but also improved its inherent mechanical strength and printability, which can be treated as an eco-friendly building material that shows promising application in the field of sustainable 3D printing. Supporting information S1 Data. https://doi.org/10.1371/journal.pone.0292951.s001 (XLSX)
TI - Utilization of solid mine waste in the building materials for 3D printing
JF - PLoS ONE
DO - 10.1371/journal.pone.0292951
DA - 2023-10-19
UR - https://www.deepdyve.com/lp/public-library-of-science-plos-journal/utilization-of-solid-mine-waste-in-the-building-materials-for-3d-LRhZynLyng
SP - e0292951
VL - 18
IS - 10
DP - DeepDyve
ER -