Additive Manufacturing of Sustainable Construction Materials and Form-finding Structures: A Review on Recent Progresses

Background

The OPC-based concrete remains popular among material selection for Additive Manufacturing (AM). However, recently there is an increasing interest in studying the properties of sustainable printable concrete in which OPC is partially replaced by supplementary cementitious material (SCM). SCM includes fly ash (FA), silica fume (SF), ground blast furnace slag (slag), rice husk ash etc. Replacing OPC with SCM, particularly FA, contributes to lowering CO₂ emission since the cement production leads to massive CO₂ generation. ³⁷ In addition, the limestone calcined clay cement (LC ³ ), a ternary blended cement, comprising calcined clay, limestone, and OPC, has also been utilized in 3DCP to promote the sustainability.^38–40 Table 2 presents a summary of the mixture design for OPC-based sustainable materials from literature. The ongoing research progress in 3DCP of OPC-based mixes shows that researchers have mostly used SCM to modify the material rheology, and in some instances, chemical admixtures, as well as nano materials, such as nanoclay have been used to tune extrudability and buildability properties.

Effect of SCM on extrusion rheology and buildability

The addition of SCM into OPC-based cementitious materials has different effects on the extrusion rheology and green strength depending on the types of SCM. Panda and Tan ⁴¹ found that increasing FA content from 50% to 80% led to the decrease in static yield stress and viscosity, whereas increasing SF dosage to 5% gave opposite results (Fig. 3a, b). The addition of FA is in practice commonly performed by mass replacement of cement, which results in change in paste volume and thus the rheology. Therefore, changes in rheology will depend on both the variable paste volume fraction and the properties of the FA. The negative influence of FA can also be linked to the particle shape effect, ⁴² while the SF-induced improvement can be explained by the large surface area of SF particles compared to FA and OPC. Meanwhile, SF speeds up stiffening development of cementitious materials because of its ionization of SF surfaces and the potential ion bridging effect, which both promote the C-S-H gel formation. ⁴¹ Yuan et al. ⁴³ found that the addition of 5% SF nearly doubled the buildup rate compared with pure cement mortar, while the incorporation of 20% FA and 30% slag slowed down the yield stress development rate (Fig. 3c).

OPEN IN VIEWER

FIG. 3.

Effect of FA and SF on extrusion rheology of OPC-based sustainable concrete: (a) static yield stress; (b) apparent viscosity (both reproduced from Ref. ⁴¹ ); (c) Effect of SCMs and water/cement ratio (w/c) on structure buildup rate ⁴³ ; (d) effect of nSiC on structure buildup rate at reflocculation stage (R_thix) and structuration stage (A_thix) ⁴⁸ (reproduced from original data). FA, fly ash; nSiC, nano silica carbide; OPC, ordinary Portland cement; SCM, supplementary cementitious material; SF, silica fume.

Effect of additives on extrusion rheology and buildability

Different categories of additives have been added to tailor the rheology and fresh behaviors of OPC-based sustainable concrete. Panda et al. ⁴⁴ studied the effect of nanoclay on the rheology of cementitious materials with 70% FA, and they found that nanoclay up to 0.5% significantly increased both static yield stress and viscosity. In another study, the structure buildup rate of cementitious material with nearly 70% FA was improved by 50% with the addition of 0.5% nanoclay. ⁴⁵ The increase in rheological properties can be explained by the electrical attraction force induced by the oppositely charged nanoclay surface, which densified the microstructures. ⁴⁶ The improved structural buildup is attributed to the additional nucleation sites provided by nanoclay, in which the hydration process is accelerated as evidenced by the higher rate of hydration heat and cumulative heat release.^44,47 Van den Heever et al. ⁴⁸ studied the effect of nano silica carbide (nSiC) on the yield strength development of OPC-based material incorporated with FA (20%) and SF (10%). Interestingly, despite the positive correlation between nSiC dosage and yield stress, the increasing dosage of nSiC resulted in a decreased value of A t h i x (rate of yield strength development at structuration stage) (Fig. 3d), indicating that the buildability was negatively influenced by nSiC.

Extrusion rheology and buildability of LC ³ -based material

For LC ³ -based sustainable material, research has shown that the percentage of metakaolin (MK) in calcined clay plays a vital role in the fresh properties. Calcined clay with a high proportion of MK (about 75%) is termed high-grade calcined clay (HCC), whereas that with 40–50% MK is classified as low-grade calcined clay (LCC). According to Beigh et al., LC ³ -based concrete exhibited higher structure built-up rate compared with ordinary concrete despite the decrease in flowability, and the improvement in yield strength development was more profound in concrete with HCC (Fig. 4a). ³⁸ Similar results were reported by Chen et al. ³⁹ who found that materials with a higher proportion of HCC instead of LCC exhibited shorter initial setting time and higher growth rate of green strength (Fig. 4b), indicating an improved buildability. This result was explained by the higher surface area of MK particle nucleation sites to accelerate early-age cement hydration. ⁴⁹ Meanwhile, some extra MK particles in the concrete matrix accelerated the phase transition from flocculation to structuration, ³⁹ resulting in a decreased setting time.

OPEN IN VIEWER

FIG. 4.

(a) Comparison between LC ³ -based material and ordinary concrete regarding flowability and structure buildup rate ³⁸ ; (b) effect of calcined clay grade on the green strength development of LC ³ -based material. ³⁹ LC ³ , limestone calcined clay cement.

Mechanical anisotropy and fiber effect

Due to the inherent features of layer-by-layer stacking, the printed objects exhibit directional dependence on mechanical properties, known as mechanical anisotropy. For OPC-based sustainable material, this property has also been investigated. As illustrated in Figure 5a, the defined directions include: (1) D1, loading is parallel to the printing direction, that is, the direction of the extrusion (longitudinal load); (2) D2, loading is parallel to the layer stacking direction (vertical load); and (3) D3, loading is perpendicular to both D1 and D2 (lateral load). Paul et al. ³³ reported that for the printed filaments with 60% OPC substituted by SCM, the scale of mechanical anisotropy was more significant at 28 days than 7 days (Fig. 5b). In contrast, Rahul et al. ⁵⁰ investigated the hardened properties of mortar with 25% FA and 5% SF. They found that the printed specimens exhibited little anisotropy in compressive strength, whereas the compressive strength in all directions was 12–20% lower than the cast counterparts.

OPEN IN VIEWER

FIG. 5.

(a) Definitions of loading direction for investigating mechanical anisotropy; (b) comparison of mechanical anisotropy of the printed OPC-based sustainable specimens at 7 and 28 days ³³ (reproduced from original data); fiber effect on the mechanical properties of printed OPC-based sustainable specimens; (c) flexural strength and uniaxial tensile strength; (d) tensile strain capacity and strain energy (both reproduced from Ref. ⁵¹ ).

Limited publications are found regarding the effect of fibers on hardened mechanical properties of OPC-based sustainable materials. Zhu et al. ⁵¹ investigated the impact of polyethylene (PE) fibers on the mechanical properties of printed specimens with about 60% of cement replaced by FA. They found that the flexural strength was improved as PE fibers increased to 2% (Fig. 5c). The addition of PE fibers also increased the specimen resistance against tensile loading and ductility (Fig. 5d), which was attributed to the increasing proportion of fibers intersecting cracks that improved the overall crack-bridging force.

Interlayer bond strength and printing parameters

The interlayer bonding strength between deposited concrete filaments is a crucial feature distinguishing 3DCP from cast concrete. The layer adhesion is generally influenced by printing parameters. For OPC-based sustainable material, the interlayer bonding strength was investigated by changing the print time gap ⁴⁴ and standoff distance (distance between the nozzle and top of filament). ⁴⁵ As reported in Panda et al., ⁴⁴ for a printed layer with 70% FA, increasing the time gap from 5 to 30 min led to a significant drop in tensile bond strength (Fig. 6a), which was attributed to the higher moisture loss from the substrate. Another study focused on the adhesion behavior of 20-mm thick filaments with 70% cement replacement and reported that decreasing standoff distance by 5 mm improved the layer bonding by up to 30% (Fig. 6b). This can be explained by the overlay compaction caused by lowering nozzle height, which generated a sufficient stress by overlay layer such that it could initiate the flow into the substrate layer which triggered an intermix between two adjacent layers. ⁴⁵

OPEN IN VIEWER

FIG. 6.

Effect of print parameters on OPC-based sustainable material: (a) time gap ⁴⁴ ; (b) standoff distance ⁴⁵ (reproduced from original data).

Background

Alkali-activated material refers to such material where a solid aluminosilicate undergoes chemical reaction under alkaline conditions, resulting in the production of hardened binder based on a combination of hydrous alkali-aluminosilicate and/or alkali-alkali earth-aluminosilicate phases. Furthermore, additional terminology applied to the alkali-activated material includes “geopolymer” specifically describing low-calcium alkali-activated aluminosilicate binders, although this nomenclature has been sometimes used indiscriminately. ⁵² From point of view of sustainability, alkali-activated/geopolymer materials can replace ordinary concrete and thus help minimize the global CO₂ emissions. Meanwhile, the rheological properties of geopolymer materials throughout the geopolymerization process have also been studied. ⁵³ Nowadays, 3D-printable alkali-activated/geopolymer material also gets increasing popularity among researchers compared to other cementitious sustainable materials. Table 3 provides the information of mixture design of 3D-printable alkali-activated/geopolymer from different publications.

Effect of raw material on extrusion rheology and buildability

The mixing proportions of raw materials in printable geopolymer, including FA, slag, SF, and so on, can significantly affect the fresh properties with alkali activators unmodified. According to Panda et al., ⁵⁴ when the proportion of slag replacing by FA (by mass) increased from 15% to 40%, the static yield stress and viscosity of materials increased by 80% and 20%, respectively (Fig. 7a). The results were explained by change in paste volume fraction as a result of scaling the mixture and angular shape of slag particles that enhanced the interlocking effects, and the increased yield stress could improve the buildability of the printed layer. Thus, it would be preferable to determine the optimum replacement of the FA by volume, thus keeping the paste volume fraction at a constant value. In the literature, however, most if not all of the investigations are performed utilizing a mass replacement. In another study, Alghamdi et al. ⁵⁵ investigated rheological properties of sodium alkali-activated FA-based material and found that replacing FA with limestone significantly decreased shear yield stress and viscosity (Fig. 7b).

OPEN IN VIEWER

FIG. 7.

(a) Effect of slag and FA on the rheology of geopolymer ⁵⁴ ; (b) effect of FA (F), slag (S), cement (C), and limestone (L) on the rheology of alkali-activated FA binder ⁵⁵ ; (c) effect of slag and SF on structure buildup rate of FA-based geopolymer ⁵⁷ ; (d) effect of slag and SF on thixotropy of FA-based geopolymer ⁵⁷ (all reproduced from original figures and data).

The concept of structural/thixotropic parameter (λ), which is an indicator of the thixotropic behavior of fresh material, was initially introduced by Roussel et al. ⁵⁶ A higher λ means a more profound thixotropic behavior, indicating better buildability. Panda et al. ⁵⁷ studied the effects of slag and SF on the structure buildup rate and thixotropy of FA-based geopolymer. They found that 10% slag outweighed 10% SF in terms of improving yield strength development in 10–30 min after mixing (Fig. 7c), which was attributed to the acceleration in the mixture hardening activated by slag particles. ⁵⁸ Meanwhile, both SF and slag were found to promote the thixotropy of geopolymer due to the higher value of λ, with slag exhibiting a more profound improvement than SF as the resting time increased (Fig. 7d).

Effect of alkali activators on extrusion rheology and buildability

Alkali activators play a critical role in the formation of geopolymer materials from the initial stage of precursor dissolution to the final step of polymerization, in which silicon and aluminum atoms are extracted from the source materials to generate polymeric Si-O-Al bonds. ⁵⁹ Panda et al. ⁶⁰ conducted investigations on the effects of molar ratio (MR) and activator solution-to-binder ratio on the rheology of geopolymer material. With a water-to-solid (w/s) ratio of 0.3, increasing MR from 1.8 to 2 resulted in a significant increase in both static yield stress and viscosity (Fig. 8a, b), which was explained by a higher activator viscosity with the increasing MR. The dominant influence of alkaline solutions is reflected in the experimental results illustrated in Figure 8c and d. Regardless of the change in MR, the static yield stress and viscosity consistently decreased as the activator solution-to-binder ratio increased, which was mainly attributed to the decrease in particle concentration.

OPEN IN VIEWER

FIG. 8.

Effect of molar ratio on the rheology of printable geopolymer: (a) static yield stress, (b) viscosity; effect of activator solution-to-binder ratio on the rheology of printable geopolymer: (c) static yield stress, (d) viscosity (all reproduced from Ref. ⁶⁰ ); effect of alkali type and mass ratio of silicate solution to hydroxide solution on the fresh properties of geopolymer: (e) flowability, (f) shape retention ability (both reproduced from Ref. ⁶² ).

Apart from MR, the rheological properties of geopolymer were also influenced by the type of alkali activator and the mass ratio of silicate solution (SS) to hydrate solution (HS). According to Bong et al., the Na-based activator was compared with K-based activator in terms of workability and shape retention of geopolymer for 3D printing.^61,62 The shape retention ratio (SRR) was defined to characterize the shape retention ability, and a higher SRR indicates that the extruded filament shows less deformation, thus exhibiting improved buildability.^61,62 In general, when the mass ratio of SS to HS was constant, the Na-based activator showed more positive effects on the flowability of fresh geopolymer compared with K-based activator. If the mass ratio of SS to HS increased, Na-based activators further improved the workability, whereas K-based activator decreased the flowability (Fig. 8e). Interestingly, K-based activator was found to outperform Na-based activator regarding shape retention because of higher SRR value under the category of K-based groups (Fig. 8f). One possible explanation could be the difference of alkaline solution viscosity depending on the type of alkali source. ⁶²

Effect of additives on extrusion rheology and buildability

The primary additive supplemented to printable geopolymers and alkali-activated binders is nanoclay, as reported in Refs.^63,60,64 As illustrated in Figure 9a, the addition of nanoclay improved both yield stress and viscosity. However, the positive effect of nanoclay on yield stress was much more profound than that on viscosity. Meanwhile, the added nanoclay was observed to enhance the recovery ability of fresh geopolymer soon after shearing was removed, ⁶⁰ indicating the better thixotropic behavior and improved buildability. Apart from nanoclay, other additives such as sodium carboxymethyl starch (CMS) and hydromagnesite seeds were also investigated. According to Sun et al., ⁶⁵ the addition of CMS (up to 8%) promoted both yield stress and viscosity at different rates (Fig. 9b), which could reduce the risk of segregation while avoiding filament collapse. However, the porosity of printed filaments rose with increasing CMS dosage, leading to weak internal structures and lower strength. In addition, the addition of 1–2% hydromagnesite seeds was found to exert minor influence on the rheological properties of the alkali-activated slag binders. ⁶⁴

OPEN IN VIEWER

FIG. 9.

(a) Effect of nanoclay on rheological properties of printable geopolymer ⁶⁰ ; (b) effect of CMS content on rheological properties of printable geopolymer ⁶⁵ (reproduced from original data). CMS, carboxymethyl starch.

Mechanical anisotropy and fiber effect

The anisotropy in compressive strength of printed geopolymer specimens was first investigated by Panda et al., ⁶⁶ who found that the compressive strength in D1 direction (longitudinal loading) was slightly higher than that in D2 and D3 directions (Fig. 10a). Unlike compressive strength, the flexural strength showed more significant directional effect since the printed specimens were found to be much stronger in resisting bending loads from D2 direction (perpendicular loading), compared with other two directions (Fig. 10b). This was attributed to the well compacted central area in which the maximum tensile stress commonly occurred. In contrast, due to tensile force exerted perpendicular to the layer interfaces, ⁶⁶ the printed geopolymer specimen was weak to resist flexural loading from longitudinal direction (D1).

OPEN IN VIEWER

FIG. 10.

Mechanical anisotropy in printed geopolymer specimens: (a) compressive strength ⁶⁶ ; (b) flexural strength^66,62 (both reproduced from original figures); Fiber effects on printed geopolymer blocks: (c) flexural strength in different directions; (d) uniaxial tensile strength (both reproduced from Ref. ⁶⁷ ).

Regarding the fiber effect, an investigation was conducted by Panda et al. ⁶⁷ to probe into relationships between glass fiber size/proportion and flexural strength. Apart from the positive correlation between glass fiber content and bending resistance, they also found that with increasing fiber length the improvement in flexural resistance was more profound (Fig. 10c). The results were attributed to an enhanced “bridging effect” by increasing fiber size such that the growth rate of microcracks initiated by tensile stress was slowed down. ⁶⁸ Similar results were also reported by Nematollahi et al. ⁶⁹ They found that the fracture energy of 3D-printed geopolymer specimens under three-point bending tests experienced a fivefold increase with polypropylene (PP) fiber content rising from 0% to 1%. Besides, the increasing glass fiber percentage also resulted in an improved resistance against longitudinal tensile forces (Fig. 10d).

Interlayer bond strength and printing parameters

Through limited publications on the study of layer adhesion of geopolymer specimens, it is found that the tensile bond strength between stacked geopolymer filaments is mainly influenced by printing parameters. Panda et al. ⁷⁰ found that increasing both time gap and standoff distance exerted much more negative impacts on layer bonding (Fig. 11a). The negative effects of time gap were mainly attributed to the inherent sticky feature of activator solution, which made the surface lubricating film most effective in exchanging moisture within a few minutes after extrusion. Influenced by polycondensation of geopolymer and moisture loss, the extruded filaments stiffened rapidly with air entrapped into layer surface, resulting in low bond strength.

OPEN IN VIEWER

FIG. 11.

Interlayer bonding of printed geopolymer filaments influenced by: (a) printing parameters ⁷⁰ ; (b) PP fiber content ⁶⁹ (both reproduced from the original figures). PP, polypropylene.

Fiber content was also found to have an impact on the bonding behaviors of geopolymer filaments. According to Nematollahi et al., ⁶⁹ the inclusion of PP fibers generally decreased the interlayer bond strength (Fig. 11b). They proposed a hypothesis that an increasing fiber content stiffened the geopolymer filament with a more porous surface, thus weakening the layer adhesion.

Background

Regarding the sustainable materials with other types of cement for 3DCP, there are two categories that have been studied so far, including sulfoaluminate cement (SAC) and calcium sulfoaluminate cement (CSAC). The main feature of these two types of cement is that their manufacturing process releases less CO₂ compared with the production of OPC,^71,72 which shows great potential in terms of moving 3DCP toward sustainability. Table 4 lists publications of 3D printing with the utilization of sustainable material containing SAC and CSAC.

Extrusion rheology and buildability

A study by Huang et al. suggested that the buildability of cementitious material containing CSAC could be influenced by the percentage of CSAC in the binder. ⁷³ They found an evident positive correlation between the ratio of CSAC replacing OPC and structure buildup rate at structuration stage (Fig. 12a). This result was explained by the positive effect of CSA on the hydration kinetics of hydrates and the formation of network between needle-like AFt and rod-like gypsum, which increased interparticle frictional force. ⁷³ The fresh property of CSAC-based materials for printing purposes can also be tailored through the supplement of additives. According to Chen et al., the addition of bentonite (up to 3%) significantly improved the thixotropic behavior and increased material viscosity simultaneously (Fig. 12b). The optimal bentonite dosage was determined to be 2% since the significant increase in viscosity due to the excessive amount of bentonite could cause blockage in the nozzle. ⁷⁴

OPEN IN VIEWER

FIG. 12.

(a) Relationship between CSA content and structure buildup rate ⁷³ ; (b) effect of bentonite content on the viscosity and thixotropy of CSAC-based material ⁷⁴ (reproduced from original data and figures); (c) effect of water/cement ratio and HPMC on the setting of printable SAC material; (d) effect of sand/cement ratio and HPMC on the flowability of printable SAC material (both reproduced from Ref. ⁷⁵ ). CSAC, calcium sulfoaluminate cement; HPMC, hydroxypropyl methylcellulose.

The fresh properties of concrete containing SAC can be influenced by water/cement ratio (W/C), sand/cement ratio, and the dosage of hydroxypropyl methylcellulose (Fig. 12c, d). ⁷⁵ Meanwhile, Chen et al. ⁷⁶ also found that the addition of either borax acid or sodium gluconate decreased yield stress and plastic viscosity, whereas the addition of diatomite showed adverse effects.

Hardened mechanical property

In general, research in the direction-related strength and fiber effect is still scarce. Regarding the interlayer adhesion, Ma et al. ⁷⁷ investigated the effects of cellulose fibers on the interlayer bond strength of printable materials containing up to 15% SAC, and they found that with different printing time interval the bond strength (both tensile and shear) between adhesive layers was improved with the increased proportion of cellulose fibers. They inferred that cellulose fibers helped to bridge the interface of layers and promoted the hydration due to internal curing induced by entrained water, which contributed to an improved layer adhesion behavior. In another study, the printable material with 15% SAC was further modified by two types of polymers (epoxy resin and chloroprene latex), and the test results showed an acceptable level of interlayer bond strength compared with the reference mixture without polymer modification. ⁷⁸

SIMP method

The SIMP algorithm was first proposed by Bendsøe ⁸⁹ to optimize structure with different design variables in a nondiscrete solution. The SIMP is also known as a method of topology optimization in virtue of density because the penalty factor does not allow fractional densities to occur during the optimum design process. Several complex concrete structures have been printed successfully using SIMP algorithm. A concrete topology-optimized slab (Fig. 14a) was developed with SIMP to reduce structure mass by 70% from the original solid concrete slab while having similar load-bearing capacity. ⁹⁰ Another example includes a 3DCP 4m-span girder with post-tensioning cable designed with SIMP algorithm, in which the volume of concrete is reduced by 20% compared to the original one and yielding a similar deflection. ¹²

BESO method

The BESO algorithm is a TOP method that was initially proposed to improve the convergence issue and results of both evolutionary structural optimization (ESO) and additive ESO by Querin et al. ⁹¹ Then, some other research groups updated the method to a new algorithm version to address compliance problems.^92,93 The BESO algorithm is also known as a discrete method that redundant material is iteratively cleaned up from an object while the efficient part is added simultaneously. The technique was utilized to optimize specific types of structures for either traditional or AM like shape optimization for underground openings, ⁹⁴ shell structure optimization, ⁹⁵ and topology optimization of 3D continuum structures. ¹³ In some cases, BESO method has been demonstrated to be less efficient than SIMP algorithm. ⁹⁶ Recently, Shao ⁹⁶ has proposed a hybrid method that combines the advantages of the two algorithms to achieve better efficiency in structural optimization with various types of structures for AM design.

Commercial structural optimization software

Apart from the abovementioned structural optimization algorithms (code based), there are other commercial softwares able to optimize the structural design for AM. Besides, the software tools can take into account constraints of extruding direction, minimum and maximum size, symmetry, and integrated postprocessing of outcomes. ⁹⁷ Among those, Abaqus topology optimization module (ATOM) is among the top commercial software packages capable of optimizing the structure for heat transfer and fluid flow problems. ⁹⁸ For example, a 3DCP beam optimized in terms of stiffness and volume has been conducted with the support of ATOM. ⁸⁸ OptiStruct software running on Hyperworks platform can be used for structural optimization for AM design with regard to component size, thickness, and other factors such as extrusion constraints and draw directions. ⁹⁹ A three-dimensional complex three-branch joint was optimized toward structural topology using the OptiStruct tool, which then was 3D printed. ¹⁰⁰

Finally, a recent propagation-based optimization algorithm known as Grasshopper tool—a commercial software—was developed by Robert McNeel & Associates. This software has been demonstrated as an excellent tool in addressing continuous optimization problems, yet structural optimization for binary problems cannot be resolved directly. ¹⁰¹ In structural optimization for 3D printing of concrete structures, there is a variety of required input data to evaluate its relations, that is, material properties, boundary conditions, and geometry. In terms of these considerations, the combination of Grasshopper and Abaqus software was an optimal choice in the study of Ref. ¹⁰² In addition, several specific CAD optimization tools for 3DCP,^103,104 which are exclusive of commercial packages, have been developed recently as well.

Bioinspiration concepts

In recent years, the bioinspired method has been commonly used in 3DCP field to manufacture some novel architectures.^{90,105,106,107} Based on this idea, some bioinspired architectures like a honeycomb cellular panel and Bouligand shapes (Fig. 15) were printed by cement material, which not only enhanced the crack issues but also reduced the failure and inelastic deflection in the structure by over 50% in comparison with traditional casting components. ¹⁰⁸ Besides, bioinspired cellular cementitious blocks that are naturally inspired with continuous nonself-intersecting surfaces known as triply periodic minimal surface (TPMS) structure were investigated to demonstrate their exceptional mechanical properties compared to traditional blocks with traditional lattice with the same volume fraction.^107,109, However, such kinds of structures normally contain overhang parts or curved surfaces that are difficult to be printed in a normal way by 3DCP technology.^90,109

OPEN IN VIEWER

FIG. 15.

3DCP structures based on bioinspired concepts of (a) honeycomb, (b) cellular sandwich panel, (c) Bouligand architecture with pitch angles 2° ¹⁰⁸ (with permission from John Wiley and Sons), and (d) Gyroid-TPMS cementitious structure using 3D printing mould ¹⁰⁹ (with permission from Elsevier). TPMS, triply periodic minimal surface.

Prefabricated and assembling

Although the capability of concrete printer has been developed to print various geometries, the manufacturing of a large-scale optimized structure is still a challenge because of its highly complex architecture. Dividing it into each smaller part to print then assembling them together seems to be a feasible solution for such optimized objects. The assembly process is normally carried out by digital devices like robot or by human workers. The assembly in 3DCP process is classified into 04 types, that is, assembling of printed parts to form the final structure, handling of printed element(s) to put it into the final one, and assembling of external object(s) after and during printing. ¹¹⁰ A problem in the assembling process of 3DCP elements is the interface bonding and tolerance between each separate section. The incompletely smooth connection between printed surfaces may cause local peek stresses. ¹¹¹ A research on the material used for binding each smaller part together in manufacturing a prestressed bridge was conducted (Fig. 16), in which synthetic epoxy was applied as the interface material. ¹¹¹ The bridge was assembled of many smaller 3DCP sections that were optimized in terms of bearing capacity.

OPEN IN VIEWER

FIG. 16.

(a) Before and (b) after assembly of a prestressed bridge's segment from separate 3DCP optimized sections.

3D printing formwork

Concrete casting using 3D printing formwork enables the construction industry to manufacture complex and optimized structure like lightweight concrete slab (Fig. 14a), which not only helps to save time and costs but also resources. The complexity, that is, curvatures, overhangs that mentioned earlier no longer plays a role with the support of such hybrid approach. Lightweight concrete blocks based on the bioinspired concept of TPMS structure were fabricated with the aid of 3D printed polylactic acid (PLA) moulds. ¹⁰⁹ However, using PLA or Acrylonitrile Butadiene Styrene (ABS) materials for printing formwork of concrete casting process is facing some problems as these kinds of materials are shrunk and brittle, which result in warping and transformation of final products. Another 3D printing formwork, fabric formwork that can overcome the abovementioned issues, has been studied, ¹¹² in which they used 3D printed fabric shuttering formwork for casting some complex concrete objects. In recently published work, Triveni et al. ¹¹³ have continued applying 3D printed PLA formwork in fabricating complex reinforced concrete walls. Two small-scale sample walls were obtained through the topology optimization algorithm to show the outperformance to standard walls in terms of stiffness and cracking.

Shotcrete

A new method of 3DCP was introduced to manufacture complex freeform architectures of concrete structures known as shotcrete 3D printing (SC3DP) technology, in which sprayed concrete was used for automated process to produce complex and optimized concrete components without formwork. ¹¹⁴ Compared to other conventional 3DCP methods, SC3DP technology offers some distinct benefits such as strong interlayer connection, printing with traditional reinforcement, and possible to print overhang parts. ¹¹⁵ Like other techniques, the SC3DP has its own challenge as variation of nozzle distance, delivery pressure, rate of concrete volume flow, application angle, etc. has considerable impact on the final product. Some limitations of using SC3DP techniques have been pointed out by Dressler et al. ¹¹⁶ as different material volumes need different kinetic energy to have the same compaction level, yet in the spraying process they receive the same amount of energy. This makes the printed structure surfaces rough, causes the interface tortuosity and, therefore, influences on the flexural strength of final products.