Blending performance of helical static mixer for concrete printing

Cement and Concrete Composites

Volume 134, November 2022, 104741

Blending performance of helical static mixer used for twin-pipe 3D concrete printing

Author links open overlay panel Yaxin Tao ^a b, A.V. Rahul ^c, Manu K. Mohan ^a, Kim Van Tittelboom ^a, Yong Yuan ^d, Geert De Schutter ^a b

https://doi.org/10.1016/j.cemconcomp.2022.104741Get rights and content

Abstract

A twin-pipe pumping system has been developed to overcome the conflicting requirements in 3D concrete printing between pumping and deposition. In the twin-pipe pumping system, a helical static mixer, consisting of a series of mixing elements, is used to blend a cement-based mixture and a limestone-based mixture right before extrusion. As these two mixtures go through the static mixer, non-moving mixing elements continuously blend the materials in a flow-division pattern.

However, the blending performance of the helical static mixer used for twin-pipe 3D concrete printing has not yet been reported. This paper presents the results of an experimental study on the relation between the mechanical behavior and the blending efficiency of the helical static mixer. Based on the binary images of the polished specimens printed with a different number of mixing elements, the blending performance was characterized by the coefficient of variation of the row-wise intensity distribution. A reasonable linear relationship was established between the mechanical strength (flexural strength and tensile strength) and the mixing homogeneity of the two mixtures. In addition, a higher number of mixing elements was observed to lead to a more dense pore structure, most probably due to the better compaction of the material by the higher pumping pressure.

Introduction

3D concrete printing (3DCP) has progressed rapidly over the past years, benefiting from its automated manufacturing process, free of formwork, and high flexibility allowing the realization of complicated structures [[1], [2], [3]]. As the construction process is distinct from mold-cast concrete, the applications of 3D concrete printing come with several difficulties, among which the conflicting requirements between pumping and deposition might be the greatest challenge [[3], [4], [5]]. Without any accelerating process, the maximum yield stress that can be achieved usually does not allow the structure to be built much higher than about three-quarters of 1 m. Several approaches can be used to increase the stiffening rate, such as the use of high-strength cement, decreasing the water-to-cement ratio, or the use of an accelerator.

However, these mixtures may cause problems while pumping due to very limited opening time. For example, decreasing the water-to-cement ratio can cause an increase in solid volume fraction, thereby increasing yield stress and plastic viscosity leading to an increased pumping pressure [6], [7], [8]].

Instead, injecting an accelerator close to the nozzle with an injection quill and waking up the ‘sleeping’ concrete right before extrusion via inline mixing is an alternative solution. As such, the accelerator would soon take effect in contact with the cement and give rise to a fast stiffening process after extrusion. Different types of inline mixing systems have been developed by researchers and most of them adopt a dynamic mixer with rotational blades and screws [9,10].

However, this leads to a complicated design and the potential formation of dead zones inside the mixing chamber. Moreover, previous research has demonstrated that merely injecting a liquid accelerator into fresh concrete can cause the blockage of the accelerator inlet or backflow of the liquid accelerator [6]. Considering the drawbacks of using a dynamic mixer and merely injecting liquid accelerator, a specially designed twin-pipe pumping (TPP) system has been developed by the authors, where a helical static mixer is used and a limestone-based mixture is used as a carrier material for the accelerator [11,12]. During the TPP process, two mixtures including a cement-based mixture (without accelerator) and a limestone-based mixture (without cement but with a high dosage of accelerator) are delivered via two pumps and blended via a helical static mixer just before extrusion through the nozzle. In this system, the limestone mixture is used as a ‘carrier fluid’ for the accelerator.

For a smooth pumping operation, the cement-based mixture is designed to have a high open time (over 2 h) while the limestone-based mixture has an indefinite open time as there is no cement present in this mixture. However, once the two mixtures combine in the static mixer, the accelerator comes in contact with the cement resulting in rapid ettringite formation and fast stiffening [6,11].

Although this system is primarily designed for Portland cement, recently this was also applied to calcium sulfoaluminate (CSA) cement systems by the same authors [13]. A 1.5 m high column with an internal diameter of 30 cm made by using this calcium sulfoaluminate cement mixture was printed in a very short duration of less than 10 min.

Such a high stiffening rate after extrusion allows the printing of overhanging structures [14]. In addition, the concept of using two streams during printing can also be employed to create functionally graded concrete materials [15].

The mechanical performance of printed concrete in the hardened state is very different from that of mold-cast concrete. Previous research has shown that printed concrete in the hardened state presents anisotropic behavior, caused by the weak interface between two printed layers [16,17].

Moreover, the structural performance is influenced by process parameters (e.g. the time gap between two layers, the printing speed, and the layer thickness) [18,19] and environmental conditions during and immediately after printing (e.g. temperature and relative humidity) [[20], [21], [22]].

The loss of surface moisture before the deposition of the following layer might be the most prominent factor controlling the layer interface bond of printed concrete [23,24]. Other factors such as air entrapment between the layers and the layer surface roughness were also pointed out to influence the layer interface bond and associated mechanical performance of printed concrete [[25], [26], [27]].

Unlike conventional printed concrete, another weak region can exist in elements printed by using the TPP system. The mixing of two different fluids in a static mixer can result in the formation of striations due to the flow divisions occurring in the helical static mixer (see Fig. 1).

The striations consisting of the limestone layer, where no binder is present, can become a weak zone adversely influencing the mechanical behavior [28]. However, the influence of mixing homogeneity on the mechanical performance of printed concrete has not been systematically investigated in previous studies.

Therefore, methods need to be formulated to assess the mixing homogeneity that can be achieved by using the helical static mixer and to evaluate its influence on the mechanical performance. This is essential to ensure the adequate mechanical performance of printed concrete produced by TPP.

In the current study, we examine the influence of the number of mixing elements present in the static mixer on the mechanical behavior of printed concrete using the TPP technology. Helical static mixers with a varying number of mixing elements were used to print straight walls. Prisms, cubes, and cylinders were extracted from the printed walls for flexural, compression, and tensile testing.

Moreover, mixing homogeneity was evaluated based on an image analysis technique and correlated to the mechanical strength of the printed concrete. Finally, the pore structure of printed concrete was studied by using mercury intrusion porosimetry (MIP) to better understand the blending performance of the helical static mixer.

Materials and mixing procedure.

In this study, calcium sulfoaluminate cement (CSA cement, i.tech ALI CEM GREEN® by Italcementi, Italy) and limestone powder (Calcitec 2001 S from Carmeuse, Belgium) were used. The chemical composition and loss on ignition of calcium sulfoaluminate cement and limestone powder are shown in Table 1. The specific gravity of calcium sulfoaluminate cement and limestone powder was 3150 kg/m³ and 2710 kg/m³, respectively. Silica sand with a maximum particle size of 2 mm and a specific gravity of

Flexural strength

The flexural strength of the prismatic specimens for different loading directions is shown in Fig. 6. An anisotropic behavior was observed, i.e., flexural strength was found to depend on the loading direction. For example, the flexural strength of the prismatic specimens produced with 18 mixing elements was 1.6, 9.7, and 8.6 MPa for the loading directions F1, F2, and F3, respectively. As also reported in the previous work of the authors, limited hydration occurred inside the limestone-based.

Conclusions

The blending performance of the helical static mixer used for twin-pipe 3D concrete printing is investigated in this study.   According to the results and discussions, the following conclusions can be drawn:

Mixing elements used in the helical static mixer led to the increase of the mechanical strength, especially of the flexural strength (loading direction F1) and the tensile strength. Moreover, less fracture was observed in the printed specimens produced by more mixing elements.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

The authors acknowledge the financial support provided by the BOF starting grant (BOF.STG.2018.0017.01.) and by the Ministry of Science and Technology of China (No. 2021YFE0114100).

References (46)

• B. Khoshnevis, Automated construction by contour crafting—related robotics and information technologies, Autom. ConStruct., (2004)
• P. Bosscher et al. Cable-suspended robotic contour crafting system, Autom. ConStruct., (2007)
• M.K. Mohan et al.Extrusion-based concrete 3D printing from a material perspective: a state-of-the-art review Cem. Concr. Comps.(2021)
• D. Marchon et al.Hydration and rheology control of concrete for digital fabrication: potential admixtures and cement chemistry, Cement Concr. Res.(2018)
• L. Reiter et al.Setting on demand for digital concrete – principles, measurements, chemistry, validation Cement Concr. Res. (2020)
• Y. Tao et al. Stiffening control of cement-based materials using accelerators in inline mixing processes: possibilities and challenges Cem. Concr. Comps. (2021)
• D. Jiao et al. The structural build-up of cementitious paste with nano-Fe3O4 under time-varying magnetic fields Cement Concr. Res. (2019)
• M.K. Mohan et al. Rheological and pumping behavior of 3D printable cementitious materials with varying aggregate content Cement Concr. Res. (2021)