https://doi.org/10.3311/PPci.

37602
Creative Commons Attribution b |1
Periodica Polytechnica Civil Engineering

3D Printed Concrete: Fresh and Hardened Properties

Marwah M. Thajeel1*, György L. Balázs1

1
Department of Construction Materials and Technologies, Faculty of Civil Engineering, Budapest University of Technology and
Economics, Műegyetem rkp. 3., H-1111 Budapest, Hungary
*
Corresponding author, e-mail: [email protected]

Received: 04 June 2024, Accepted: 08 August 2024, Published online: 21 August 2024

Abstract
3D concrete printing (3D CP) is an advanced form of additive manufacturing (AM), that allows for the creation of complex geometrical
structures using concrete. This technology has the potential to reduce construction time and labour costs while minimizing material
waste. However, there are also challenges related to material properties, structural integrity, and standardization of construction
practices. The primary challenge in developing 3D printable cementitious materials lies in engineering specific fresh properties
that enable extrusion-based printing. Unlike cast concrete, where formwork provides dimensional stability during the printing
process, the absence of formwork in 3D printing necessitates fresh mixtures with low-slump and high-thixotropy to be suitable
for concrete 3D printing. Additionally, the orthotropic nature of the resulting 3D printed object due to the extrusion process
impacts its mechanical properties. Therefore, this research aims to study the effect of water content and loading directions on
3D printed concrete. A pre-mix material from Sika was used in this study. Three mixes were prepared by using three different
water to dry material (w/d) ratios, 0.140 l/kg, 0.145 l/kg and 0.150 l/kg. The experiments focused on analysing flowability, green
strength, mechanical properties (compressive strength, flexural strength, and modulus of elasticity), water absorption, porosity,
and chemical transformations at high temperatures through thermoanalytical measurement. The results showed that the mixture
containing 0.145 l/kg (w/d) exhibited satisfactory 3D printability, optimal mechanical performance, lower porosities compared with
mixtures 0.140 and 0.150, and same total porosity compared with cast specimen.
Keywords
3D concrete printing, compressive strength, flexural strength, loading directions, modulus of elasticity, water content

1 Introduction
Additive manufacturing (AM), commonly referred to as methods have been developed to utilize AM in concrete
3D printing, encompasses a group of innovative meth- construction industry. Extrusion-based and selective bind-
ods designed for fabricating three-dimensional objects ing (also known as powder-based) techniques are the main-
directly from digital models by depositing successive lay- stays of these technologies [4, 5]. This article delves into
ers of material, thereby minimizing waste production [1]. extrusion-based 3D concrete printing (3DCP). Many com-
The American Society for Testing and Materials (ASTM) panies, like Contour Crafting, CyBe, XtreeE, WinSun,
defines AM as "the process by which materials are joined Apis Cor, and more, have been exploring ways to bring this
to construct objects from 3D model data, typically through innovative technology to practical use in the real world [6],
a layer-by-layer approach" [2]. 3D concrete printing thereby demonstrating the potential of this technology.
(3D CP), represents an advanced form of AM. It facili- One of the key advantages of 3D CP is its potential to
tates the creation of complex geometrical structures, which reduce construction time and labour costs while minimiz-
were challenging to achieve using conventional construc- ing material waste [7, 8]. Traditional construction methods
tion techniques. This innovative approach permits the often involve significant manual labour and are susceptible
extrusion of concrete through a nozzle in a sequential lay- to errors, leading to delays and cost overruns [9]. In the
er-by-layer manner, enabling the construction of struc- report of Safe Work Australia [10], 35 construction workers
tural elements without the need for formwork or additional are seriously injured every day in Australia. Furthermore,
vibration processes [3]. In recent years, a variety of 3D CP falls from a height account for more than a quarter of all

Cite this article as: Thajeel, M. M., Balázs, G. L. "3D Printed Concrete: Fresh and Hardened Properties", Periodica Polytechnica Civil Engineering, 2024.
https://doi.org/10.3311/PPci.37602
2|Thajeel and Balázs
Period. Polytech. Civ. Eng.

construction deaths. This is despite the fact that Australia ratio [23, 26]. Additionally, admixtures such as superplasti-
has some of the strictest construction site safety rules in cizers, viscosity modifiers, and accelerators can be used to
the world. 3D CP, on the other hand, offers the potential for fine-tune the rheological performance [23].
rapid construction of building components, thereby stream- On the other hand, the other challenge of the 3D CP,
lining the overall construction process [11]. Additionally, lies in the orthotropic nature of the resulting 3D printed
this technology enables the customization of building object due to the extrusion process. Orthotropy denotes
components, thereby facilitating the implementation of the anisotropic behaviour of a material along different
novel architectural designs and structural solutions [12]. axes resulting from its inherent microstructural charac-
Furthermore, the use of 3D concrete printing has the poten- teristics. In the context of 3D printed concrete, the lay-
tial to enhance sustainability in construction by optimizing er-by-layer deposition process introduces variations in
material usage and reducing carbon emissions associated material composition and microstructure, thereby impact-
with traditional construction methods [13]. ing its mechanical properties. The study of the orthotro-
While 3D concrete printing (3DCP) holds great prom- pic behaviour of 3D printed concrete enables engineers
ise, it also faces several hurdles, particularly concerning and researchers to accurately evaluate its strength, stiff-
material properties, structural integrity, and the standard- ness, and durability under diverse loading conditions [27].
ization of construction practices. One of the biggest chal- Studies conducted on 3D printed concrete involved vari-
lenges in creating 3D printable cement-based materials ous tests such as tensile, compression, splitting, and flex-
is engineering them to have the right fresh properties for ural tests [28]. These tests confirmed the evident aniso-
extrusion-based printing. Unlike traditional cast concrete, tropic behaviour of 3D printed concrete in comparison to
which uses formwork to maintain shape during the curing ordinary concrete [29]. Additionally, it was observed that
process, 3D printing lacks this support. Therefore, the fresh the mechanical properties of 3D printed concrete are influ-
concrete mixtures need to have low slump and high thixot- enced by the loading direction, printing direction, and
ropy to ensure they are stable and suitable for the 3D print- printing parameters [25, 30].
ing process [14, 15]. Various studies [4, 8, 16] have investi- The present study aims to investigate the influence of
gated the fresh state behaviours of cementitious materials water content and loading directions on 3D printed con-
suitable for 3D printing, such as extrudability, open time, crete. A series of tests have been conducted to explore
and buildability. For instance, Le et al. [8] proposed tests to the effects of water content on the fresh state behaviour
develop a high-performance fibre-reinforced material for of 3D CP, encompassing flowability and green strength.
3D concrete printing, while Kazemian et al. [17] demon- Furthermore, the study delves into the impacts on the
strated ways to assess the printing quality and shape sta- mechanical characteristics, including compressive
bility of fresh mixtures. Ma et al. [14] used existing tests strength, flexural strength, and modulus of elasticity test-
to find the best mix design for concrete with copper tail- ing at 28 days. Additionally, mechanical tests have been
ings for 3D printing. Moreover, recent advancements have extended to encompass measurements of water absorp-
led to new methods for measuring and evaluating the fresh tion, apparent porosity, total porosity, and closed porosity.
properties of 3D printable cementitious materials [18–22]. Moreover, the chemical transformations at high tempera-
Expecting 3D printable cementitious materials to have tures have been studied through thermoanalythical mea-
both good extrudability and buildability might seem contra- surement. Finally, the paper discusses the obtained results.
dictory [23]. However, it's crucial to find a balance between
the right flowability for extrusion and sufficient strength 2 Experimental programme and methods
after deposition [24]. The performance of fresh mixtures 2.1 Raw material and fresh mixture preparation
under different conditions depends on their rheology and To prepare the 3D printed and cast specimens, a pre-mix
thixotropy. Roussel [25] provided a summary of the nec- material called Sikacrete®-751 3D from Sika was used.
essary rheological properties of cementitious materials for This material consists of Portland cement, aggregate, and
3D construction printing. Typically, the rheology of con- additives. Sikacrete®-751 3D is light grey-white in colour
crete is influenced by the mix design, which involves the and has an accelerated setting time, making it ideal for bet-
volume and composition of materials like cement, fillers, ter buildability. It features a maximum aggregate size of
and other supplementary components, as well as particle 1 mm and only requires mixing with water to be ready for
shape, content, aggregate gradation, and the water-to-binder use. Fig. 1 gives X-ray diffraction (XRD) patterns of raw
 Thajeel and Balázs
Period. Polytech. Civ. Eng. |3

Fig. 1 XRD analysis results of raw pre-mix material

pre-mix material for the passing and retained on sieve size

0.063 mm. XRD was conducted on PANanalytical Xpert
Pro MPD powder X-ray diffractometer with X'celerator
detector, with Cu-Kα radiation operated at 40 kV and
30 mA, for 2θ values between 3° and 85°.
The water to dry material ratio (w/d) was the main param-
eter of the different mix designs. Three mixes were prepared
by using three different w/d ratios, 0.140 l/kg, 0.145 l/kg and
0.150 l/kg. The fresh mixtures were prepared using a plane-
tary mixer (Hobart) by following the procedures:
1. homogenize the dry components by mixing for
2 min at a low speed,
2. add the water while mixing at a low speed,
3. after 3 min, pause, scrape the walls and blade, (a)
4. mix the mixture again at high speed for 2 min,
5. stop, start the test.

The time zero was defined as the moment when the

water was added to the dry mixture.

2.2 Concrete printing process

A mixing pump from MAI (Fig. 2 (b)), was used in this
paper. Only the pumping system of this mix was used since
the fresh mixtures were prepared by using outside Hobart
mixture to make sure adding the exact w/d ratio, addition-
ally to keep the same mixing procedure. After preparing
the fresh mixture manually loaded to the feeder system
which consist of a hopper (volume: 24 l), and extruder with
a barrel and die. A six-axis robot (Fig. 2 (a)) was used for
printing process. The robot used for printing can rotate (b)

360 degrees and cover a horizontal diameter of 3 meters Fig. 2 3DCP setup; (a) six-axis robot; (b) mixing pump system

and a vertical height of 3 m. With extensions, it can reach

up to 6 m. Its printing speed is adjustable, ranging from the hose and mounted on the robotic print head on the end
0 to 200 mm/s. The robotic arm is connected to a con- part of the z-axis. The flow rate, printing speed, and layer
troller that manages its head movements, ensuring pre- height were kept the same for the three mixes as 1.5 l/min,
cise operation. A flexible hose with an inner diameter of 84 mm/s, and 10 mm, respectively.
25 mm and a length of 5 m is used to connect the die to the In this study, we carefully considered the characteristics
pump. A nozzle with a circular opening is connected to of anisotropy. Therefore, it is necessary to establish a coor-
4|Thajeel and Balázs
Period. Polytech. Civ. Eng.

dinate system based on the printing process. We defined fresh mortar. To minimize air bubbles, all samples were
the horizontal print plane as the X-Y plane, with the verti- manually compacted. They were then stored in a sealed
cal direction across layers as the Z-axis. The X-axis aligned plastic bag and demoulded just before testing. The green
with the direction of the printed layer, following the move- strength tests were performed using an Instron universal
ment of the nozzle, while the Y-axis was perpendicular to testing machine equipped with a 10 kN load cell, Fig. 4.
the X direction within the printing plane, where the nozzle The test was conducted with constant compression and
moved between layers (Fig. 3). variable load, as follows:
• Pre-test: speed of 5 mm/min until reaching 4 N.
2.3 Test methods • Test: speed of 15 mm/min until either a compression
2.3.1 Flow table test of 25 mm or a reduction of the compressive force to
The test was carried out in accordance with EN 1015- 95%, the chosen test limitations were selected based
3:1999 [31]. In this test, a mini-cone mould with an upper on practical considerations to ensure consistent and
internal diameter of 70 mm, a bottom internal diameter of reliable testing of the green strength. Since there is
100 mm, and a height of 60 mm was used. Before start- no established standard for testing 3D printed con-
ing, the inside of the mould was lubricated with mould oil. crete, these parameters were determined through pre-
The mould was half-filled with the fresh mixture, which liminary trials to provide stable and repeatable condi-
was then stirred about 10 times with a wooden stick. This tions. The pre-test speed allows for a controlled initial
process was repeated for the second half of the mould. application of load, while the test speed is designed to
Any excess material was gently scraped off with a trowel. simulate realistic loading conditions and capture the
After removing the mould, the table was dropped 15 times material's response effectively. These settings ensure
within 15 seconds. The spread diameter was measured in that the material is neither subjected to excessive
two perpendicular directions. This test was performed for deformation too quickly nor allowed to settle for too
each mixture at 5 minutes, 15 minutes, and 30 minutes, long, which could affect the accuracy of the results.
with two repetitions for each time interval.

2.3.2 Green strength test

The green strength test was carried out to evaluate how
the strength of the mixtures developed over the first hour.
Fresh mixtures were prepared and poured into cylindrical
moulds with a 60 mm internal diameter and 60 mm height.
While there is no standard for the green strength test, the
mould dimensions were chosen based on the available
equipment in our laboratory. The moulds were initially
coated with silicone spray before filling them with the

Fig. 4 Green strength test setup

Fig. 3 Coordinate system of the 3D printed specimens (mixture 0.140)

 Thajeel and Balázs
|5
Period. Polytech. Civ. Eng.

For each mixture, the test was conducted at 5 min,

15 min, 30 min, 45 min and 60 min, with three repetitions
for each time interval.

2.3.3 Compressive strength test

A 580 × 286 × 110 mm3 wall was printed using the printing
system described in Section 2.2, (Fig. 5 (a)). The printed
samples were cut from the printed wall. To account for
the anisotropic mechanical properties of the printed mor-
tar [32], the samples were tested in the X, Y, and Z direc-
tions (Fig. 5 (b)), with three repeated tests conducted in
each direction. The average value was then calculated
from these repeated tests. Thus, for each mixture, nine
printed samples were prepared and tested in total. Six cast (a)
samples of each mixture were also prepared and tested
for comparison purposes. All samples had a dimension of
100 × 100 × 100 mm3 and were cured in a room tempera-
ture. Compressive strength at 28 days was measured con-
forming EN 12390-3:2019 [33], at a loading rate of 5 kN/s.

2.3.4 Flexural strength test (b)

A 580 × 320 × 90 mm3 wall was printed (Fig. 6 (a)). Small Fig. 5 Schematic diagram illustrating the loading directions for
prisms were cut in two perpendicular directions for the conducting compressive strength tests on printed cubes; (a) 3D printed
flexural strength test. The printed samples were tested in wall for compressive strength test; (b) Loading directions for
X, Y, and Z directions (Fig. 6 (b)). The average value was 3D printed cubes

calculated through 3 repeated tests in each direction. Thus,

for each mixture, nine printed samples were prepared
and tested in total. Six cast samples of each mixture were
also prepared and tested. All samples had a dimension of
40 × 40 × 160 mm3 and were cured in a room tempera-
ture. flexural strength at 28 days was measured conform-
ing EN 12390-5:2019 [34], at a loading rate of 0.05 kN/s.

2.3.5 Modulus of elasticity test

A 70 × 70 × 250 mm3 prisms were cut from the same wall
that printed in Section 2.3.4 (Fig. 6 (a)). The printed sam-
ples were tested in two directions were the layers horizon-
tal (H) and vertical (V), (Fig. 7 (a)). The average value
was calculated through three repeated tests in each direc-
(a)
tion. Thus, for each mixture, six printed samples were pre-
pared and tested in total. Three cast samples of each mix-
ture were also prepared and tested. The test was performed
by EN 12390-13:2013 standard [35]. The surface of the
specimen is marked with paint for appropriate contrast,
(b)
as the strain is measured by a contactless (video) technique
Fig. 6 Schematic diagram illustrating the loading directions for
(Fig. 7 (b)). The specimen is subjected to a compressive
conducting flexural strength tests on printed prisms; (a) 3D printed wall
force with maximum third of the compressive strength.
for flexural strength test; (b) Loading directions for 3D printed prisms
The data for stress and strain values were collected, secant
6|Thajeel and Balázs
Period. Polytech. Civ. Eng.

with size 50 × 50 × 50 mm3 for each mix tested four printed

cubes and four cast cubes. For water absorption calculation
the specimens were stored in water until constant mass.
After saturation the cubes were dried in a drying furnace at
60 °C until constant mass. Therefore, the amount of water
uptake could be measured and the volume of open pores
could be determined [36]. For the total porosity calcula-
tion, the material density was measured by using pycnom-
eter. The dried cubes crushed to powder and measured the
material density four times for each mix (printed and cast).
Water absorption refers to the ability of a material to
absorb water when immersed in it. It is an important prop-
erty to assess as excessive water absorption can lead to
degradation of the material over time. Apparent porosity
is a measure of the volume of open pores in a material rel-
ative to its total volume. It is an indicator of the material's
ability to absorb water and other liquids. Total porosity,
(a) on the other hand, includes both open and closed pores in
the material and provides a more comprehensive under-
standing of its porosity. Closed porosity specifically refers
to the volume of pores that are not interconnected and
are therefore not accessible to external fluids. The water
absorption, open porosity total porosity, and close porosity
were calculated by using Eqs. (2) to (5):

mwet  mdry
Water absorption: w  m%   100, (2)
mdry

Apparent porosity:
(3)
pa  w  m%    dry density water density  ,

Total porosity  1   dry density material density  , (4)

(b)
Fig. 7 Loading directions and test setup for conducting modulus of (5)
Close porosity  total porosity  apparent porosity,
elasticity test on printed prisms; (a) Loading direction of 3D printed
prisms; (b) Test setup
where mwet is the saturated mass and mdry is the dried mass.
modulus of elasticity was calculated using Eq. (1) from the To support the porosity results scanning electron micro-
standard: scope investigations were conducted (SEM), for cast (C)
and printed specimens. The 50 mm cubes were crushed,
  am   bm and the tested samples have been taken from the core of the
Ec ,0   , (1)
  a ,1   b ,0 cross-sectional areas of the specimen, then the samples faces
were coated with golden spray for 30 s in a vacuum cham-
where Ec,0 is modulus of elasticity, σ am is upper stress (MPa), ber and thereafter, the samples are investigated under SEM.
σ bm is lower stress (MPa), εa,1 is strain from upper stress
(mm/mm), and εb,0 is strain from lower stress (mm/mm). 2.3.7 Thermoanalythical test
The chemical changes in the selected mixtures as they
2.3.6 Water absorption and porosity tests were exposed to increasing temperatures were examined
Mechanical tests were extended by measurements of water using a MOM Q-1500 D derivatograph. In this test, the
absorption and porosity, which were measured on cube powdered material was heated up to 1000 °C, while the
 Thajeel and Balázs
|7
Period. Polytech. Civ. Eng.

device recorded thermogravimetric (TG), derivative ther- Moreover, during printing the three mixtures by using
mogravimetric (DTG), and differential thermoanalytical the printing system described in Section 2.2, the extru-
(DTA) data [37]. Thermoanalythical tests were made at sion pressure was recorded. It is obvious that the mixture
the age of 28 days and the test parameters were: containing a lower w/d ratio requires higher extrusion
• Reference substance: aluminum oxide; pressure to reach a similar material flow rate compared to
• Heating rate: 10 °C/min; mixture containing higher w/d ratios. The extrusion pres-
• Temperature range: 20–1000 °C; sure for the mixture 0.140 was ranging from 18–20 bar.
• TG sensitivity: 500 mg; While for the mixtures 0.145 and 0.150 were ranging from
• Crucible: corundum; 12–15 bar and from 6–10 bars respectively.
• Atmospheric pressure.
3.2 Green strength results
3 Results and discussion The green strength development within the first hour could
3.1 Flowability be used as an indicator of the proper structural buildability
Fig. 8 shows the results of the flow test results of mixtures of the mixture. Fig. 9 (a)–(c) shows the average load-dis-
with three w/d ratios at different ages. The spread diame- placement curves of the mixtures 0.140, 0.145, and 0.150
ter of all mixtures gradually decreases over time, indicat- from the age 5 min to 60 min. while Fig. 9 (d) shows the
ing that the workability of the fresh mixtures diminishes green strength of each test defined as the peak load divided
as resting time increases. This reduction in workability is by the surface area of the cylinder. The green strength of
likely due to the gradual stiffening of the mixture. This the mixtures 0.145 and 0.150 were close at younger ages
stiffening occurs because calcium silicate hydrate (C-S-H) 5 min – 15 min and lower than mixture 0.140. After that,
bridges form between particles, creating a rigid, intercon- the mixture 0.145 tends to be close to the mixture 0.140
nected network within the material [38]. Mixture 0.140 and much higher than the mixture 0.150. Indicating that this
showed lower flowability at any specific age compared ratio may be optimal for achieving good early strength with
with mixtures 0.145 and 0.150. Thus, may lead an issue to low pressure for pumpability requirement and great strength
pump and extrude concrete with low fluidity. While mix- development for buildability purpose in 3D printed concrete.
ture 0.150 demonstrated the highest fluidity after mixing
by about 7% and 17% as compared with mixtures 0.145 3.3 Compressive strength results
and 0.140, respectively. However, it’s noteworthy that Fig. 10 demonstrates the results referring to the compres-
the decrease in flowability with age is more pronounced sive strength of the casted and printed cubes of the three
for mixture 0.150. For instance, the flowability decreases mixtures after 28 days. The results showed that the com-
by 20% after 30 min as compared with other mixtures. pressive strength of the cast cubes (C), is greater than the
Mixture 0.145 seems to provide a good balance between compressive strength of the printed cubes. Similar results
flowability and stability, as it offers relatively good initial have also been reported in previous studies [39–41]. This
flowability (144.5 mm) and maintains reasonable flowabil- mechanical variation in printed samples may be due to the
ity after 30 minutes (120.8 mm). inadequate interface strength and printing precision [39].
In particular, the anisotropy in the compressive strength of
the printed cubes was observed, depending on the loading
direction. Importantly, the X direction is parallel to the
printing direction, which generally indicates higher par-
ticle placement and compaction compared to the Y and
Z directions. This is evident in the results for the mix-
tures with w/d ratios of 0.145 and 0.150, where the X direc-
tion shows superior performance in terms of compressive
strength. For the stiff mixture with a w/d ratio of 0.140,
however, the highest compressive strength was observed
in the Y direction. This deviation could be due to the
specific characteristics of the stiffer mixture, such as its
Fig. 8 Flowability test results of mixtures with different w/d ratios at reduced workability, leading to variations in layer bonding
different ages and compaction across different directions. Furthermore,
8|Thajeel and Balázs
Period. Polytech. Civ. Eng.

(a)

Fig. 10 Compressive strength test results at 28 days

Nerella et al. [27] commented that the compaction induced

by the layer weight should also be considered when assess-
ing the variation of the compressive strength.
(b)
In general, the compressive strength of concrete
tends to decrease as the water-to-dry material (w/d) ratio
increases, similar to conventional concrete where too
much water weakens the structure. However, for 3D print-
ing concrete, the compressive strength tends to increase
with a higher w/d ratio. The mixture with a 0.140 w/d ratio
had the lowest compressive strength for printed cubes but
the highest for cast ones. This mixture was too stiff, caus-
ing many cracks in the extruded layers, which negatively
affected its overall mechanical properties (see Fig. 11 (a)).
The lower water content altered the anisotropic behaviour
(c)
of the printed cubes, with the Y loading direction showing
about 5% and 3% higher compressive strength than the
X and Z directions, respectively.
The 0.145 mixture showed the best compressive strength
at 28 days in all three loading directions compared to the
0.150 and 0.140 mixtures (see Fig. 11 (b)). The average
compressive strength in the X direction was about 17%
higher than in the Y direction and 7% higher than in the
Z direction. Meanwhile, the 0.150 mixture had results
similar to the 0.145 mixture, with the average compressive
strength in the X direction being about 19% higher than
in the Y direction and 10% higher than in the Z direction.
(d)
3.4 Flexural strength results
Fig. 9 Load-displacement curve and green strength of fresh mixtures;
(a) Load-displacement curve of mixture 0.140; (b) Load-displacement Previous research on mechanical properties has demon-
curve of mixture 0.145; (c) Load-displacement curve of mixture 0.150; strated that 3D printing using cementitious materials
(d) Green strength of the three mixtures results in noticeable anisotropy behaviour once the mate-
rial hardens [42, 43]. Fig. 12 shows the flexural strength
 Thajeel and Balázs
|9
Period. Polytech. Civ. Eng.

to dry material ratio, 3D printed prisms loaded in Y and

Z directions showed the highest flexural strength 7.5 MPa
and 7.3 MPa, respectively. While the flexural strength of
the specimens when loaded in X direction were 5.1 MPa.
The flexural strength of the cast specimens was 8.8 MPa
and higher than that of printed specimens tested in X, Y, and
Z directions by about 42%, 15%, and 17%, respectively.
Same anisotropic behaviour was showed in the other mix-
(a) tures. Mixture 0.150 showed close results to mixture 0.145,
the flexural strength of the cast specimens and printed
specimens loaded in X, Y, and Z directions were 8.7 MPa,
4.9 MPa, 6 MPa, and 5 MPa, respectively. Meanwhile,
the mixture 0.140 showed the highest flexural strength
of the cast specimens 9.4 MPa and the lowest flexural
strength of printed specimens tested in X, Y, and Z direc-
tions 2.9 MPa, 4.3 MPa, and 4 MP, respectively. This can be
explained due to the extruded concrete in this mixture was
too stiff and the outer parts of the layers sheared off and the
bonding between the excessive layers were weak and that
(b) effect negatively on the overall mechanical characteristics.
Fig. 11 Quality of the printed walls; (a) w/d = 0.140; (b) w/d = 0.145 The primary cause of the pronounced anisotropy in 3D
printed specimens is the layered structure formed during
the printing process. Within this structure, voids and weak
contact interfaces typically develop between adjacent lay-
ers aligned with the printing direction. These imperfec-
tions significantly influence the bending performance of
the specimen when subjected to different loading direc-
tions. During bending, the specimen’s ultimate strength is
typically dictated by the maximum normal stress at its mid-
point. When loaded in the X direction, the tensile stress at
the midpoint runs perpendicular to the layer-to-layer inter-
face of the specimen, where the bond strength is weakest.

3.5 Modulus of elasticity results

The modulus of elasticity results of the cast specimens (C),
and printed specimens tested in two directions when
the layers vertical (V, parallel to the loading direction),
Fig. 12 Flexural strength test results at 28 days
and horizontal (H, perpendicular to the loading direction),
test results in this study. It's observed that using the were shown in Fig. 13.
0.145 w/d ratio tends to enhance the flexural strength of The results showed a higher modulus of elasticity for
the printed specimens. However, they still exhibit signif- cast specimen compared with printed specimens. With
icant anisotropy, which differs notably from traditionally increasing w/d ratio from 0.140 l/kg to 0.150 l/kg, a slight
cast specimens. This characteristic can be seen as inherent reduction of the modulus of elasticity for cast specimens
property to 3D printed layered structures. was observed from 34 GPa to 30 GPa. Meanwhile, for 3D
When subjected to loading in the X, Y, and Z directions, printed specimens, the modulus of elasticity was found to
the flexural strength is highest in the Y and Z directions, be significantly influenced by the water content and load-
showing similar performance, while the flexural strength in ing directions. All mixtures showed the same anisotropic
the X direction is relatively lower. With 0.145 l/kg water behaviour, a higher modulus of elasticity was found when
10|Thajeel and Balázs
Period. Polytech. Civ. Eng.

Fig. 14 Body and material densities of cast (C) and printed (P) specimens

process's pressure. Additionally, the selection of samples

from the central portion of the printed wall, which is sub-
ject to greater compaction relative to outer layers, further
Fig. 13 Modulus of elasticity test results at 28 days
contributes to this observed density increment.
the loading directions parallel to the printing directions (V) Fig. 15 delineates the water absorption rates, along with
compared to when the loading directions perpendicular to apparent, total, and closed porosity values for both cast and
the printing directions (H). During the 3D printing process, 3D printed specimens. It is evident that the water absorp-
layers are deposited sequentially, creating a structure with tion rates for 3D printed cubes were higher than those of
distinct directional properties. When the load is applied the cast cubes by approximately 33%, 14%, and 43% for
parallel to the printing direction, the stress is distributed mixtures 0.150, 0.145, and 0.140, respectively. Notably, the
along the continuous filaments of the printed layers, which mixture with w/d ratio 0.145 exhibited the lowest water
enhances the material's stiffness and results in a higher absorption rate among the tested specimens, indicating
modulus of elasticity. In contrast, when the load is applied a reduction of approximately 35% and 37% for 3D printed
perpendicular to the printing direction, the stress must tra- cubes, and 5% and 10% for cast cubes, in comparison with
verse the interfaces between layers, which are typically mixtures 0.150 and 0.140, respectively.
weaker due to the presence of micro-gaps, lower bonding Furthermore, the study found that the apparent poros-
strength, and potential defects. This interlayer weakness ity (open porosity) in 3D printed cubes was significantly
leads to lower stiffness and, consequently, a lower modulus higher than in cast cubes, with variations of about 32%,
of elasticity. This behaviour is a common characteristic of 15%, and 42% for the respective mixtures. Conversely,
additively manufactured materials and reflects the direc- the total porosity for printed cubes exceeded that of cast
tional dependency of their mechanical properties. cubes by approximately 10% and 11% for mixtures 0.150
The 0.140 mixture showed the lowest modulus of elas- and 0.140, respectively, while the values for mixture 0.145
ticity 19 GPa and 12 GPa in V and H directions, respec-
tively. Mixtures 0.145 and 0.150 showed close results
28 GPa, 21 GPa, 26 GPa, and 20 GPa in V and H direc-
tions, respectively.

3.6 Microstructure results

Fig. 14 presents the variations in body and material densi-
ties between conventionally cast cubes and those fabricated
through 3D printing techniques. A comparative analysis
reveals that the average hardened density of 3D printed
cubes marginally exceeds that of their cast counterparts.
This increase in density for 3D printed cubes, corrobo- Fig. 15 Water absorption, apparent porosity, total porosity and closed
rated by findings in literature [41, 44], can be attributed porosity for cast (C) and printed (P) specimens

to the enhanced compaction achieved under the extrusion

 Thajeel and Balázs
|11
Period. Polytech. Civ. Eng.

remained comparable between both types. Interestingly, These results are supported by the scanning electron
the closed porosity observed in printed cubes was mark- microscope (SEM). As shown in Fig. 16, the stiff and wet
edly lower than in cast cubes for all mixtures, showcasing mixes 0.140 and 0.150, respectively showed more voids
reductions of about 75%, 25%, and 109%, respectively. than the mixture with w/d ratio 0.145.

(a) (b)

(c) (d)

(e) (f)

Fig. 16 SEM images for cast (C) and printed (P) specimens; (a) 0.140 C; (b) 0.140 P; (c) 0.145 C; (d) 0.145 P; (e) 0.150 C; (f) 0.150 P
12|Thajeel and Balázs
Period. Polytech. Civ. Eng.

These observations underscore the distinct porosity system within the 3D printed elements, particularly at the
characteristics of 3D printed concrete, predominantly open interlayer junctions, thereby exerting a substantial influ-
porosity, in contrast to the predominantly closed porosity ence on the mechanical properties of the material.
observed in cast concrete. This distinction likely stems
from the layer-by-layer extrusion process inherent to 3D 3.7 Thermoanalythical results
printing. Moreover, the process of printing with either stiff Thermoanalythical tests were made at the age of 28 days.
or wet concrete mixtures introduces a more extensive void The results of the thermoanalythical test (Fig. 17) show

(a) (b)

(c) (d)

(e) (f)
Fig. 17 TG/DTG/DTA curves for cast (C) and printed (P) specimens; (a) 0.150 C; (b) 0.150 P; (c) 0.145 C; (d) 0.145 P; (e) 0.140 C; (f) 0.140 P
 Thajeel and Balázs
|13
Period. Polytech. Civ. Eng.

that for all mixtures, three mean peaks appear on the pressure and green strength. Mixture 0.145 seems
DTG and DTA curves. Two series of calcium alumi- to provide a good balance between flowability and
nates, AFm (mono-substituted) and AFt (tri-substituted), stability, as it offers relatively good initial flowabil-
appear, with the first main peak occurring at a tempera- ity (144.5 mm) and maintains reasonable flowability
ture interval of about 50 °C to 200 °C; on the second peak after 30 minutes (120.8 mm), which provide longer
appears at the temperature around 500 °C, represents cal- open time. Good early strength with low pressure for
cium hydroxide Ca(OH)2 one hydration product of cement. pumpability requirement and great strength develop-
Meanwhile, the derivative thermogravimetry (DTG) peak ment for buildability purpose.
between 700 °C to 900 °C indicates the presence of cal- 2. Compressive strength
cium carbonate CaCO3. CaCO3 can be either the reaction Anisotropy of the printed specimens appears to
product of the Ca(OH)2 and airborne CO2, or present in the be critical concerning compressive strength tests.
original pre-mix. For mixtures 0.145, and 0.150, printed samples were
Fig. 18 shows the three mean peaks as a mass loss as well stronger in the loading direction X in comparison
as shows the loss of ignition (LOI) which represents the with directions Y and Z. The X direction is parallel
total TG mass deficit during heating up to 1000 °C (m/m%). to the printing direction; this indicates higher par-
Based on the thermoanalythical test results, we can ticle placement and compaction compared to the Y
conclude, that the higher total amount of AFm and Aft and Z directions. Mixture 0.145 showed the highest
phases were formed in the mixtures of 0.145 (both cast and compressive strength at 28 days by about 60.9 MPa
printed), which can help in the setting and achieve the sta- at X direction. Meanwhile, the 0.140 mixture showed
bility. Also, the LOI was the higher for specimens of 0.145, the lowest compressive strength for printed cubes by
which shows the optimum w/d ratio (Fig. 18). about 48.3 MPa in X direction. The mixture was too
stiff and lots of cracks happened to the extruded lay-
4 Conclusions ers and that effect negatively on over all the mechan-
An extensive experimental study was carried out on fresh ical characteristics.
and hardened properties of 3D printed concretes including 3. Flexural strength
flowability, green strength, compressive strength, flexural Increasing the w/d ratio tends to enhance the flex-
strength, modulus of elasticity, water absorption, poros- ural strength of the printed specimens. However,
ity, and chemical transformations at high temperatures they still exhibit significant anisotropy, which dif-
through derivatographic measurements. Test parameters fers notably from traditionally cast specimens. This
were the amount of mixing water and testing direction. characteristic can be seen as inherent property to
A pre-mix material from Sika was used in this study. Three 3D printed layered structures. The flexural strength
mixes were prepared by using three different water to dry is highest in the Y and Z directions, showing sim-
material (w/d) ratios, 0.140 l/kg, 0.145 l/kg and 0.150 l/kg. ilar performance, while the flexural strength in the
The following conclusions could be obtained in this study: X direction is relatively lower. With 0.145 l/kg water
1. Flowability to dry material ratio, the flexural strength of the cast
Increasing w/d ratio from 0.140 to 0.150 could specimens was higher than that of printed specimens
increase the flowability and reduce the extrusion tested in X, Y, and Z directions by about 42%, 15%,
and 17%, respectively.
4. Modulus of elasticity
A higher modulus of elasticity was observed in cast
specimens compared to printed specimens. With
increasing w/d ratio from 0.140 l/kg to 0.150 l/kg,
a slight reduction of the modulus of elasticity for
cast specimens was observed from 34 GPa to
30 GPa. Meanwhile, for 3D printed specimens, the
modulus of elasticity was found to be significantly
influenced by the water content and loading direc-
tions. All mixtures showed the same anisotropic
Fig. 18 Mass loss and LOI of the cast (C) and printed (P) specimens
14|Thajeel and Balázs
Period. Polytech. Civ. Eng.

behaviour, a higher modulus of elasticity was found 6. Thermoanalythical test results

when the loading directions parallel to the printing Based on the thermoanalythical test results, we can
directions (V) compared to when the loading direc- conclude, that the higher total amount of AFm and
tions perpendicular to the printing directions (H). Aft phases were formed in the mixtures of 0.145
The 0.140 mixture showed the lowest modulus of (both cast and printed), which can help in the set-
elasticity 19 GPa and 12 GPa in V and H directions, ting and achieve the stability. Also, the LOI was the
respectively. Mixtures 0.145 and 0.150 showed close higher for specimens of 0.145, which shows the opti-
results 28 GPa, 21 GPa, 26 GPa, and 20 GPa in V and mum w/d ratio.
H directions, respectively.
5. Water absorption and porosity Acknowledgements
Water absorption, apparent porosity (open porosity), The Stipendium Hungaricum Scholarship Program is
total porosity for printed specimens were higher than highly acknowledged for supporting the PhD study and
that for cast specimens. However, the closed porosity research work. Authors acknowledge the support by
was the lower, this indicates the porosity type in the the Hungarian Research Grant VKE 2018-1-3-1_0003
printed specimens is open porosity, while in the cast "Development of advanced concrete elements". Thanks to
specimens is closed one. Mixture 0.145 showed the Balázs Burai, Ameen H. Chalawi, as well as Dr. Sándor
lowest water absorption, apparent, total and closed Sólyom for their help in the printing process and the lab-
porosities compared with the mixtures 0.140 and oratory measurements. Special thanks to Dr. Katalin
0.150. Although, it has a higher apparent porosity and Kopecskó for her help in evaluating derivatographic mea-
lower closed porosity, however the total porosity was surements and reviewing the paper, as well as thanks to
the same as compared to the cast specimens. Prof. Éva Lublóy for her help in reviewing the paper.

References
[1] Nematollahi, B., Xia, M., Sanjayan, J. "Current Progress of 3D [7] Al Rashid, A., Khan, S. A., G. Al-Ghamdi, S., Koç, M. "Additive
Concrete Printing Technologies", In: 2017 Proceedings of the manufacturing: Technology, applications, markets, and opportuni-
34th International Symposium on Automation and Robotics in ties for the built environment", Automation in Construction, 118,
Construction (ISARC), Taipei, Taiwan, 2017, pp. 260–267. ISBN 103268, 2020.
978-80-263-1371-7 https://doi.org/10.1016/j.autcon.2020.103268
https://doi.org/10.22260/ISARC2017/0035 [8] Le, T. T., Austin, S. A., Lim, S., Buswell, R. A., Gibb, A. G. F., Thorpe,
[2] ASTM "ASTM F2792-10 Standard Terminology for Additive T. "Mix design and fresh properties for high-performance printing
Manufacturing Technologies", ASTM International, West concrete", Materials and Structures, 45(8), pp. 1221–1232, 2012.
Conshohocken, PA, USA, 2010. https://doi.org/10.1617/s11527-012-9828-z
https://doi.org/10.1520/F2792-10 [9] De Schutter, G., Lesage, K., Mechtcherine, V., Nerella, V. N.,
[3] Paolini, A., Kollmannsberger, S., Rank, E. "Additive manufactur- Habert, G., Agusti-Juan, I. "Vision of 3D printing with concrete
ing in construction: A review on processes, applications, and digi- – Technical, economic and environmental potentials", Cement and
tal planning methods", Additive Manufactiring, 30, 100894, 2019. Concrete Research, 112, pp. 25–36, 2018.
https://doi.org/10.1016/j.addma.2019.100894 https://doi.org/10.1016/j.cemconres.2018.06.001
[4] Nerella, V., N., Näther, M., Iqbal, A., Butler, M., Mechtcherine, V. [10] Australia, S. W. "Work-related injuries and fatalities in construc-
"Inline quantification of extrudability of cementitious materials tion", [online] Available at: https://www.abs.gov.au/ [Accessed:
for digital construction", Cement and Concrete Composites, 95, 20 October 2023]
pp. 260–270, 2019. [11] Arunothayan, A. R., Nematollahi, B., Ranade, R., Bong, S. H.,
https://doi.org/10.1016/j.cemconcomp.2018.09.015 Sanjayan, J. G., Khayat, K. H. "Fiber orientation effects on ultra-
[5] Chen, Y., Veer, F., Çopuroğlu, O. "A critical review of 3D concrete high performance concrete formed by 3D printing", Cement and
printing as a low CO2 concrete approach", Heron, 62(3), pp. 167– Concrete Research, 143, 106384, 2021.
194, 2017. https://doi.org/10.1016/j.cemconres.2021.106384
[6] Chaves Figueiredo, S., Romero Rodríguez, C., Ahmed, Z. Y., Bos, [12] Lloret, E., Shahab, A. R., Linus, M., Flatt, R. J., Gramazio, F.,
D. H., Xu, Y., Salet, T. M., Çopuroğlu, O., Schlangen, E., Bos, F. P. Kohler, M., Langenberg, S. "Complex concrete structures: Merging
"An approach to develop printable strain hardening cementitious existing casting techniques with digital fabrication", Computer-
composites", Materials & Design, 169, 107651, 2019. Aided Design, 60, pp. 40–49, 2015.
https://doi.org/10.1016/j.matdes.2019.107651 https://doi.org/10.1016/j.cad.2014.02.011
 Thajeel and Balázs
|15
Period. Polytech. Civ. Eng.

[13] Nematollahi, B., Sanjayan, J., Shaikh, F. U. A. "Synthesis of [26] Chen, Y., Veer, F., Copuroglu, O., Schlangen, E. "Feasibility of
heat and ambient cured one-part geopolymer mixes with differ- Using Low CO2 Concrete Alternatives in Extrusion-Based 3D
ent grades of sodium silicate", Ceramics International, 41(4), Concrete Printing", In: First RILEM International Conference on
pp. 5696–5704, 2015. Concrete and Digital Fabrication – Digital Concrete 2018, Zurich,
https://doi.org/10.1016/j.ceramint.2014.12.154 Switzerland, 2018, pp. 269–276. ISBN 978-3-319-99518-2
[14] Ma, G., Li, Z., Wang, L. "Printable properties of cementitious https://doi.org/10.1007/978-3-319-99519-9_25
material containing copper tailings for extrusion based 3D print- [27] Nerella, V. N., Hempel, S., Mechtcherine, V. "Effects of layer-in-
ing", Construction and Building Materials, 162, pp. 613–627, 2018. terface properties on mechanical performance of concrete ele-
https://doi.org/10.1016/j.conbuildmat.2017.12.051 ments produced by extrusion-based 3D-printing", Construction and
[15] Lim, S., Buswell, R. A., Le, T. T., Austin, S. A., Gibb, A. G. F., Building Materials, 205, pp. 586–601, 2019.
Thorpe, T. "Developments in construction-scale additive manufac- https://doi.org/10.1016/j.conbuildmat.2019.01.235
turing processes", Automation in Construction, 21, pp. 262–268, [28] Yu, K., McGee, W., Ng, T. Y., Zhu, H., Li, V. C. "3D-printable
2012. engineered cementitious composites (3DP-ECC): Fresh and hard-
https://doi.org/10.1016/j.autcon.2011.06.010 ened properties", Cement and Concrete Research, 143, 106388,
[16] Wangler, T., Roussel, N., Bos, F. P., Salet, T. A. M., Flatt, R. J. 2021.
"Digital Concrete: A Review", Cement and Concrete Research, https://doi.org/10.1016/j.cemconres.2021.106388
123, 105780, 2019. [29] Pasco, J., Z. Lei, Z., Aranas Jr., C. "Additive Manufacturing in
https://doi.org/10.1016/j.cemconres.2019.105780 Off-Site Construction: Review and Future Directions", Buildings,
[17] Kazemian, A., Yuan, X., Cochran, E., Khoshnevis, B. "Cementitious 12(1), 53, 2022.
materials for construction-scale 3D printing: Laboratory testing of https://doi.org/10.3390/buildings12010053
fresh printing mixture", Construction and Building Materials, 145, [30] Roussel, N., Spangenberg, J., Wallevik, J., Wolfs, R. "Numerical
pp. 639–647, 2017. simulations of concrete processing: From standard formative cast-
https://doi.org/10.1016/j.conbuildmat.2017.04.015 ing to additive manufacturing", Cement and Concrete Research,
[18] Wolfs, R. J. M., Bos, F. P., Salet, T. A. M. "Early age mechanical 135, 106075, 2020.
behaviour of 3D printed concrete: Numerical modelling and exper- https://doi.org/10.1016/j.cemconres.2020.106075
imental testing", Cement and Concrete Research, 106, pp. 103–116, [31] European Committee for Standardization "EN 1015-3:1999
2018. Methods of test for mortar for masonry Part 3: Determination of
https://doi.org/10.1016/j.cemconres.2018.02.001 consistence of fresh mortar (by flow table)", European Committee
[19] Reiter, L., Wangler, T., Roussel, N., Flatt, R. J. "The role of early age for Standardization, Brussels, Belgium, 2000.
structural build-up in digital fabrication with concrete", Cement and [32] Panda, B., Paul, S. C., Hui, L. J., Tay, Y. W. D., Tan, M. J. "Additive
Concrete Research, 112, pp. 86–95, 2018. manufacturing of geopolymer for sustainable built environment",
https://doi.org/10.1016/j.cemconres.2018.05.011 Journal of Cleaner Production, 167, pp. 281–288, 2017.
[20] Wolfs, R. J. M., Bos, F. P., Salet, T. A. M. "Triaxial compression https://doi.org/10.1016/j.jclepro.2017.08.165
testing on early age concrete for numerical analysis of 3D concrete [33] European Committee for Standardization "EN 12390-3:2019
printing", Cement and Concrete Composites, 104, 103344, 2019. Testing hardened concrete - Part 3: Compressive strength of test
https://doi.org/10.1016/j.cemconcomp.2019.103344 specimens", European Committee for Standardization, Brussels,
[21] Alghamdi, H., Nair, S. A. O., Neithalath, N. "Insights into mate- Belgium, 2019.
rial design, extrusion rheology, and properties of 3D-printable [34] European Committee for Standardization "EN 12390-5:2019
alkali-activated fly ash-based binders", Materials & Design, 167, Testing hardened concrete - Part 5: Flexural strength of test
107634, 2019. specimens", European Committee for Standardization, Brussels,
https://doi.org/10.1016/j.matdes.2019.107634 Belgium, 2019.
[22] Rahul, A. V., Santhanam, M., Meena, H., Ghani, Z. "3D printable [35] European Committee for Standardization "EN 12390-13:2013
concrete: Mixture design and test methods", Cement and Concrete Testing hardened concrete - Part 13: Determination of secant
Composites, 97, pp. 13–23, 2019. modulus of elasticity in compression", European Committee for
https://doi.org/10.1016/j.cemconcomp.2018.12.014 Standardization, Brussels, Belgium, 2013.
[23] Marchon, D., Kawashima, S., Bessaies-Bey, H., Mantellato, S., [36] European Committee for Standardization "EN 1936:2006 Natural
Ng, S. "Hydration and rheology control of concrete for digital fab- stone test methods - Determination of real density and apparent
rication: Potential admixtures and cement chemistry", Cement and density, and of total and open porosity", European Committee for
Concrete Research, 112, pp. 96–110, 2018. Standardization, Brussels, Belgium, 2007.
https://doi.org/10.1016/j.cemconres.2018.05.014 [37] Biró, A., Hlavička, V., Lublóy, É. "Effect of fire-related tempera-
[24] Perrot, A., Rangeard, D., Pierre, A. "Structural built-up of cement- tures on natural stones", Construction and Building Materials, 212,
based materials used for 3D-printing extrusion techniques", pp. 92–101, 2019.
Materials and Structures, 49(4), pp. 1213–1220, 2016. https://doi.org/10.1016/j.conbuildmat.2019.03.333
https://doi.org/10.1617/s11527-015-0571-0 [38] Roussel, N., Ovarlez, G., Garrault, S., Brumaud, C. "The origins of
[25] Roussel, N. "Rheological requirements for printable concretes", thixotropy of fresh cement pastes", Cement and Concrete Research,
Cement and Concrete Research, 112, pp. 76–85, 2018. 42(1), pp. 148–157, 2012.
https://doi.org/10.1016/j.cemconres.2018.04.005 https://doi.org/10.1016/j.cemconres.2011.09.004
16|Thajeel and Balázs
Period. Polytech. Civ. Eng.

[39] Panda, B., Singh, G. V. P. B., Unluer, C., Tan, M. J. "Synthesis and [42] Wolfs, R. J. M., Bos, F. P., Salet, T. A. M. "Hardened properties of
characterization of one-part geopolymers for extrusion based 3D 3D printed concrete: The influence of process parameters on inter-
concrete printing", Journal of Cleaner Production, 220, pp. 610– layer adhesion", Cement and Concrete Research, 119, pp. 132–140,
619, 2019. 2019.
https://doi.org/10.1016/j.jclepro.2019.02.185 https://doi.org/10.1016/j.cemconres.2019.02.017
[40] Chen, Y., Chaves Figueiredo, S., Li, Z., Chang, Z., Jansen, K., [43] Ma, G., Li, Z., Wang, L., Wang, F., Sanjayan, J. "Mechanical anisot-
Çopuroğlu, O., Schlangen, E. "Improving printability of lime- ropy of aligned fiber reinforced composite for extrusion-based 3D
stone-calcined clay-based cementitious materials by using viscos- printing", Construction and Building Materials, 202, pp. 770–783,
ity-modifying admixture", Cement and Concrete Research, 132, 2019.
106040, 2020. https://doi.org/10.1016/j.conbuildmat.2019.01.008
https://doi.org/10.1016/j.cemconres.2020.106040 [44] Pham, L., Tran, P., Sanjayan, J. "Steel fibres reinforced 3D printed
[41] Le, T. T., Austin, S. A., Lim, S., Buswell, R. A., Law, R., Gibb, A. concrete: Influence of fibre sizes on mechanical performance",
G. F., Thorpe, T. "Hardened properties of high-performance print- Construction and Building Materials, 250, 118785, 2020.
ing concrete", Cement and Concrete Research, 42(3), pp. 558–566, https://doi.org/10.1016/j.conbuildmat.2020.118785
2012.
https://doi.org/10.1016/j.cemconres.2011.12.003