< previous project

Design and Production of a 3D-Printed Concrete Beam for LCA Comparison

PROJECT TITLE

Design and production of a topology-optimized, additively manufactured concrete beam as a basis for a life cycle assessment comparison of structural variants

AUTHOR

Nina Sam

RESEARCH PROJECT

C3PO – Concrete 3D Printed Objects

Funded by MA23

PROJECT TEAM

Marc-Patrick Pfleger, Elisabeth Radl, Nina Sam, Markus Vill, Sebastian Geyer, Christian Hölzl

 

INSTITUTION

Department Bauen und Gestalten of Hochschule Campus Wien

 

Text by Nina Sam:

 

This Master’s thesis explores the intersection of advanced digital design methods and sustainable construction practices by focusing on the design and production of a topology-optimized concrete beam using additive manufacturing (3D concrete printing). The primary objectives were to investigate how topology optimization can reduce material consumption in bending elements while preserving structural performance, and to assess the environmental impacts of various optimized structural variants through a life cycle assessment (LCA).

Background and Motivation

The construction sector is a significant contributor to global carbon emissions, with cement production alone accounting for approximately 5 percent of total emissions. With the increasing demand for concrete and a growing shortage of skilled labor, there is a pressing need for more sustainable and automated methods of construction. Additive manufacturing offers potential solutions by enabling precise material placement and eliminating the need for formwork, which results in less material waste and shorter construction times.
The use of additive manufacturing together with digital tools like topology optimization algorithms makes it possible to create complex shapes that use material efficiently and follow the flow of internal forces. This work focuses on bending elements, where both tension and compression occur and smart design and reinforcement methods are needed.

Figure 1 - Overview of the Grasshopper file of the topology optimization with Karamba3D Software

Methodology and Conception of the Variants


The thesis combines theoretical, computational, and experimental approaches. A literature review established the state of the art in additive manufacturing, topology optimization (SIMP and BESO), and reinforcement strategies for 3D concrete printing, forming the basis for defining design constraints and boundary conditions for a beam under bending.

A conventional reference beam was analyzed using finite element analysis (FEA) as a baseline. Several optimized variants were then generated in Grasshopper with Karamba3D using different topology optimization algorithms, prioritizing material placement along internal stress paths. Six topologies were developed, and one was selected for fabrication based on structural feasibility and printability. The chosen variant was produced as a full-scale prototype using a 6-axis robotic arm with a concrete extrusion system and a custom fine-grain concrete mix optimized for extrusion performance.

Figure 2 - Various analysis parameters and representations of the topology-optimized structure. Left: FEM analysis of the stress level in the structure. Right: Material distribution in the topology-optimized structure: white = empty space; black = material
Table 1 - Results of the structural variants and assessment of convergence using graphs of mean compliance and volume

Figure 2 shows a general representation of the result structure in black and white on the right, while the left side shows the stress level of the structure. However, the left-hand representation is misleading, as the empty spaces are also colored (stress), even though no material exists here after topology optimization. For this reason, the black-and-white representation is used in the following. The feasibility of the three structural variants is assessed using the graphs shown in Table 1, which provide information about the convergence of the topology optimization. BESO_A_m was excluded because convergence was not achieved. BESO_A_u and BESO_A_o show no significant differences in terms of average compliance and volume. BESO_A_o has a lower average compliance and would therefore be the stiffer structure. Nevertheless, variant BESO_A_u was chosen for the prototype because the internal geometry consists of fewer struts and therefore appears less complex to manufacture. Variant BESO_A_u was subsequently subjected to a post-design process in order to adapt the geometry to the real conditions of the printing system.

Post-Design-Process

Before printing, the selected geometry underwent a post-design process to ensure compatibility with the fabrication system. This included geometric smoothing, reinforcement placement planning, and segmentation adjustments to avoid overhangs and ensure print stability.

Figure 3 shows the theoretical strand width and the placement of the strands in order to maintain the organic shape of the topology-optimized beam. The topology-optimised beam shape exhibits characteristics of several known static structural forms. The generated structure is reminiscent of a truss, characterised by organic strut lines. During the topology optimisation process, the solid cross-section was reduced to compression and tension members, which are connected by diagonal struts. The diagonal struts at the center of the beam also resemble the design of an arched structure with tension members.

Table 2 – Schematic representation of the steps in the Post-Design-Process
Figure 3 – Final prototype model and theoretical strand width

Additive Manufacturing and Experimental Testing

The printing process was carried out using a 6-axis robotic arm mounted on a linear track, with the concrete extruded layer by layer from a custom-designed nozzle to ensure precision in material placement and adherence between layers. In order to facilitate the processing of the generated data by the robot, it is necessary to translate the G-code into the robot’s executable programming language, known as RAPID code. This step is performed using RobotStudio software and the 3D Printing PowerPac from manufacturer ABB. Within the software, it is possible to position the object at any point on the printing platform, define the printing speed, and subsequently perform a printing simulation in order to identify any potential errors in the printing process.

Figure 4 – Screenshot of the printing simulation from Robot Studio

Additive Manufacturing and Experimental Testing

The printing process was carried out using a 6-axis robotic arm mounted on a linear track, with the concrete extruded layer by layer from a custom-designed nozzle to ensure precision in material placement and adherence between layers. In order to facilitate the processing of the generated data by the robot, it is necessary to translate the G-code into the robot’s executable programming language, known as RAPID code. This step is performed using RobotStudio software and the 3D Printing PowerPac from manufacturer ABB. Within the software, it is possible to position the object at any point on the printing platform, define the printing speed, and subsequently perform a printing simulation in order to identify any potential errors in the printing process.

Figure 5 - Printing process for the topology-optimized beam.
Figure 6 - Printing process for the topology-optimized beam. First reinforcement layer inserted (left) and subsequent printing of the next layers over the reinforcement (right)

After printing and curing, the prototype beam was subjected to a three-point bending test to assess its mechanical performance. The test confirmed the structural viability of the component, with results indicating that the optimized geometry maintained adequate stiffness and load-bearing capacity. Despite the open, lattice-like structure and reduced material volume, the beam successfully resisted bending moments, but then local node failure occurred as shown in Figure 7, which prevented further force absorption. The beam mainly failed at the nodes near the supports, where the structure was very stiff and couldn’t deform much. This caused cracks and splitting in the concrete, especially because the steel reinforcement was blocked from stretching. Possible improvements include removing the compression strut at the support to allow more flexibility (Figure 8) or using a hybrid girder design, in which a more flexible lower chord made of steel cables is combined with an additively manufactured concrete upper chord shown in Figure 9.

Figure 7 - Fracture pattern after component failure / Buckling of the upper chord in the left support area with further increase in load / Final failure pattern of the topology-optimized beam
Figure 8 - Optimized structure in the support area
Figure 9 - Optimized beam structure with a lower chord made of steel

Life Cycle Assessment

The LCA followed ISO 14040 and 14044 standards and focused on the production phase (A1 to A3) in accordance with EN 15643. Environmental data were gathered in form of environmental product declarations (EPDs) from databases such as baubook, ecoinvent and Oekobaudat. The functional unit was defined as one beam with identical load-bearing performance across all variants.

The environmental impacts of each structural variant were assessed based on Global Warming Potential (GWP in kg CO2-equivalents). As shown in Figure 12, the production of raw materials (A1 material) has the greatest impact. Transport and manufacturing account for only about 5% of the total emissions. Additive manufacturing is therefore a highly efficient production method in terms of emissions.

Results revealed that the most geometrically complex variant (BESO_A_o) showed the lowest GWP due to its high material efficiency. This confirmed that complex geometries do not necessarily result in higher environmental burdens when produced additively. In contrast, a seemingly efficient but overly filigree variant with too many small elements (BESO_A_m) resulted in higher material use due to rework and geometric inefficiencies, leading to increased emissions. The selected prototype exhibited a balance between material reduction, structural performance, and feasibility for additive manufacturing.

Schematic representation of the cradle-to-gate phases A1-A3 for the additive manufacturing of the prototype
GWP values for manufacturing phases A1-A3 of the structural variants and of the materials

Key Findings

• Topology optimization in combination with 3D concrete printing enables significant material savings
• Additive manufacturing facilitates the realization of highly complex geometries that are otherwise unfeasible with traditional formwork-based construction.
• LCA results demonstrated that material savings directly translate into environmental benefits, particularly in terms of GWP, provided that optimized geometries are feasible to manufacture.
• Even though 3D printable concrete often require higher binder contents (leading to initially higher CO2 footprints per volume), the net impact can be reduced through force-flow geometry design.

see more

Submit Project