1 Introduction

The use of cementitious materials within construction can be traced back to ancient Egyptian times, with the use of gypsum as a cementing material. Ancient Greeks and Romans used calcinated limestone mixed with water and crushed bricks and stones to form buildings and infrastructure which are still standing today. These structures show the strength and durability of concrete as a constructing material. At the turn of the twentieth century, modern concrete was introduced (Bussel, 2015). This revolutionized the construction industry, as reinforced concrete replaced the unreliability of wood and natural stone. With its low production costs, increase in strength, and the worldwide availability of raw materials, concrete has now become the most used building material in the world (Anton et al., 2019, 2021; Chen et al., 2017; Wangler et al., 2016).

Because of these qualities, cement has become the largest manufactured product on earth by mass, and the second most used substance in the world after water (Scrivener et al., 2018a, b). In 2021, it was estimated that globally, 32 billion tonnes of concrete were being produced per year, with this number expected to rise (ISO/TC 71, 2021). An increasing scarcity of base aggregates such as sand, and the growing awareness of concrete’s environmental impact across the construction, operation, and demolition of buildings, is leading to a rapid change in awareness of the effect concrete has on our planet. Coupled with new regulations, it means the Architecture, Engineering, and Construction Industry must adapt (Dixit, 2017; Jipa et al., 2019).

With increasing global urbanization, the concrete sector is expected to continue to grow (Habert et al., 2020). Concrete is an irreplaceable material of modern construction (Imbabi et al., 2012), and whilst there may be disruptive technologies and material compositions on the far horizon to revolutionize the industry, there is an urgent need to implement new methods for using concrete in a more sustainable way. 3D concrete printing (3DCP) has emerged as the most popular solution to solve some of these issues (Anton et al., 2021).

1.1 What and why 3DCP

3DCP is an Additive Manufacturing (AM) method of Digital fabrication. Digital fabrication, at an architectural scale, is based on computational design methods and robotic construction processes (Agustí-Juan et al., 2017). AM is a fabrication process, transferring 3D model data and creating a 3D object by the joining of materials, usually layer on layer, to create a physical 3D form. This is as opposed to subtractive manufacturing methodologies which remove material from a stock to create a form (Chen et al., 2017).

3DCP includes two different printing techniques: extrusion and powder-bed (Chen et al., 2017). In the extrusion method, concrete is extruded out of a nozzle positioned via a gantry or robotic system. The extrusion occurs along a predetermined path that is defined from a 3d digital file and is built-up layer upon layer to form a final 3-dimensional element. This element can by printed in-situ or as a prefabricated component. The ability of extrusion-based methods to be scaled up, and the flexibility of printing material mixes that can be used, has meant that this method has become foregrounded and developed most within industry.

There are many suggested benefits to 3DCP. These include: The reduction of the need for formwork, the selected deposition of material only where it is needed, and the ability to produce complex, optimized geometries at no perceived additional cost (Gebhard et al., 2020; Wangler et al., 2019). Additional to these, there is also increased productivity and benefits to worker safety (** prefabrication options (BESIX, n.d.; XtreeE, n.d.-a), the general trend within industry is towards large scale in-situ.

Many of these enterprises offer their own platforms to aid in the design and production of elements, such as COBOD’s Slicr (COBOD, n.d.-b). These are often based off other 3D printing software such as Prusa (Prusa, n.d.), with adapted parameters specifically for 3DCP. However, these are often patented, and access is only available with use of a printer, or through subscription (CyBe, n.d.-b; RAPCAM AM, n.d.). More basic slicing software are available open source, such as SlicerXL (slicerXL, n.d.), but with limited control of the printing parameters. Anton et al (2019) in 2019 presented an open source design platform for the generation of bespoke 3DCP columns, and there is scope for more open-source software to become available.

1.3 A new challenge: sustainability in 3DCP

There is a global climate crisis, and the AEC industry must adapt. Within Denmark, a threshold limit for emissions from new constructions has been set at 12 kg CO2eg/ m2/year, calculated over a 50 year ‘standard’ building lifespan. This is set to reduce to 7.5 kg CO2eg/ m2/year by 2029, with a voluntary standard of 5 kg CO2eg/ m2/year (National Strategy for Sustainable Construktion, 2021). As these calculations are performed through Life Cycle Assessments (LCA), these tools are fast becoming an integral part of a buildings design process, assessing the impacts across the life cycle phases of a building: from raw materials, building use, through to end-of-life.

Whether 3DCP is a sustainable practice is a subject of debate (Flatt & Wangler, 2022). Many of those within the industry argue that the reduction in the need for formwork, and material optimization shows that it is a sustainable practice (Kuzmenko et al., 2020, p. 3; Mohammad et al., 2020). In 2004, Khoshnevis et al. (2004) concluded that the material efficiency, speed, and digital fabrication process of 3DCP “favorably impacts the transportation system and environment”, and that the environmental impact is noteworthy.

1.4 Current areas of research

As Fig. 2 Illustrates, since 2004, 400 papers have been published within the field of 3DCP. Ma et al. (2022) performed a comprehensive review of all published papers, and found that 80% were focused on materials, 10% on Process, 1% on Building integration and 9% on Software.

Fig. 2
figure 2

Focus areas of research. Diagram based on Ma et al. (2022)

Full size image

However, despite a frequent acknowledgement that further research into the sustainability of 3DCP is required (Wangler et al., 2019), a systematic literature review conducted in 2022 found only 15 papers that specifically look at 3DCP from an environmental sustainability perspective (Tinoco et al., 2022). Comparing this number to the 430 papers on which Ma et al. performed their analysis, it implies that only 3% of all papers focus on the environmental impacts. This is a gap that, given the current climate concerns, urgently needs to be addressed.

In the author’s own analysis of more recent literature, 156 academic papers published from 2021 were analyzed to understand the more recent trends in research focus’ in 3DCP. These papers were sourced through academic online journals and articles, as well as conference publications. The key terms used in sourcing relevant papers were ‘3D Concrete Printing’, ‘3DCP’ ‘Additive Manufacturing Concrete’. From this, a filtering process was performed by the author, based on keywords as well as what was deemed as the architectural relevance of each paper. Out of the resulting 156 papers, 12 papers were found to be focused on the sustainability impacts of 3DCP. Whilst this shows a growing recognition of the importance of the environmental impacts, it is still a nascent territory of research.

2 A framework for analyzing 3DCP research

The world of 3DCP is notably collaborative, bringing together multiple interests and disciplines, from academia and Industry (Perrot et al., 2021). To understand the field of 3DCP, and the broad interests and backgrounds of the different researchers, a framework is presented that divides the research areas of 3DCP into three main fields: Material Science, Computational Design, and Structure and Performance (Fig. 3).

Fig. 3
figure 3

Framework for understanding the different research areas and topics of 3DCP

Full size image

It should be noted that this framework has been built knowing that many projects and papers could be classified within one or more fields, and this also reflects the interdisciplinary nature of 3DCP research. Developments in 3DCP are not a linear process, and each development in one field will influence the others, pushing all fields and 3DCP technology further.

2.1 Definitions of the framework

Material Science incorporates the research areas of chemical engineering, mechanical testing and the investigations into the different compounds and recipe mixes for 3D printable concrete. In its infancy, when the technology was untested, understanding the printing material was key. More recently, as the printing processes and structural limitations were overcome, a second phase of material science research in 3DCP has begun to investigate more sustainable print mixes, which can meet the increasing goals for structure and performance (Ma et al., 2022).

3DCP digitizes the construction industry and bridges the gap between the advanced modelling tools used by designers, and the physical construction process (Anton et al., 2021). Computational design is a broad field, and as with the other fields, there is of course a lot of cross-over. Many optimization strategies could be classified as development in computational design tools, or also as increasing the structural performance, and therefore within the field of ‘structure and performance’. Here, computational design is defined as the development of design tools, modelling, simulations, and print path generation.

Finally, where material science looks at the performance of 3DCP at a material level, the field of structure and performance incorporates structural reinforcement, structural optimizations, and finite element (FE) analysis of 3DCP elements.

Analyzing the set of 156 recently published papers through this framework allows a deeper understanding on the trajectories and fields of research in 3DCP. Figure 4 shows that 47% are based within material science, 20% in structure and performance, 16% in computational design, and 17% as overall reviews on 3DCP. These were sorted firstly through a numerical analysis of the papers, using a keyword filter to categorize the papers into the framework based on a set of predefined terms, for example if a keyword of the paper had 'rheology’, it was labelled as material science. This sorting was then cross checked by the author based on paper titles and abstracts.

Fig. 4
figure 4

Proportion of recently published papers divided into framework

Full size image

Running the data through VOS viewer (VOSviewer, n.d.) further shows the dominance of material science. Figure 5 shows the keyword data extracted from the papers. Red represents tags related to material science, blue to computational design, green to structure and performance, purple to sustainability, and yellow to reviews. Each tag is weighted depending on its occurrence and connected via a line for every time there is co-occurrence, this line also being weighted based on the number of co-occurrences. The relatively small size and scale of the purple sustainability tags, and visibly low level of connections between the different research fields, further highlights the need for a holistic focus on the sustainability aspects within 3DCP research.

Fig. 5
figure 5

VOSviewer of keywords from recent academic papers, highlighting the scale and scope of different research fields, their interconnectivity and research focuses, and also the small proportion of papers which note sustainability as a key focus

Full size image

2.2 Claims of sustainability in 3DCP

When reviewing literature related to 3DCP, there are multiple claims and statements made about the sustainability of the process and environmental benefits. Yet to date, there seems a distinct lack of critical investigations into the source of these (Flatt & Wangler, 2022; Tinoco et al., 2022).

The following section will investigate and unpack a series of identified exemplar claims found in 3DCP literature. Using the framework, it is possible to divide the claims into 7 main sustainable development trajectories within 3DCP (Table 1). By analyzing each claim, it allows us to understand why each trajectory has formed, whilst also identifying potential issues yet to be fully addressed.

Table 1 Categorizing sustainability claims in 3DCP using the framework: Identifying 7 Sustainable research trajectories with their exemplar claims
Full size table

2.2.1 Sustainability trajectory 1: full size aggregates:

It is widely considered that one of the most impactful factors in the environmental sustainability of 3DCP is the high mortar content, in part due to the small aggregates which are currently used for printing. Currently, there are no full-scale solutions to printing concrete with full-size aggregates (up to 37.5 mm) (Kuzmenko et al., 2020). This is due in part to technical limitations: current printer set-ups cannot take large aggregates as they are prone to blocking the feed pump or hoses. Also, the material composition: especially during its fresh state, its pumpability and extrudability can be affected (** countries. Whilst Fly ash and Slags have been shown to reduce the CO2 emissions in concrete production, the replacements themselves are by-products from other unsustainable productions, and the supply of these replacements are limited globally (Ma et al., 2022).

A more promising replacement is calcinated clay. It is globally abundant, and the production process is significantly less than OPC. The calcination process requires fewer fuels and emits much less CO2 compared with that of clinker production (Chen et al., 2022). This reduction is due to the lower burning temperature of the calcinated clay compared to clinker. To assess the reduction in emissions, an LCA study was performed based on the Cuban cement industry. They found that the emissions are around 0.25–0-37 kg CO2/kg for the calcinated clay, compared to around 0.9 kg CO2/kg for OPC (Scrivener et al., 2018a, b). Whilst these numbers are promising, further assessments are necessary to understand whether the same figures are applicable in other geographies. Furthermore, replacing OPC with calcinated clay does not tackle the issues of resource scarcity. Whilst clay may be abundant today, it is still a finite resource. Should research not learn from the past that it is unsustainable to develop construction materials which rely heavily on these?

An emerging future alternative is engineered biology-based cements. Products such as 'Biocement', a material based on the generation of coral production, offer 'the potential to eliminate 1 kg CO2 for every kg of biocement used instead of OPC (Biomason, 2022). Whilst this technology sounds promising, it is still in early development. The only application within the AEC industry is through pre-cast tiles, and it is still unknown whether it is applicable to AM techniques. However, if these challenges are overcome, and the figures substantiated, it is an exciting future development within the field of material science.

2.2.4 Sustainability trajectory 4: optimizations

Compared to competitor building materials, concrete has a relatively low carbon footprint when normalized by volume, suggesting that it is the quantity of concrete used which is causing such a large impact on the environment (Barcelo et al., 2014). A key argument for the sustainability of 3DCP is that it will reduce the quantity of concrete used in construction, therefore reducing the CO2 emissions, making it more sustainable (De Schutter et al., 2018; Wangler et al., 2019).

This claim can be found in many articles and commentary across 3DCP, both in academia and industry. A recent promotional post for a 3DCP company stated, 'in comparison to traditional concrete walls that are 100% casted, our 3D Printed Walls only have an infill of 45% resulting in concrete material savings of 55%' (LinkedIn, n.d.). This post exemplifies this category of argument. If we unpack this statement: If a simple, straight wall element of 1 × 1x0.2 m is cast, it will have a volume of 0.2 m3. If a print path of 30 mm is used to print the perimeter of the wall (0.07 m3) with 45% infill of the interior of this (0.06 m3), this would be 0.13 m3 total material used, resulting in a 35% material saving. This only takes into account one parameter of material optimization, and does not consider other performance-based measurements, such as acoustic or thermal properties, which will also contribute to the overall environmental impact of an element.

Reduction of material use is becoming critical, as resource scarcity and production emissions grow. As a solution, digital fabrication has emerged, and is 'expected to lead to more sustainable construction, due to more efficient structural design by placing the material only where it is needed' (Wangler et al., 2016). Topology optimization, as a design tool for 3DCP, is considered a key design tool to unlock the potential of 3DCP (Gaudillière-Jami et al., 2019). A built example of this is Krypton, 3DCP structural column for a building in Aix-en-Provence. The form of the column was based on a lattice structure, designed using topology optimization to reduce the volume of material used whilst still considering the loading conditions. The project also proposed novel tangential toolpaths, to achieve layers with variable thickness and mechanically sound construction (Gosselin et al., 2016).

Toolpath manipulation offers designers the opportunity to create additional functionality through complex geometry, and some suggest it could be used to integrate and automate services within structures in an efficient way (Scrivener et al., 2018a, b). Anton et al (2021) present a series of 3DCP columns with the suggestion that the complex print paths create cavities, which could be used for secondary functions. XtreeE produced a 3DCP wall section which integrates drainage, light, and watering to make a living wall (XtreeE, n.d.-b). Whilst integration of services is an appealing thought, especially as a solution to reducing the material layering in architectural structures and simplifying and speeding up the construction process, the serviceability of these structures has not been tested. Access and repair are critical in architectural element. What would happen if a pipe burst, and the only way to repair it is to break through the 3DCP form? How would this affect the structural integrity, and visual appearance of the element? These are questions which are yet to be fully addressed.

An alternative strategy of increasing functionality could be by manipulating the toolpath to improve the thermal, acoustic, or fire performance of the elements. Aziz et al. (2021) Present an optimized slab structure, into which they have nested Helmholtz-resonators in the print path to absorb low frequency sounds. They further suggest that the surface roughness resulting from the printing process is ideally sized to diffuse the high frequencies. No data has yet been provided to substantiate these increased performance claims. If proven it suggests a promising trajectory for 3DCP within architectural design, by not only reducing the overall volume of material used but also reducing the number of additional building components (De Schutter et al., 2018).

2.2.5 Sustainable development trajectory 5: analysis tools

The task of simulating the behavior of 3DCP concrete is not a trivial one. Concrete has some specific characteristics in its fresh state. Particularly, its pumpability: the ease of transporting the material from the mixing vessel to nozzle deposition (Tay et al., 2019). Buildability: the capacity for the printed layer to support the subsequent layers without collapsing from elastic buckling or plastic collapse (Breseghello et al., 2021). And the printability window: the time in which the material state is at its optimum, to be smoothly deposited whilst achieving enough early-stage strength to sit in its intended geometrical form (Tay et al., 2019). Possibly the most challenging of these is the buildability: the stability of the material when it is deposited, layer upon layer (Vantyghem et al., 2021). This can often lead to unexpected results and failures in printing.

In a bid to reduce the failure rate of prints and predict the behavior of the printed artefacts, a series of simulation and analysis tools are being developed. VoxelPrint, a grasshopper plug-in, uses voxels to model complex 3DCP geometries, which can then serve as input files for FE analysis in Abaqus. The hope is by analyzing and predicting material behavior during the printing process could lead to less unpredicted prototype fails, and faster modelling (Vantyghem et al., 2021). Both Voxelprint, and Wolfs et al. (2019) are tools which work in parallel with commercial FE analysis tools. Whilst these tools undoubtedly provide accurate modelling, learning them takes time, are expensive, and often requires specialist knowledge.

Breseghello and Naboni (2022) introduce a method for visualization and simulation of toolpaths, to predict the behavior of a printed complex object. They suggest that this will allow for more rapid design iterations, and it is also implied that this will help address material wastage within failed prints. By reducing the number of unpredictable failures, this suggests that less material will be wasted during the prototype phase of design. However, the buildability of the form is only one factor that is tested in the prototy** phase. With external factors such as temperature and humidity having a high influence on the success of a print, the true material savings are unknown.

2.2.6 Sustainable development trajectory 6: formwork

3DCP is often promoted as a fabrication option to produce custom geometries without the need for formwork. With no formwork required in the sha** of the concrete, it is seen as 'opening the door to more economic and environmentally lean fabrication' (Anton et al., 2019).

The reduction in cost by eliminating formwork is a common claim, found in multiple research papers. 'The cost of formwork costing 30–60% of the total cost of a concrete structure' (Chen et al., 2017; Lloret et al., 2015; Perrot et al., 2016). Reducing or removing formwork costs is an appealing argument to an AEC Industry that is often operating with tight budgets and short timelines. The stated cost of formwork can be traced to a chapter ‘Design and Construction of Concrete Formwork’ in a handbook published in the US in 2008 (Johnston, 2008). Johnston explains that this cost is usually due to complex on-site formwork construction. No calculations are given as a basis for this figure, and the handbook is directed towards an American market, suggesting that the figures are based on construction and material costs in America. The cost difference for printing a concrete structure compared to traditional formwork construction is still unknown, and may vary considerably depending on the geometry being produced.

Beyond the economic cost, is the claim that the absence of formwork in 3DCP could help save large quantities of CO2 and waste emissions related to formwork use (Chen et al., 2017). The argument behind this is that by not needing formwork, the energy consumption for the production and waste processing of the formwork is not required. It would also need less onsite labor, which would reduce the fuel transport of those workers to and from site (Smith, 2012). But the location of the site would be critical: how much savings would actually be saved in fuel transportation of workers, for a building site in the city center of Copenhagen, where many commute by bicycle already? Also, most traditional formworks for prefabricated elements are highly efficient and re-useable, and the increased cement content of 3DCP material and the energy consumption of the fabrication method outweigh the material savings in simple geometries (Agustí-Juan et al., 2017).

Replacing or eliminating formwork using 3DCP has been shown to have an impact on economic and environmental costs, however it can also be seen through the analysis of these claims that these benefits are often more speculative, rather than tested through completed projects. It can also be deduced that these benefits are only found when fabricating complex geometries.

2.2.7 Sustainable development trajectory 7: compression only

Reinforcement has emerged as one of the most challenging, and limiting, factors in the implementation of 3DCP in industry. As there are currently no regulations when it comes to 3DCP structures, architects and engineers must find ways of meeting building codes through multiple prototype testing, or equating to static calculations of traditional constructions (Tay et al., 2019).

One solution to tackle this challenge, is the design and fabrication of compression only structures. Bhooshan et al.(2022) developed a design tool inspired by the paradigm of masonry based design, to facilitate the design of compression only structures. They demonstrate this workflow, which supports the design of large scale unreinforced 3DCP elements, in the design for a bridge. Striatus, a collaborative project by Block Research Group, and Zaha Hadid Architects, present the bridge as a demonstrator of a structure with no internal reinforcement or glued elements. This elimination of embedded reinforcement not only overcomes the technological challenge of how to embed it in a 3DCP process, but also means the printed components can be quickly and easily disassembled and re-used or recycled (Striatus, n.d.).

As the structure is compression only, this can allow the elimination of reinforcement from the concrete, one of the bigger impacting materials in structural concrete. By using a single material, they claim that no material sorting will be required, which will therefore make it more cost effective. They also state that all the components are designed to be disassembled and re-used (Striatus, n.d.), Whilst disassembly is considered a promising design strategy for architecture in the push for circularity in the construction industry, 3DCP elements are generally highly customized and fabricated specifically for a single purpose, as in the case of the Striatus bridge. How or if they could be re-purposed is unclear.

Striatus is fabricated using a 6-axis robotic arm and employs non-planar spicing and printing (Bhooshan et al., 2022). The use of robotic arms in fabrication brings higher costs, and more planning and expertise in the design to fabrication process. Lin et al. (2022) propose a fabrication method to produce compressive arches using a 3-axis gantry printer. This could be a step at reducing the fabrication cost of 3DCP, whilst still producing a compressive structure.

3 Discussion: What can we understand from these trajectories?

By Unpacking and analyzing each of the 7 identified research trajectories, it is clear there is no single solution to improving the environmental sustainability of 3DCP. Whilst Important steps are being taken to reduce CO2 emissions during production and the embodied carbon of the 3DCP material, there still remains large unanswered questions. Optimization has been shown to reduce the volume of the material used, and further computational tools to assist in the simulation and print path manipulations, however the environmental benefits of these seem limited to the production of complex forms. This may be seen as a limiting factor in the widespread adoption of 3DCP, especially when looking at the geometry of printed projects, which tend towards more standard architectural forms.

The analysis has identified unsubstantiated claims found tin both industry and academia. The discussions around the environmental impacts of 3DCP are highly debated (Flatt & Wangler, 2022). It is vital that in these discussions, both researchers and industrial partners be transparent with their calculations and assessments. This will not only allow for the development of the industry to be more sustainable, but by uncovering the challenges and limitations of 3DCP, it will help direct future research within the fields and allow the application of 3DCP, for instance within architecture, to only be used in appropriate situations.

3.1 Categorizing the development trajectories

Table 2 shows a matrix of the exemplar claims from the 7 identified sustainability trajectories within 3DCP. By unpacking each sustainability claim made within the framework, it is possible to characterize them into three main topics: 7 are about minimization of material, 2 about circularity, 5 about performance.

Table 2 Characterizing the research trajectories
Full size table

It can be deduced that the higher number of claims that focus on minimization is due to the fabrication process of 3DCP, which affects all three fields of 3DCP. Likewise, the performance of elements both in terms of environmental impact assessments and geometric performance is highlighted over all three fields. Circularity, which looks towards the end-of-life of the printed object, has comparatively fewer trajectories. It can be concluded that this is partly due to the assumption that concrete is not a ‘reusable’ material, but also as previously identified, the maturation of 3DCP has been focused mostly on material recipes and fabrication strategies, with less focus on the end of life of the printed element.

It can also be understood from the analysis, that whilst the different trajectories are being investigated and developed individually, there is a lack of comprehensive analyses or assessments on the overall sustainability aspects of 3DCP. This shows a clear need for further research which takes a more holistic approach to assessing the sustainability of 3DCP, understanding the relationships between minimization, circularity, and performance.

The architectural design process provides an opportunity to apply this holistic approach. Every project requires: specification of materials, geometric design, optimizations across material, structural, acoustic and thermal performance, and the development of construction strategies. If 3DCP is to develop with environmental sustainability at its core, each of these characteristics must be considered, and assessment criteria developed to understand the relationships between each, and how they perform compared to other fabrication strategies or architectural elements. Evaluation and assessment tools are key to develo** our understanding. Further claims, that 3DCP can solve economic and social problems, by using less material and transportation, building faster and cheaper (CyBe, n.d.-a), should also be considered.

4 On ‘sustainability’ and methods of measurement

The term sustainability is used by almost every discipline across the world and describes the capacity to endure at a certain level. Yet it is an essentially contested, and evolving, concept, in that it boasts numerous legitimate connotations which cannot be resolved through argumentation.

4.1 Sustainability in architecture

Within architecture and construction, sustainability thinking has been closely informed by multiple theoretical frameworks: carbon costing(What Is Carbon Pricing?, n.d.), circular design (The Circular Design Guide, n.d.), planetary boundaries (Rockström et al., 2009), life cycle thinking (What Is Life Cycle Thinking?, n.d.), absolute and relative sustainability (Hauschild et al., 2020), and more. Whilst early drivers of sustainability were linear models (Cheshire, 2021), seeking to optimize and reduce the material use through a buildings lifecycle, in 2002 Braungart and McDonough (2009) highlighted the lack of consideration for the re-use of materials and elements. This introduced the idea of circular design into mainstream architectural thinking.

In 2009, the concept of Planetary Boundaries was introduced by Rockström et al. (2009). This framework, which underlies much contemporary understanding of sustainability, outlines a set of nine planetary boundaries within which humanity can continue to develop without causing irreversible environmental changes. This framework provides architecture further guidelines in how to think about sustainability in construction.

‘The Handbook To Building A Circular Economy’ describes how Optimization, reduction and circularity could be incorporated into architecture, and presents how the concept of a circular economy can be translated into the building industry (Cheshire, 2021). Building on Braungart and McDonogh’s cradle-to-cradle (2009), The Ellen MacArthur Foundation present the circular economy as a series of cascading levels of material use, inviting us to consider how we use, repair, and recycle finite resources (The Butterfly Diagram, n.d.). What is noticeable within each of these systems of thinking, is the need to consider the wider picture of sustainability and its interconnectedness and avoid the separation of measurements and metrics.

In general, progression has been towards an understanding that environmental sustainability requires integrated and holistic, rather than siloed, measures. LCA, a systematic method for holistically analyzing environmental impacts across production, use, and end of life, though not a total solution, could help contribute to more tangible and practical discussions about what sustainability means for 3DCP.

LCA is a technique used to assess the environmental impacts and aspects of a product over its whole lifespan. This is measured using a functional unit, usually defined by the product, as a means to compare results (e.g., a window unit, or 1 m3 of wall) (Farjana et al., 2021). There are multiple factors which can be measured and produced from an LCA, including Global Warming Potential, Ozone Depletion, Eutrophication, and others. The main measurement which is considered within construction and 3DCP is the Global Warming Potential (GWP) unit of measurement in kgCO2-eq, as this is considered the most relevant impact category (Tinoco et al., 2022).

4.1.1 Unpacking the different system boundaries for LCAs

LCAs are defined by different system boundaries. The European standard for LCA is divided into four stages, see Fig. 6: Product (A1-A3), Construction (A4-A5), Use Stage (B1-B7), End of Life (C1-C4). A fifth stage (D) looks at benefits or loads beyond the system boundaries, such as re-use, recovery and recycling (Hollberg & Ruth, 2016). These system boundaries are more commonly defined as cradle-to-gate (A1-A3), gate-to-gate (A4-B7), cradle-to-grave (A1-C4), and cradle-to-cradle (A1-D).

Fig. 6
figure 6

Life cycle stages. Blue highlighted are the most common system boundaries used within architecture. (Hollberg & Ruth, 2016)

Full size image

When performing an LCA, the boundary conditions are chosen to respond to the purposes of the assessment. Figure 7, adapted from Bhattacherjee et al. (2021), presents a schematic sketch of the different gates in relation to 3DCP.

Fig. 7
figure 7

Adapted from Bhattacherjee et al. (2021)

Schematic sketch of LCA of a 3D printed structure, also highlighting the different gates within the LCA system.

Full size image

The different system boundaries are also more generally divided into two formats. Foreground systems: materials, construction and product use, and background systems: raw materials and processing. Traditionally, Environmental Product Declarations (EPDs) look at the background systems, whilst LCA focusses on the foreground.

EPDs are independently verified documents which provide quantitative data on the environmental impact of a given material (Agustí-Juan & Habert, 2017). The use of EPDs is vital for calculating accurate assessments of building impacts. However, whilst they are readily available for traditional mixtures, concrete recipes change for each 3DCP project, and there are currently no standardized measurements or EPDs which can be used in 3DCP LCA assessments.

4.2 LCA in architecture

It is now standard practice to perform LCAs on new architectural projects (Agustí-Juan et al., 2017). In Denmark, with the introduction of set emission targets coming into force on all new building construction from 2023 (National Strategy for Sustainable Construktion, 2021), architects must be able to have quantifiable data, showing what the impact of their designs will have on the environment.

These calculations are typically performed at two stages: Cradle-to-gate (A1-A5) which considers the production of raw materials, transport, and construction on/ off site. This is often used at earlier stages of a design process, to help guide overall design strategies. Further through a project, a broader cradle-to-grave analysis (A1-A5, B4, B6, C3, C4) is performed, allowing architects to consider the circularity of each material and building system at the end of the life (Hollberg & Ruth, 2016).

4.3 Application of LCA for 3DCP: current status and challenges

Whilst relatively comprehensive LCA tools have been developed for traditional construction methods, for example One click LCA and LCA byg (LCA byg, n.d.; One Click LCA, n.d.), these are targeted to standardized materials and processes and do not allow for easy comparisons of novel fabrication strategies. Relatively few projects have performed and reported one off LCA analysis on 3DCP elements for a design (Abdalla et al., 2021; Alhumayani et al., 2020; Gislason et al., 2022; Kuzmenko et al., 2020; Mohammad et al., 2020; Silva & Kaasgaard, 2022; Yao et al., 2020).

Of these assessments only Abdalla et al. (2021) performs an LCA on a full building, whilst others use varying scales of walls as their functional unit (Alhumayani et al., 2020; Kuzmenko et al., 2020; Mohammad et al., 2020).

To understand the application of LCA for 3DCP, we can highlight three particular challenges: definition of the system boundaries, definition of the functional unit, and the specification of material.

4.3.1 Definition of the system boundaries

As previously described, the system boundaries for LCA of architectural buildings are well defined. However, these are based on a manufacturing paradigm in which architecture is assembled from elements, rather than being printed. EPDs of elements are produced in line with these system boundaries which makes it easy to connect and compare different strategies.

When 3DCP, these boundaries are blurred. Does the system boundary of A1-A3 consider the mixing of raw materials directly for onsite printing part of A3- Manufacturing, or is that considered part of A5- Construction? Another considerable question is how and where to consider the impact of the fabrication unit itself (Kuzmenko et al., 2022). Both in its use, but also if and when the associated impacts to build, transport, and install the printer should be considered.

Currently, most research and LCA analysis of 3DCP structures perform a form of cradle-to-gate analyses (Agustí-Juan et al., 2017; Kuzmenko et al., 2020). This is because it is widely considered that the material composition, fabrication method, and quantity of material are considered the most impactful factors of 3DCP production (Agustí-Juan et al., 2017; Silva & Kaasgaard, 2022).

Bhattacherjee et al.. (2021) argue that as structural systems are traditionally designed to last multiple decades or even centuries, it is impossible to speculate end-of-life processes and therefore this measurement is not relevant, and only the early system boundaries need to be considered. However, end-of-life concrete waste is the main material in construction demolition. In the Netherlands, it is predicted that it will be 22 million tonnes by 2025. Although much of the concrete can be recycled as aggregates, the increase in concrete waste will surpass the requirement from these cascading roles for concrete (Hu, 2012). This implies that the end of life of concrete may become an increasing concern.

Whilst minimization through fabrication options such as 3DCP could help reduce this gap, it is still a potentially significant impact, and one which should be considered in environmental impact assessments. Circularity is also a potential considerable benefit. If assessment tools do not take this metric into account, then we will not have a full understanding of the impacts of 3DCP.

4.3.2 Definition of the functional unit

One of the challenges when comparing LCA on 3DCP elements is defining the functional unit. The functional unit defines the parameters for the analysis of a product, setting limits for which functions or requirements the product needs to fill (Nwodo & Anumba, 2019).

Often, the functional unit of a building element is measured in 1m2 of the material, with the depth of the material predefined by the product. In the case of 3DCP, the print width of the deposed material may vary within a single project, and in many instances the printer path is complex and varied across the element. This makes it extremely hard to define the functional unit of a 3DCP element in the standard way.

This can be seen when comparing LCAs architectural 3DCP elements. Alhumayani et al. (2020), when comparing cob vs concrete, used a functional unit of 1m2 of a load-bearing wall, with variable depth. Long et al. (2021) use 1 kg of concrete as their functional unit, and Abdalla et al. (2021) use a single story detached house of 90m2 and 4.5 m. Furthermore, each of these projects also have varying material compositions, which would further affect the LCA results.

These variations on the functional unit mean that it is a non-trivial task to compare data (Tinoco et al., 2022). Whilst there is considerable research and development being performed to try and allow better comparisons of the different material compositions (Bhattacherjee et al., 2021; Chen et al., 2022). There is also a risk that without having an easily comparable functional unit of 3DCP, comparisons to traditional fabrication techniques will remain limited. This therefore calls for a universal, but flexible, way of measuring a functional unit for 3DCP. Within architecture, what is potentially more important to consider is the overall impact of a 3DCP element within a whole building frame.

4.3.3 The specification of material

Adjustments and developments of the material composition are beyond the remit of architects for 3DCP. However, with LCA becoming more integrated into the design process, there is the potential to take a more holistic approach to LCA assessments of 3DCP.

These could be both early-stage comparisons between different printing strategies, as well as objective comparisons between different construction strategies. Considering the impact of the 3DCP on the whole building system, rather than a stand-alone element. This could then lead to smarter, hybrid buildings, where materials are used based on their performance both structurally and environmentally, ensuring that the future use of 3DCP in architecture is environmentally sustainable.

4.4 Other assessment metrics

These challenges show the limitations of using LCA for measuring the environmental impact of 3DCP. The focus of LCA measurements within architecture is GWP. However, as previously discussed, the reduction in the embodied carbon of elements is just one consideration in develo** environmentally sustainable elements. 3DCP is still a develo** construction strategy, and the long-term performance of these elements are unknown (Adaloudis & Bonnin Roca, 2021). Circularity: the end-of-life, and recyclability of printed elements must also be considered. There are emerging metrics associated with these concepts, including Circularity Indicators (Heisel et al., 2020; Madaster, 2021) and Recyclability assessments (Roithner et al., 2022) and This opens the question of what other measurements of sustainability could be included in environmental assessment for 3DCP.

Considering other metrics is necessary to ensure the assessment criteria allows for both relative comparisons between 3DCP strategies, and with non-printed traditional fabrication solutions. Whilst LCA provides feedback for material selection and optimization, uncertainty in the later phases of the LCA system is an issue and relying on data from one method of measurement may narrow our understanding. Architects are known for considering multiple aspects within a single design project. The same multifaceted approach should be employed when measuring the environmental sustainability of 3DCP elements.

5 Conclusions and critical reflections

This paper has highlighted that there is a gap in knowledge when considering the environmental impacts of 3DCP. As 3DCP moves from research to application within industry, there is a growing need for more comprehensive understanding of the environmental impacts of this unique process within construction, and for this to be incorporated into the design workflow (Kuzmenko et al., 2020).

The multi-disciplinary nature of 3DCP research and development is highlighted, and a framework is proposed to help understand and unpack the different associated development fields. This framework defines three overlap** fields: Material Science, Computational Design, and Structure and Performance. This framework is used to identify exemplar claims surrounding the environmental sustainability of 3DCP, and identify 7 research trajectories related to sustainability. These are further categorized into three main sustainable development tracks of 3DCP: minimization, circularity, and performance.

The categorization has revealed that although strongly present, sustainability concerns are siloed. It also highlights that many of these claims are untested, and that further data and analysis must be made to substantiate them, to understand what the true impacts of 3DCP are. The categorization further shows there is a general understanding that through minimization, more specifically the reduction of material use through selected deposition, this will directly lead to a lower environmental impact of the 3DCP element. However, designers must also consider the learnings from research within the Material Science field. Specifically, the higher embodied carbon per m3 of 3DCP mixtures compared to traditional large aggregate cast concretes, and the printing characteristics, strengths, and properties of the emerging new 3DCP concrete mixtures.

Until we understand the balance and relationship of these factors within the design process, it will limit our understanding of how geometry affects the overall environmental impact of 3DCP elements. How this feedback between material choice and geometric design is incorporated into the architectural design process, whether through advanced computational tools or embedded in the design space as feedback, is a topic for further investigations.

The minimization of material as a means to reduce the embodied carbon is just one parameter that must be considered for architectural structures. Acoustic and thermal performance are also key considerations during the design phase. It is therefore vital that the balance of material optimization with thermal and acoustic performance, and structural stability is understood early in the design process of 3DCP architectural elements.

Furthermore, a key and under-explored consideration is the use-phase performance. If a 3DCP wall element has lower thermal performance, this will have an impact on the energy required for heating and cooling the building. Therefore, although the element itself may be optimized to have a lower embodied carbon, this might be offset by higher energy consumption during the use phase.

To understand the full environmental impacts of 3DCP elements in an architectural project, this paper proposes that comprehensive measurement tools be developed and incorporated into a design workflow. As an assessment tool, LCA is already used and integrated into architectural workflows, and allows for comparisons across different design strategies and material choices. This leads us to suggest that if LCA could be applied to 3DCP elements in the same way, it may give us new opportunities to guide the design process of these elements.

However current LCA measurement tools cannot be straightforwardly applied to 3DCP. In particular, defining the system boundaries, the functional unit, and material specification are each challenges that need to be overcome. Furthermore, the end-of-life processing of 3DCP elements remains an uncertainty, and further consideration should be given to how architects can factor this into design.

Through the development of LCA tools specific to 3DCP, academia and industry can understand what the impactful factors of 3DCP on the environment are, and when in a design process these factors should they be considered. If this is understood, it can lead to a focus on research resources within each research field of 3DCP. Not only ensuring that as 3DCP reaches maturity and becomes more widely used across the AEC industry, but that it is developed and implemented in an environmentally sustainable way.