Optimisation of 3D Printable Cement- and Lime-Based Mortars for Built Heritage Rehabilitation

Abstract

Three-dimensional printing (3DP) represents a significant innovation in the construction sector, offering substantial benefits in terms of efficiency, customisation, and sustainability. In the context of built heritage rehabilitation, it is capable of accurately reproducing architectural elements, facilitating conservation efforts, while minimising waste and resource consumption. However, in this field, ensuring material compatibility with original structures is essential. This study explores the development and optimisation of lime and cement-based mortars for 3DP applications, focusing on their physical and mechanical performances (on moulded specimens) for use in replicating elements of a renowned Portuguese theatre. Laboratory testing supports the selection of suitable mortar compositions, aiming to balance performance and fidelity to historical construction practices. This research seeks to contribute to explore the potential of 3DP for heritage conservation, promoting innovative, durable, and culturally sensitive restoration strategies.

1. Introduction

The construction sector currently faces significant challenges, including the urgent need to adopt more sustainable and efficient practices. It is one of the world’s largest consumers of natural resources and a major contributor to global carbon dioxide emissions [1]. Alongside these environmental concerns, the traditional construction methods remain labour-intensive and inefficient, highlighting the demand for more automated solutions, both on site and in industrial environments.

Three-dimensional printing (3DP) in construction has emerged as a transformative technology, offering improved efficiency, reduced material waste and labour costs, and the ability to fabricate complex geometries without formwork [2]. While most applications remain limited to experimental buildings, façade components, and small-scale prototypes [3], the technology shows promise for specialised applications requiring precision and customisation.

However, the application of 3DP to built heritage rehabilitation remains underexplored, despite the urgent need for innovative restoration solutions. Historic buildings possess high architectural, cultural, and historical value and require restoration methods that respect the original construction materials and techniques, many of which are undocumented or no longer in use [4,5]. These structures inevitably deteriorate due to pollution, increased salt levels, biodeterioration, and mechanical stress [4], with their complex architectural features often surpassing the capabilities of traditional manual restoration techniques [6].

Historically, restoration was carried out by artisans using manual techniques such as carving, moulding, and applying lime-based mortars, relying on generational knowledge and empirical expertise [7]. In contrast, 3DP coupled with 3D scanning offers precise, rapid, and potentially non-invasive replication of damaged architectural elements, such as sculptures, mouldings, and façade components [4,8].

A critical research gap exists in the development of 3DP mortars specifically compatible with historic construction materials. Most current 3DP mortars are cement-based with a higher cement content than traditional mixtures to accommodate equipment constraints, resulting in an increased environmental impact [2]. While Portland cement offers durability, good setting times, and mechanical strength (20–150 MPa) [3], its incompatibility with historic lime-based systems can cause mechanical stress, chemical incompatibility, and accelerated deterioration of original substrates [9].

Historic mortars typically included lime, which provided flexibility and compatibility with historic substrates, reducing the risk of cracking and mechanical incompatibility [9]. Lime-based mortars date back to 7000 B.C. and were widely used in classical Greece and Rome for structural purposes, often enhanced with pozzolans [10,11]. Portland cement, introduced in the late 19th century, eventually replaced lime as the most commonly used binder [10], dominating construction practices in the 20th century [12]. However, its production accounts for up to 9% of global greenhouse gas emissions [1]. As a result, researchers are investigating alternative additives, such as silica fume [13], fly ash [14], marble powder [15], sugarcane bagasse ash [1], and even graphene [16], aiming to reduce their environmental impact, while maintaining a good performance.

Recent research has begun to explore lime-based mortars for 3DP heritage applications. While Dias et al. [17] conducted systematic material optimisation of hydraulic lime mortars for 3DP, Jesus et al. [18] focused on the methodological aspects of heritage 3DP, demonstrating the complete digital workflow from 3D survey data acquisition to 3D-printed prototypes, establishing the technical feasibility of the scan-to-print process for cultural heritage applications.

However, a significant gap remains in the optimisation of hybrid formulations with lime and cement for heritage 3DP applications. While lime offers excellent historic compatibility, it may present limitations in terms of early strength development and printing performance. Conversely, cement provides adequate printing properties, but compromises historic compatibility. No systematic investigation has addressed the optimisation of lime-cement formulations that could potentially balance the heritage compatibility of lime with the technical advantages of cement for 3DP applications.

Therefore, this study addresses the specific research gap of developing and optimising lime-cement mortar formulations for Delta WASP 40100 through systematic laboratory testing. This research gap is investigated through a case study involving the replication of architectural elements from a renowned Portuguese theatre, where lime-cement mortars were historically used in the original renovation of some of the building’s architectural elements, providing both historical precedent and practical justification for the material optimisation approach. The mortars’ performance was evaluated in both fresh and hardened states, where key properties, such as extrudability (ability of the mixture to continuously pass through an extrusion nozzle, without blocking the system, while producing uniform filaments), buildability (extrusion of filaments into stacked layers, capable of retaining their shape), mechanical strength, and water absorption by capillarity, were assessed. The main objective is to identify the mortar compositions that are simultaneously compatible with 3DP technology and heritage conservation, contributing to more precise and culturally respectful restoration practices. The technical requirements for 3DP mortars are highly dependent on the specific printing system employed, as different printers have varying nozzle geometries and extruder that directly influence material the abovementioned fresh properties.

2. Experimental Methods

2.1. Work Plan

The main objective of this study is to develop 3D printable mortar compositions suitable for application in heritage buildings, including the effect of binders, aggregates, and admixtures on their composition and their properties in both fresh and hardened states.

Figure 1 presents a flowchart summarising the experimental approach adopted for mortar optimisation. The process began by defining the mortar compositions to be tested, establishing appropriate proportions for each constituent material. After mixing, fresh-state properties were evaluated, followed by a 3DP trial. If the mortar failed to meet the required criteria (flowability, extrudability, and buildability), it was adjusted based on three main factors: mixture proportions (through water content), printing parameters (mainly extrusion factor, meaning the amount of material that is deposited at each time), and ambient conditions (air temperature and relative humidity, which make it harder to control). The procedure, fresh-state testing and printing was then repeated. Once the mortar demonstrated a proper behaviour, samples were moulded for mechanical strength tests (compressive and flexural strength) and water absorption by capillarity tests. Finally, the results were compared with reference values reported in the literature.

2.2. Equipment

A Delta WASP 40100 3D printer (Figure 2a), from the company WASP, established in Massa Lombarda, Italy, was used for printability (preservation of extrudability and buildability throughout the entire printing) assessment, coupled with a WASP Manual Feeding Extruder (Figure 2b) with a round Ø 8 mm nozzle. The printing technology employed is known as LDM (Liquid Deposition Modelling), which allows for printing with various materials, such as ceramics, clays, cement, geopolymers, and other mixtures.

To ensure proper homogenisation of the developed mortars, the Kenwood Titanium (Havant, UK) Chef Baker XL KVL65.001WH electric mixer [19] (Figure 4b), from Kenwood, founded in Woking, England was used. Although originally designed for culinary use, it proved effective in preparing small mortar batches. The mixer features six adjustable speed settings, which were modified throughout the mixing process.

In addition to the main materials and equipment, auxiliary tools were also used, including a digital scale with a precision of ±0.05 g to ensure accurate measurement of the components. The printing base was made of wood, coated with a melanin film, chosen for its stability and smooth surface, providing good printing conditions.

Figure 2. (a) Delta WASP 40100 3D printer [20] and (b) WASP Manual Feeding Extruder [21].

Figure 2. (a) Delta WASP 40100 3D printer [20] and (b) WASP Manual Feeding Extruder [21].

2.3. Mortars’ Preparation

2.3.1. Materials and Formulations

To meet 3DP requirements, while ensuring compatibility with the rehabilitation of the theatre, material selection was based on their suitability for use in a historical construction context.

The main components used in the mortar formulations were as follows:

• Binders: Hydraulic lime HL5 and white Portland cement CEM I 52.5R (BR), both from Secil;
• Aggregate: River sand with controlled particle size below 1 mm, ensuring a homogeneous distribution;
• Water: Regular tap water;
• Admixture: A superplasticiser (SP) (Chryso Optima 206, polycarboxylate-based) was used to reduce the water content in the mixes.

Two main mortar formulations were defined based on the abovementioned constituents: one with 70% binder and 30% aggregates, and another with 60% binder and 40% aggregates. For each of these, three combinations of binders were tested, resulting in a total of six formulations (as pictured in Figure 3) without SP. The best ones were then tested with SP. For each printing trial, a fixed quantity of 1 kg (binder + aggregates) was prepared, sufficient to print each test element.

2.3.2. Mixing Process

Mix preparation is a critical step to ensure the quality and consistency of the extruded material. Although the procedure is relatively simple, it requires careful attention to avoid dosing errors that could compromise this study’s reliability.

The first step involves the accurate weighing of all the mortar constituents (binders, aggregates, admixtures, and water) using a digital scale with a precision of ±0.05 g as previously mentioned (Figure 4a). Once weighed, the dry materials were placed in the mixer and blended for approximately 1 min at the lowest speed (30–60 rpm) to ensure a homogeneous distribution of components, reducing the risk of segregation and improving workability (right balance of extrudability and buildability to be printable). Then, water was gradually added and mixed for 2.5 min, while progressively increasing the mixer speed (100–200 rpm). Next, the sides of the bucket and the mixing paddle were scraped to ensure complete incorporation of all materials, followed by an additional 2.5 min of mixing (Figure 4b).

Figure 4. (a) Constituents weighing. (b) Mixing process.

Figure 4. (a) Constituents weighing. (b) Mixing process.

Fresh state tests were then performed, and finally the mortar was mixed for another 2 min before being placed in the stainless steel cone of the WASP Manual Feeding Extruder, with a capacity of 2.5 L. The extruder also has a stepper motor connected to a screw responsible for regulating the material flow toward the nozzle. The printer was operated through an interactive control panel, which allowed the user to start and stop the extrusion, adjust the printing speed, and the extrusion rate during printing (screw rotation). The G-code files defining the print path were uploaded via SD card. Before printing, a rapid extrusion test was conducted to check the extrudable capacity of the mortar.

During deposition, the control panel was used to adjust the extrusion rate, since layer height and printing speed were fixed to ensure continuous flow and layer stability. Continuous monitoring was required to detect flow irregularities or spreading, since these directly affected printing parameters. Once printing was completed, the specimens were left to set under controlled laboratory conditions (24 °C and 60% RH).

This preparation protocol was strictly followed for all the tested compositions to maintain consistency and control of experimental variables.

2.4. Test Procedures

2.4.1. Fresh State Tests

Fresh state tests are essential for assessing mortar consistency, particularly through slump and flow tests.

In the tests conducted, the mortars were printed immediately after mixing, typically from 60 to 120 s, the time required to load the material into the printer’s feeding cone.

The slump and flow table tests were both performed in accordance with EN 1015-3:1999 [22] to evaluate the workability, consistency, and flowability of the mortars under study (Figure 5). These tests can provide valuable indicators of a material’s behaviour during extrusion, particularly in assessing whether the mixture possesses adequate workability for continuous and uniform extrusion. While the slump test focuses on the mortar’s ability to retain its shape, which can be related to buildability, the flow table test evaluates how easily the mortar spreads over a surface, offering complementary insight into its flowability and overall workability.

2.4.2. Hardened State Tests

Hardened state tests are essential for evaluating the mechanical performance and durability of mortars. In this study, these tests were conducted only on the mortars that showed the best fresh-state performance, focusing on compressive strength, flexural strength, and capillary water absorption, with moulded samples rather than printed elements since moulded specimens allowed for efficient and consistent evaluation without the variability introduced by printing parameters.

It is acknowledged that 3D-printed elements can exhibit anisotropic behaviour due to factors, such as interlayer adhesion (quality of the bond between successive layers), extrusion geometry, and printing paths, so future work will extend this research to include systematic testing of printed specimens to fully capture anisotropy, interlayer adhesion, and structural performance under realistic printing conditions.

The compressive strength test was carried out in accordance with EN 196-1:2016 [23], with minor adaptations for 3DP (Figure 6a). The objective is to determine the mechanical resistance of hardened mortar under compressive loads, expressed in megapascals (MPa). This test is crucial for evaluating the material’s capacity to bear structural loads.

The flexural strength test followed the same standard as the compressive strength test (EN 196-1:2016 [23]), and the results are also expressed in MPa (Figure 6b). This test is important for assessing the structural integrity of printed mortar elements.

As with the compressive strength test, the load application rate was slightly modified from the standard to align with previous tests, ensuring comparability across all the results.

The capillary water absorption coefficient, A_w, expressed in kg/(m²·s^0.5), represents the amount of water absorbed by a material over time and corresponds to the mass of water absorbed per unit area of the material in contact with the water (Figure 6c). The test follows the standard ISO 15148:2002 [24]. The test procedure comprises the following steps: prismatic specimens measuring 40 mm × 40 mm × 100 mm are used; their sides are sealed with epoxy resin so that absorption occurs only vertically through the base of the specimen; the specimens are stored in a laboratory environment at a temperature of 23 ± 2 °C and relative humidity of 50 ± 5% until they reach a constant mass; after stabilisation, the specimens are placed with their base submerged in water at a height of 5 ± 2 mm; and the mass of each specimen is recorded at different time intervals (5 min, 10 min, 20 min, 40 min, 60 min, 2 h, 4 h, 8 h, and 24 h), always maintaining a constant water level during the tests.

3. Data Analysis

This section presents the analysis and interpretation of the experimental results obtained from the laboratory procedures described previously. The goal is to assess the performance of the developed mortar formulations and their feasibility for simultaneous use in 3DP and heritage building rehabilitation. The evaluation covers both fresh-state and hardened-state behaviours.

Key variables, such as binder, aggregate, and water contents—as well as water-to-binder (w/b) and aggregate-to-binder (a/b) ratios—were analysed to understand how formulation changes influence mortar performance. Particular attention was given to extrudability, buildability, layers stability, and surface finish, as these properties directly affect print quality and structural reliability.

The results were compared with values reported in the literature to benchmark performance and validate the findings. However, due to the limited number of studies specifically addressing lime- and cement-based mortars in 3DP for heritage applications, comparisons were made with related studies.

3.1. Printing Validation

The main objective of this step was to evaluate the printing performance of the mixtures developed, which relies on a careful balance between the materials, printing equipment, design, and ambient conditions, with at an air temperature of 24 °C and a relative humidity of 60%, which are usual laboratory conditions. To assess printability, a standardised 3D model was used, consisting of 20 layers with a height of 2.50 mm each, except for the first layer, which measured 3 mm, resulting in a total expected height of 50.50 mm and a printing speed of 40 mm/s, resulting in printing time for each specimen of approximately 5 min. The layer height was chosen based on nozzle diameter, resulting in layer widths ranging from 10 mm to 15 mm. This combination of layer height and print speed was defined in previous tests, resulting in an appropriate interlayer adhesion and a print without blockages and where the material is not dragged. This model proved adequate for observing key printing characteristics and is shown in Figure 7.

To ensure consistent and objective evaluation, a set of qualitative criteria was established, focusing on surface finish, dimensional consistency, structural integrity, and overall stability of the printed element.

Table 1. Printed pieces evaluation.

Criteria	1 (Very Poor)	2 (Poor)	3 (Acceptable)	4 (Good)	5 (Very Good)	Evaluation Method
Finish	Surface with flaws and visible defects	Irregular surface with occasional flaws	Ok surface with small flaws	Smooth surface, few visual defects	Smooth and uniform surface with no defects	Visual inspection
Layer dimensions (thickness)	<80% or >200% of nozzle diameter (8 mm)	80–99% or 171–200% of nozzle diameter	100–110% or 160–170% of nozzle diameter	111–159% of nozzle diameter	120–150% of nozzle diameter	Visual inspection + calliper measurement
Structural integrity	Cracks, layer displacements	Some cracks or misalignments	Good bond between layers	Stable structure, no visible cracks	Fully cohesive and intact structure	Visual inspection
Printability	Impossible to print	Printing with frequent interruptions	Continuous printing with minor defects	Smooth and continuous printing	Perfect printing, no interruptions	Visual inspection
Collapse	Specimen collapses completely	Partial collapse evident	Deformed but printable structure	Stable piece with good geometry	No collapse, complete stability	Visual inspection

presents these criteria, distinguishing between acceptable and non-acceptable outcomes, along with the methods used for their assessment.

The criteria were defined from observations during the printing trials. For instance, when the layer widths were very close to the nozzle diameter (100–110%) or smaller, the upper layers often became less stable. The narrower base reduced contact and support, which sometimes led to misalignment, small edge defects, and even local buckling. For this reason, these cases were considered “Acceptable” or “Poor” rather than “Good.” By contrast, layers in the 111–159% range provided a wider base, improving support and interlayer adhesion. This generally produced smoother surfaces and more regular geometry, which justified their classification as “Good.” However, once the width exceeded this range, the mixtures tended to be too fluid, causing spreading, loss of dimensional accuracy, and in some cases partial collapse.

Table 2. Mortars developed and tested without SP.

Mixture		Cement (g)	Lime (g)	Sand (g)	Water (g)	w/b	a/b
1	1.1	350	350	300	280	0.400	0.429
	1.2	350	350	300	285	0.407	0.429
2	2.1	420	280	300	260	0.371	0.429
	2.2	420	280	300	265	0.379	0.429
	2.3	420	280	300	270	0.386	0.429
3	3.1	525	175	300	260	0.371	0.429
4	4.1	300	300	400	260	0.433	0.667
	4.2	300	300	400	250	0.417	0.667
5	5.1	360	240	400	250	0.417	0.667
	5.2	360	240	400	240	0.400	0.667
6	6.1	450	150	400	250	0.417	0.667
	6.2	450	150	400	240	0.400	0.667
	6.3	450	150	400	230	0.383	0.667

presents the experimental results obtained for mortar. In addition to the dosage of each constituent, the w/b and a/b ratios were calculated, as these are essential for controlling fresh-state properties and estimating the required water content for the mixtures. Visual inspection of the samples allowed for assessment of the criteria established in Table 1, as will be further discussed throughout this section.

Table 3. Evaluation of 3D-printed pieces without SP according to Table 1.

Mixture	F	LD	SI	C	P	Observations	FE [1,2,3,4,5]
1.1	3	3	3	3	3	Acceptable result; some deformations; layers not fully uniform; exudation observed	3
1.2	2	3	2	1	3	Uniform layers; lack of buildability; collapse due to excess water	2.2
2.1	1	2	3	2	2	Poor extrudability; visible mortar gaps and defects	2.0
2.2	3	3.5	3.5	3	3	Still showing flaws and imperfections compared to mix 2.1	3.2
2.3	4	4	4.5	4.5	4	Uniform layers and minimal deformation; almost perfect	4.2
3.1	5	5	5	5	5	Perfect result; uniform layers; good buildability; no deformation; no excess water	5
4.1	3	3	3.5	3.5	3.5	Acceptable; visible exudation and some deformations	3.3
4.2	4	3	3	4	4	Good final surface finish	3.6
5.1	3	4	4	4	4	Uniform layers; good printability; some deformations	3.8
5.2	4.5	4.5	4.5	4.5	4.5	Uniform layers; good surface finish	4.5
6.1	3	2	3	3	3	Good printability; exudation observed	2.8
6.2	3	3	3	3	3	Some imperfections and deformations	3
6.3	3.5	4	4	4	3.5	Good finish; good printability; good flowability	3.8

F—finish; LD—layer dimension; SI—structural integrity; C—collapse; P—printability; FE—final evaluation.

summarises the quantified individual performances of only the mortars that were extrudable. Each criterion was quantified from one to five, where one is unacceptable, two is poor, three is acceptable, four is good, and five is very good, and an arithmetic average was calculated to generate a final rating.

All the specimens were printed using the same extrusion factor, which was adjusted as needed for each mixture. The w/b closely aligned with the values reported in the literature for mortars without admixtures, typically ranging from 0.35 to 0.40 [25]. Another critical factor influencing performance is the particle size distribution of the materials used; finer particles, such as those in cement, significantly increase the water demand due to their higher surface area, which is necessary to maintain workability [26].

The a/b varied across the formulations. This ratio directly affects both the mechanical strength of the mortar and the quality of the printed component. A higher aggregate content can compromise surface finish, resulting in a rougher texture. Conversely, increasing the binder content generally enhances buildability.

Out of the first six tested mixtures, 2.3 and 3.1 produced excellent results. Mixture 2.3 performed very well, with minimal deformation, consistent layer geometry, and smooth surfaces. Composition 3.1 achieved a perfect score (5.0) across all the categories, showing an excellent finish, highly uniform layers, strong structural integrity, complete resistance to collapse, and flawless printability. Notably, both exhibited very similar w/b ratios. Additionally, it was observed that increasing the proportion of lime relative to cement led to higher water demand for successful extrusion.

In the second round of tests, the main variable altered was the binder content, which was reduced from 70% to 60%. This adjustment resulted in a less water demand, as previously discussed. From this analysis, it can be concluded that the mortars with a higher binder content generally outperformed those with a higher aggregate content, supporting the findings from Malla et al. [27], who noted that while higher aggregate volumes can improve buildability, they tend to reduce extrudability and print quality. Moreover, increasing aggregate content raises the pressure required for pumping, if a pump is used [26]. In this round, composition 5.2 was the best, with a similar evaluation to 2.3.

In both the testing rounds, the mortars that produced better surface finishes tended to have a higher proportion of cement relative to lime. This may be attributed to the distinct properties of each material. Cement, with finer particles than lime, enhances cohesion (ability of the mixture to hold together without segregation or bleeding during mixing, printing, and curing) and reduces porosity, resulting in more defined and uniform layer deposition [28]. A higher cement content also accelerates green strength gain, improving buildability, a critical factor for print success. Lime, on the other hand, improves workability, but reduces initial stiffness, making the mix more susceptible to deformation or collapse after extrusion, as observed with specimen 1.2.

Several other mortars performed well, but showed minor limitations. Compositions 4.2, 5.1, 6.3, and 4.1 could potentially be improved through fine-tuning of the w/b ratio or additive content.

At the lower end of the performance spectrum, formulations 1.2, 2.1, and 6.1 underperformed. Mix 1.2 collapsed due to excess water content, despite producing relatively uniform layers. Mix 2.1 suffered from poor extrudability and visible voids, while mix 6.1 exhibited a combination of exudation and poor surface finish.

Overall, collapse resistance and finishing quality were the most sensitive indicators of performance variation across all the mixes. The best results were consistently associated with a higher binder content, particularly higher cement-to-lime ratios and a well-controlled water content. These findings confirm that increasing binder content not only improves early strength and cohesion, but also enhances the structural buildability and visual definition of printed elements. Conversely, a higher aggregate content, while being beneficial for buildability, often compromised surface quality and extrudability. These trends are consistent with previous findings in the literature and reinforce the need for balanced formulations tailored specifically for 3DP in heritage applications.

After characterising the mortars in both fresh and hardened states, the best-performing formulations were selected for the next phase, printing tests incorporating a SP, as shown in

Table 4. Mortars developed and tested with SP.

Mixture		Cement (g)	Lime (g)	Sand (g)	Water (g)	SP (g)	w/b	a/b
2	2.1	420	280	300	210	3.50	0.300	0.429
	2.2	420	280	300	220	3.50	0.314	0.429
3	3.1	525	175	300	225	3.50	0.321	0.429
	3.2	525	175	300	220	3.50	0.314	0.429
	3.3	525	175	300	215	3.50	0.307	0.429
	3.4	525	175	300	205	3.50	0.293	0.429
5	5.1	360	240	400	180	3	0.300	0.667
	5.2	360	240	400	200	3	0.333	0.667
	5.3	360	240	400	205	3	0.342	0.667
	5.4	360	240	400	204	3	0.340	0.667
	5.5	360	240	400	202.50	3	0.338	0.667
6	6.1	450	150	400	180	3	0.300	0.667
	6.2	450	150	400	185	3	0.308	0.667
	6.3	450	150	400	187.50	3	0.313	0.667
	6.4	450	150	400	195	3	0.325	0.667
	6.5	450	150	400	192.50	3	0.321	0.667
	6.6	450	150	400	190	3	0.317	0.667

. The primary goal of adding SP was to improve workability and flowability without increasing the w/b ratio. The same performance properties previously evaluated were reassessed in this phase.

The SP was added at a fixed dosage of 0.5% relative to the total binder mass content in each mix. The selected formulations for this phase were 2, 3, 5, and 6. Multiple laboratory sessions were carried out to optimise these mortars, each with distinct characteristics.

Once again,

Table 5. Evaluation of 3D-printed pieces with SP according to Table 1.

Mixture	F	LD	SI	C	P	Observations	FE [1,2,3,4,5]
2.2	3	3	4	4	4	Extrudable mortar with no clogging; flow adjustment noticeable in layer thickness	3.6
3.4	4	4	4	4	5	Very uniform and nearly perfect finish; some variation in layer thickness observed	4.2
5.2	2.5	3	3	4	3.5	Moderately consistent layers with preserved geometry; rough finish, but structurally solid and with good strength—suitable for functional use where aesthetics are not a priority	3.2
5.3	2	3	3.5	4	3.5	Structurally robust; poor finish with thick and irregular layers; noticeable roughness on the top surface	3.2
5.4	1.5	2	2.5	2	2	Highly inadequate mix	2
5.5	3	3.5	3	4	3.5	Some surface roughness; consistent texture; uniform layers; increased flowability	3.4
6.2	2	2.5	3	3	3	Some irregularities; lack of water	2.7
6.3	3	2	3	3	3	Acceptable finish despite clogging at the end; some layer thickness variation	2.8
6.5	3	4	4	4	4	Easy extrusion; consistent flow; acceptable finish	3.8
6.6	4	4	4	4	4	Good mortar; uniform layer thickness; best among the 6.3 formulations	4

F—finish; LD—layer dimension; SI—structural integrity; C—collapse; P—printability; FE—final evaluation.

summarises the quantified individual performances of only the mortars that were extrudable.

Firstly, attempts were made to optimise formulation 1, but after several unsuccessful trials, it was concluded that mortars with a higher lime content (such as 1 and 4) were unsuitable, and therefore excluded from SP optimisation. Given this, multiple printing trials were conducted on the four previously selected formulations. The optimisation process was iterative and somewhat empirical, resulting in several non-viable mixtures before reaching suitable ones. The initial w/b ratio was based on values reported in the literature, approximately 0.30.

The w/b ratios of SP-based mixes ranged from 0.29 to 0.34, aligning with values commonly found in the literature. SP was added at a dosage of 3.5 g for mortars with 70% binder and 3.0 g for those with 60% binder, representing 0.5% of the binder mass.

Although the use of SP allows for a significant reduction in water content within the mixtures, this study also revealed some drawbacks associated with its application. The mortars incorporating SP exhibited a faster loss of workability, commonly referred to as a reduction in open time. This effect can negatively impact the print performance by causing interruptions and delays in the printing process, potentially leading to extruder clogging.

This reduction in open time appears to be directly related to the lower w/b ratios in the SP-modified mixes, which become less workable over time. Similar findings were reported by Zhang et al. [29], who also linked shorter open times to lower w/b ratios. While SP improves initial flowability, its use demands careful control, particularly in terms of SP type and dosage.

Among the tested formulations, 3.4 stood out, achieving an FE of 4.2. This mix demonstrated very uniform layers, an excellent surface finish, and minimal variation in layer thickness, confirming its consistency and suitability for 3DP when properly adjusted. Similarly, formulation 6.6 achieved an FE of 4.0, with uniform layers and reliable structural behaviour. Mixture 6.5 also performed well, with a score of 3.8, indicating that this water range is optimal for the formulation. These findings highlight the importance of precise control of water content to achieve a balance between flowability and structural stability.

The other formulations showed solid intermediate results. Formulation 2.2 achieved a final score of 3.6, showing good extrudability and manageable layer thickness variation. Mixtures 5.5 and 5.3 scored 3.4 and 3.2, respectively. These mixes presented acceptable surface finishes and structurally sound layers, although minor surface roughness and irregularities were observed.

On the other hand, some mixes underperformed. Mixture 5.4 exhibited poor cohesion and inadequate print quality, likely caused by excess water. Mixtures 6.2 and 6.3 lacked sufficient water, which led to reduced flowability, extrusion difficulties, and layer irregularities.

Overall, the results demonstrate that even small variations in water content (as little as 2.5 g) can significantly affect mortar performance. Mixtures 6 showed consistent improvement across the trials, with an optimal performance at 190 g, suggesting this as the most balanced formulation within that series. The results achieved confirm that SP alone is not sufficient to ensure ideal printability. Instead, it requires a retarder and precise water control to ensure a balance between early workability and buildability.

These findings reinforce that the success of 3D printable mortars depends not only on the presence of admixtures, but also on the precise calibration of dosages, especially water content, tailored to the behaviour of each formulation.

The best printed samples are shown in Figure 7.

3.2. Test Results

3.2.1. Fresh State Tests

The analysis of fresh-state behaviour is a critical first step in evaluating a mortar formulation. For successful application in 3DP, the mortar must be extruded in a continuous and stable manner, without clogging or segregation, and must retain its shape after deposition to ensure the integrity of subsequent layers. To assess these characteristics in the early stages, slump and flow table tests were performed.

The flow table test provides an insight into the mortar’s workability and flowability, while the slump test offers a preliminary measure of the material’s buildability, that is, its ability to support additional layers without deformation or collapse.

Figure 8 presents the results of the fresh-state tests performed on all the optimised formulations as a function of the w/b ratio.

From the results presented above, it can be observed that the slump flow values for the mortars without SP ranged from 201 mm to 220 mm, which aligns well with the values reported in the literature for this type of material. According to Ma et al. [30], values between 174 mm and 210 mm are considered suitable to ensure good extrudability and buildability in mortars for 3DP. Tay et al. [31] further suggests that flow values between 150 mm and 190 mm are associated with improved surface finish and reliable structural build-up.

Regarding the slump test results, the recorded values ranged between 10 mm and 16 mm, which also fall within acceptable limits. These results indicate satisfactory shape retention, a positive indicator for buildability. The mixes with the lowest slump, specifically 3.1 and 4.2, exhibited higher cohesion and structural resistance, which were reflected in their superior printing performance.

However, variations in constituent materials lead to distinct rheological behaviour, explaining some of the deviations from literature values. Factors such as the binder composition, particularly the cement-to-lime ratio, the particle size distribution of aggregates, and the w/b ratio directly influence the mixture’s flowability. In general, a higher w/b ratio results in a higher slump and flow values, indicating greater fluidity, but less cohesion.

3.2.2. Hardened State Tests

Only the mortars that demonstrated the best performance in the fresh state were selected for hardened strength testing. The results for the compressive strength testing are indicated in Figure 9.

The results clearly show that increasing the cement content leads to higher compressive strength. The highest strengths for formulations without SP were recorded for 3.1 and 6.3, which contain 75% cement and 25% lime, reaching 64.2 MPa and 62.2 MPa, respectively, exceptionally high values even when compared to the other mortars developed specifically for 3DP.

In contrast, formulations 2.3 and 5.2, with a lower cement content (60% cement and 40% lime), displayed the expected reduction in compressive strength, reaching 50.7 MPa and 48.9 MPa, respectively.

It is important to note that while compressive strength in the literature is typically assessed at 28 days, the values reported in this study correspond to tests performed at 7 days. Even so, when compared to conventional mortars, the performance of these formulations stands out. For example, Zagaroli et al. [32] evaluated two traditional mortar compositions—1:1:6 and 1:2:9 (cement/lime/sand)—and obtained compressive strengths of 7.91 MPa and 4.16 MPa at 28 days, significantly lower than the values reported here.

This disparity is expected, as 3DP mortars require more strength early on to support the successive deposition of layers. In contrast, conventional mortars do not demand immediate mechanical stability. Some specialised 3DP formulations, such as carbon fibre-reinforced mortars developed by Li et al. [33], have achieved compressive strengths between 48 and 52 MPa at 7 days in cast specimens. However, when tested on printed specimens, the values are typically lower due to lack of compaction and the quality of the interlayer adhesion [34].

All mortars containing SP outperformed those without it in compressive strength, confirming the positive influence of SP on mechanical performance. This improvement is primarily due to its ability to reduce the water content, which decreases porosity and increases material density. The highest compressive strengths coincided with the lowest w/c ratios, close to 0.3, a considerable decrease compared to the mortars without SP, which had w/c ratios near 0.4. The mixtures that achieved the highest strength values, with and without SP, were those with the highest cement content, namely series 3 and 6. Moreover, the type 3 mixtures performed slightly better than the series 6, which can be attributed to their lower a/b ratio, as series 3 contains 30% aggregate, while series 6 incorporates 40%, as shown in Figure 3.

These elevated compressive strengths can be attributed to several formulation factors. The use of high-strength cement as the primary binder (CEM I 52.5R (BR)), combined with relatively low aggregate percentages, produced dense, paste-rich matrices that enhanced the mechanical performance. The a/b ratios optimised for 3DP favour a higher paste content compared to traditional heritage mortars due to printing equipment constraints, which typically employ higher aggregate volumes for economic and compatibility reasons. Also, water is essential for hydration; however, when present in excess (high w/c ratio) it reduces compressive strength by weakening the molecular structure and limiting the mechanical interlocking between cement and aggregate [35]. While these strengths exceed those of conventional heritage mortars, they reflect the material adjustments necessary to achieve adequate printability characteristics for the Delta WASP 40100 3D printer.

Figure 10 shows the results for the flexural strength test.

The highest flexural strength (R_f) values for mortars without SP were observed in the mortar formulations with the highest cement content, specifically, 3.1 and 6.3, consistent with the trends noted in the compressive strength tests. However, unlike the compressive strength results, an increase in aggregate content and a corresponding decrease in binder did not lead to improved flexural strength.

When compared to the literature values, Zagaroli et al. [32] reported flexural strengths of 2.21 MPa for a 1:1:6 (cement/hydraulic lime/sand) mix and 1.23 MPa for a 1:2:9 mix, significantly lower values, but reasonable given their traditional mortar compositions. In contrast, Li et al. [33] achieved flexural strengths between 6 and 8.3 MPa in cement-based mortars for 3DP reinforced with fibres, confirming the substantial performance gain from fibre incorporation.

Unlike compressive strength, SP did not consistently improve the flexural strength across all the tested mixtures. In fact, a decrease was observed in formulation 6.6, primarily due to one outlier specimen (specimen 3), which showed a significantly lower strength compared to the other two. This outlier skewed the average downward, resulting in an unexpected drop. In contrast, the flexural strength of formulation 5.5 remained unchanged with the addition of SP.

These findings suggest that mortars with a higher aggregate content may not benefit as significantly from SP, at least in terms of flexural performance. On the other hand, formulations 2.2 and 3.4 showed notable improvements of 13.8% and 18.3%, respectively, compared to their non-admixtured equivalents.

In conclusion, the aggregate content plays a significant role in flexural strength; the lower the aggregate proportion is, the higher the flexural strength is, assuming the other factors remain constant. Similarly, a higher cement content consistently correlates with an improved performance. It is also important to note that like in the compressive strength tests, relative humidity and temperature were not controlled during testing. These environmental conditions can have a significant impact on the mechanical properties of mortar and should be considered in future studies.

The eight main mortar formulations were tested for capillary water absorption, with the results presented in

Table 6. Capillary water absorption test results.

SP?	Mixture		Absorption Coefficient, Aw, 24h (kg/m2s0.5)	Mean Absorption Coefficient, Aw, 24h (kg/m2s0.5)	BS EN 998-1 Classification [36]
No	2.3	1	0.00099	0.00109 ± 0.00007	W2
		2	0.00118
		3	0.00109
	3.1	1	0.00050	0.00072 ± 0.00021	W2
		2	0.00065
		3	0.00100
	5.2	1	0.00111	0.00139 ± 0.00028	W2
		2	0.00778
		3	0.00167
	6.3	1	0.00112	0.00117 ± 0.00015	W2
		2	0.00101
		3	0.00137
Yes	2.2	1	0.00043	0.00047 ± 0.00003	W2
		2	0.00048
		3	0.00050
	3.4	1	0.00045	0.00039 ± 0.00010	W2
		2	0.00025
		3	0.00048
	5.5	1	0.00069	0.00063 ± 0.00004	W2
		2	0.00061
		3	0.00059
	6.6	1	0.00042	0.00039 ± 0.00003	W2
		2	0.00039
		3	0.00035

.

This test was carried out over a 24 h period to evaluate the water infiltration behaviour of each composition. As previously discussed, the hydric performance of mortars is closely linked to their porosity, which in turn is influenced by several factors, including the binder content, aggregate grading, and the w/b ratio.

Among the tested mixtures without SP, formulation 3.1 stood out with the lowest capillary absorption coefficient, A_w = 0.00072 kg/m²·s^0.5, indicating a denser and less-porous matrix than the other mixes. This superior performance is likely attributed to its higher cement content, which contributes to a more cohesive and compact structure, as well as its low w/b ratio (0.371).

Formulation 2.3 also demonstrated good resistance to capillary absorption, with an average A_w of 0.00109 kg/m²·s^0.5. While its w/b ratio (0.386) was not the lowest, it remained within the range considered acceptable for reduced porosity and controlled absorption.

Formulation 6.3, which is compositionally similar to 3.1, showed intermediate behaviour with an A_w value of 0.00117 kg/m²·s^0.5, consistent with its w/b ratio of 0.383. The slightly higher absorption value compared to 3.1 may be explained by its higher aggregate content, which can increase pore connectivity and reduce matrix density.

In contrast, formulation 5.2 recorded the highest average capillary absorption coefficient, 0.00139 kg/m²·s^0.5, with noticeable variation between the specimens. One specimen (sample 2) showed an unusually high value of 0.00778 kg/m²·s^0.5, likely due to operational issues (such as sealing of specimens) and possible material uniformity variations. Additionally, the high w/b ratio (0.400) of this mix may have contributed to increased porosity and water uptake.

The reduction in water content with SP led to a corresponding decrease in w/b ratio, directly affecting matrix porosity, and consequently capillary water absorption.

As shown in

Table 7. Capillary water absorption coefficients’ comparison.

Mixture	Coefficient Without SP (kg/m2s0.5)	Mixture	Coefficient with SP (kg/m2s0.5)	Coefficient Variation	Reduction (%)
2.3	0.00109	2.2	0.00047	–0.00062	56.88
3.1	0.00072	3.4	0.00039	–0.00033	45.83
5.2	0.00139	5.5	0.00063	–0.00076	54.68
6.3	0.00117	6.6	0.00039	–0.00078	66.67

, all the mixtures with SP exhibited lower absorption coefficients compared to their non-admixtured counterparts, with reductions ranging from 45.83% to 66.67%. These values are consistent with expectations for mixtures with w/b ratios between 0.29 and 0.34, confirming that lower water content results in reduced porosity and water absorption, as also reported by Van der Putten et al. [37].

Among the eight tested formulations, 3.4 and 6.6 achieved the lowest absorption coefficient, both registering 0.00039 kg/m²·s^0.5.

By lowering the mortar’s ability to take in and hold water, the chances of ice forming inside during freezing cycles are reduced, which helps to limit microcracking and the loss of cohesion. In the same way, with less water being absorbed, fewer salts are carried into the pores, reducing the risk of crystallisation pressures that can lead to scaling, flaking, or surface loss in historic elements.

4. Discussion of Test Results

The study by Soares et al. [38] provides a comprehensive overview of mortar formulations reported in the literature, grouping them according to key parameters, many of which align with those presented in

Table 8. Compilation of best mortar’s fresh state test results.

SP?	Mixture	w/b	a/b	Flow (mm)	Slump (mm)
No	1.1	0.400	0.429	206.00	16
	2.3	0.386	0.429	207.00	13
	3.1	0.371	0.429	205.00	10
	4.2	0.417	0.667	207.00	10
	5.2	0.400	0.667	212.50	14
	6.3	0.383	0.667	218.00	13
Yes	2.2	0.314	0.429	201.00	13
	3.4	0.293	0.429	201.00	15
	5.5	0.338	0.667	216.00	15
	6.6	0.317	0.667	220.00	19
Without SP		0.371–0.417	0.429/0.667	205–218	10–16
With SP		0.293–0.338	0.429/0.667	201–220	13–19

of this work.

The w/b ratios reported closely reflect those obtained in this study. Specifically, 24% of the mortars reviewed had w/b ratios between 0.36 and 0.41, 18.7% between 0.31 and 0.36, and 13.3% between 0.26 and 0.31. Altogether, 56% of the mixtures fall within the 0.26 to 0.41 range, confirming that the values adopted in this work are representative of common practice.

In contrast, regarding the a/b ratio, only 12.35% of the mortars in their dataset fell within the range from 0.40 to 0.65, and just 7.4% ranged between 0.65 and 0.90. The majority (27.16%) featured a/b ratios between 0.9 and 1.15, values that are generally incompatible with the printer used in the present study due to limitations related to extruder size and nozzle diameter.

With respect to aggregate size, this study found that 37% of mortars used aggregates with a maximum particle size of 1 mm, more than any other category. This preference is largely due to equipment constraints, particularly the dimensions of the extruder and nozzle, and is consistent with the choices made in the current research.

In terms of flow behaviour, the most frequently reported values in the literature ranged between 142 and 202 mm, accounting for 44% of the mortars. Only 15.63% exhibited higher flow values, between 202 and 222 mm. Although increased flow generally improves extrudability, it can also lead to shrinkage and deformation between layers.

Table 9. Literature review results for hardened state tests.

Laboratory Test	Mortar Type	Result	Unit	Reference
Compressive strength	Traditional mortar *	4.16–7.91	MPa	Zagaroli et al. [32]
	3DP mortar with fibres	48–52		Li et al. [33]
	3DP concrete	23.50		Joh et al. [39]
	3DP cement mortar	30.40		Jesus et al. [1]
	3DP cement mortar	55–87		Matos et al. [40]
	3DP mortar	48.90–64.20		Present study
	3DP mortar with SP	52.70–74.00		Present study
Flexural strength	Traditional mortar *	1.23–2.21	MPa	Zagaroli et al. [32]
	3DP mortar with fibres	6.00–8.30		Li et al. [33]
	3DP concrete	6.50		Joh et al. [39]
	3DP cement mortar	5.30		Jesus et al. [1]
	3DP cement mortar	4.00–7.00		Matos et al. [40]
	3DP mortar	6.40–7.40		Present study
	3DP mortar with SP	6.40–8.40		Present study
Water absorption by capillarity	Masonry mortars	0.05–0.10	kg/m2s0.5	Yedra et al. [41]
	Cement mortar	0.024		Ka et al. [42]
	3DP mortar	0.00071		Pessoa et al. [43]
	Traditional coating mortar	0.13–0.14		Borges et al. [44]
	3DP mortar with SP	0.00039–0.00063		Present study
	3DP mortar	0.00072–0.00139		Present study

* Traditional mortar—cement, lime, and sand.

gathers values from different studies found during the literature review stage. The results show a clear evolution in performance between traditional mortars and those developed for 3DP.

Compressive strength of the developed 3DP mortars ranged from 48 to 64 MPa and increased to 52–74 MPa with SP addition. These values far exceed those of traditional mortars (4.16–7.91 MPa) and are comparable to or better than most 3DP mixes reported in the literature, confirming the effectiveness of the lime–cement formulations for high-performance, heritage-compatible printing.

Flexural strength values of the developed mortars ranged from 6.40 to 8.40 MPa, exceeding those of traditional mortars (1.23–2.21 MPa) and matching or surpassing unreinforced 3DP mixes in the literature. The results highlight strong interlayer adhesion and good tensile performance, even without fibre reinforcement.

Water absorption by capillarity was significantly lower than that of conventional mortars. The values ranged from 0.00072 to 0.00139 kg/m²·s^0.5, dropping to 0.00039 kg/m²·s^0.5 with SP, indicating low-level porosity and high durability, superior to most reported 3DP and traditional formulations. These comparisons reinforce that the mortars developed in this study exhibit less porosity and denser microstructures, making them particularly suitable for applications requiring reduced water ingress.

While the improved compressive and flexural strengths can enhance durability and resistance to mechanical stresses, it also raises concerns regarding compatibility with original substrates, since new repair mortars may induce stress concentrations and potential detachment from weaker historic materials. In conservation practice, repair mortars are expected to remain mechanically compatible with the original, to avoid long-term damage, as recommended by ICOMOS. Additionally, the reduced water absorption by capillarity observed in the optimised formulations represents an important improvement, since it minimises risks of freeze–thaw damage and salt crystallisation, as better described above.

5. Conclusions

This study investigated the development and optimisation of lime–cement mortars for Delta WASP 40100, particularly in the context of heritage rehabilitation. The experimental results led to several key conclusions.

All the tested mortars in the hardened state demonstrated compressive and flexural strength values consistent with requirements for non-structural 3DP applications, such as façade components and ornamental elements. The formulations with a higher cement content, especially those with 75% cement and 25% lime, consistently showed an improved mechanical performance and surface quality when compared to formulations with more lime content. The ones with a 50/50 lime-to-cement ratio exhibited poor extrudability and buildability and were excluded from further testing. Flexural and compressive strength tests on moulded samples highlighted the impact of binder composition, the aggregate ratio, and admixture use.

The incorporation of a SP effectively reduced the w/b ratio, improving flowability without compromising strength. However, it also led to a reduced open time, meaning the mortar stiffens more quickly, shortening the workable window during which it can be extruded. This can limit the maximum size of printable elements and increased the risk of nozzle clogging, interlayer discontinuities, and surface defects. Thus, in real-world contexts, this introduces logistical challenges, as smaller batches might need to be prepared and loaded more frequently. To overcome this, adjustments to printing speed or path planning are necessary, as well as optimising SP dosage or combining it with setting retarders.

Nevertheless, eight formulations demonstrated overall suitability for 3DP, balancing workability, buildability, extrusion stability, surface finish, and mechanical strength. Mortars with a lower aggregate content (30%) generally produced better finishes and cohesion.

The optimised mortars showed w/b ratios between 0.29–0.33 (with SP) and 0.37–0.40 (without SP), aligning with values reported in the literature. The capillary absorption tests confirmed that a higher cement content and lower w/b ratios led to reduced porosity, and therefore reduced water absorption.

This study established an initial framework for optimising lime-cement mortars for 3DP applications in built heritage rehabilitation. The experimental program primarily focused on fresh-state properties, early hardened-state performance, and capillary water absorption as durability indicators. The results demonstrated that the mortars with a higher lime content achieve greater visual and textural resemblance to traditional materials, but have poor results in buildability. On the other hand, the cement-rich mixtures performed better in terms of extrusion stability and mechanical strength, though with less aesthetic compatibility.

Future research should pursue targeted solutions to tackle buildability issues. The potential strategies include incorporating nanoclays and viscosity-modifying additives (VMA) to improve cohesion, dimensional stability, and resistance to segregation, as well as applying setting accelerators immediately before extrusion to enhance shape retention [45]. Further optimisation of mixture design should also consider compatibility with different printing systems, as these mortars were developed for Delta WASP 40100.

Beyond mechanical optimisation, broader compatibility requirements must be addressed. Comprehensive testing should include thermal expansion, porosity coordination, freeze–thaw resistance, and salt crystallisation effects to ensure durability under real environmental conditions. Moreover, the compatibility of these mortars with various historic substrates, such as stone masonry, brickwork, and mixed-material systems, needs to be evaluated to extend applicability across diverse building contexts.

Equally important are aesthetic and visual aspects. Dedicated assessments of colour, texture, and finish will be necessary to ensure that printed components are compatible with heritage buildings. While visual fidelity may be less critical for certain applications, such as corbels placed at considerable heights, surface roughness caused by the layered printing process remains a limitation. Post-processing techniques, like sanding and plastering, can mitigate these issues, improving definition and detail reproduction [18].

In this first stage, surface roughness (Ra) was not directly measured, as this study focused mainly on fresh-state behaviour, mechanical performance, and water absorption. Nonetheless, Ra is an important parameter for heritage rehabilitation, since it influences the adhesion of repair mortars and coatings, as well as the visual integration of printed elements with historic substrates. Future work should therefore incorporate systematic roughness characterisation to better understand its role in durability, compatibility, and conservation outcomes.

Looking ahead, full-scale applications, such as the replication of the theatre’s corbel, will provide a practical testbed for the developed mortars, enabling validation under semi-real environmental conditions. Microstructural characterisation of cohesion and interlayer adhesion will also be essential for understanding failure mechanisms and long-term performance. Expanding durability testing, including accelerated aging and chemical interaction studies, will further confirm the suitability of 3D-printed mortars for respectful conservation practices.

At last, while this study focused primarily on the fresh and hardened performances of lime–cement mortars for 3DP in heritage rehabilitation, their environmental performance should be also addressed. Recent life cycle assessment (LCA) studies highlight the crucial role of cement content in the environmental impact of 3DP, with cement quantity being more decisive than cement type, as variations between CEM I 32.5, 42.5, and 52.5 were limited to about ±2% [46,47]. The adoption of supplementary cementitious materials, such as limestone calcined clay, has shown substantial benefits, reducing greenhouse gas emissions by 41.6–50.2% and energy consumption by 38.2–45.2% [48]. Similarly, the use of recycled aggregates in place of natural aggregates has been demonstrated to further reduce impacts, including global warming (−2.5%), acidification (−5.9%), photochemical pollution (−37.5%), and eutrophication (−13.0%) [47]. These findings reinforce the potential of this technology to integrate sustainable practices through material selection and optimisation, encouraging future research in this direction [49,50].