The Influence of Mobility Parameters on the Rheological Behaviour and Mechanical Properties of Low-Carbon Mortar Mixtures

Abstract

Environmental targets towards net-zero carbon concrete are increasing the demand for eco-efficiency in concrete production. Promising measures to increase sustainability include the combination of high levels of limestone fillers (LFs) and the use of advanced mix-design techniques, such as particle packing models (PPMs). However, there is still a limited understanding of the fresh and hardened state properties of eco-efficient mixtures; the literature suggests that mobility parameters (MPs; interparticle separation distance—IPS; maximum paste thickness—MPT) can help explain the fresh behaviour of concrete mixtures. Yet, the impact of MP values on fresh properties is still not fully understood. To address this gap, this study evaluates a reduced-complexity system comprising twelve concrete mortar fractions developed with distinct MP ranges and high LF contents (up to 52%). The use of mortar mixtures was intended to reduce the number of variables in the system and provide a clearer assessment of the role of mobility parameters. Time-dependent rheological behaviour (flow behaviour factor, torque, and viscosity) is analyzed and correlated with MP ranges to identify governing fresh state mechanisms. In addition, the relationships of IPS and MPT with compressive strength and porosity are evaluated to examine their relevance to the hardened state behaviour of low-carbon mixtures with reduced cement content. Results indicate that MPT and IPS can be used as practical indicators of rheological behaviour, with MPT showing the strongest influence on rheological response across all mixtures. Based on compressive strength and porosity measurements, empirical models are proposed to describe the effect of mobility parameter-based spacing concepts on hardened properties. Finally, the environmental performance of the optimized mixtures is assessed, confirming the potential of LF-rich, MP-tailored mixtures to contribute to low-carbon, net-zero concrete production.

1. Introduction

Net-zero carbon is currently a significant topic in the concrete industry, as concrete production corresponds to about 8% of anthropogenic carbon dioxide emissions [1,2]. Portland Cement (PC), one of the main components of concrete, is the most significant contributor to the embodied energy of cement-based mixtures, accounting for 92% of their carbon footprint [3]. Amongst potential solutions towards net zero, two distinct approaches have demonstrated efficiency: a) using high volumes of supplementary cementitious materials (SCMs) [4,5,6] and limestone fillers (LFs) [7,8,9] as PC replacements, and b) optimizing concrete mixtures through the use of advanced mix-design techniques (e.g., Particle Packing Models—PPMs) [10,11,12] which can reduce the system’s porosity, increasing its inner quality. While the use of SCMs is largely studied and applied, the increasing demand for concrete outpaces the availability of SCMs [7,13]. Alternative binder systems such as alkali-activated materials have also been investigated as low-carbon solutions [14,15]; however, their widespread adoption remains limited by raw material availability and processing requirements, reinforcing the relevance of LF-based approaches using widely available materials. Conversely, LF has shown to be one of the most promising types of inert filler due to its proximity to concrete plants and abundance as a by-product of limestone aggregate production [8,9]. The use of LF at amounts of 15 wt% has demonstrated a negligible impact on the properties of cement-based mixtures [16]. Nevertheless, increasing filler contents based on direct replacement methods has shown adverse impacts on both fresh and hardened states. When LF optimization is based on PPMs, LF can contribute towards microstructural quality enhancement, known as the “filler effect” [17]. In this utilization, LF can accelerate cement hydration due to the formation of nucleation sites combined with a reduction in the system’s porosity. Conversely, at high LF replacement levels, a physical dilution of the binder occurs (i.e., the proportion of reactive cement decreases and the paste contains more inert solids, limiting hydration product formation), negatively impacting the mechanical properties of the material due to the increased amount of water required. While the use of PPMs have proven to be one of the best methodologies for optimizing limestone replacement through the reduction of the system’s porosity, mixtures containing LF designed with PPMs generally display challenging fresh state performance, and trial-and-error experimentation is often required for improving fresh properties [18,19,20]. Previous studies demonstrated that mobility parameters (e.g., Interparticle Separation Distance—IPS [21] and Maximum Paste Thickness [22]—MPT) present a good correlation with the fresh and hardened state behaviours (i.e., mixing energy, slump, compressive strength) of PPM-designed mixtures containing high amounts of LF [23]; however, the lack of consistency in investigated IPS and MPT ranges in past studies limits the full understanding of the impact of each parameter on distinct fresh and hardened state properties. Further investigations are therefore required to discover ideal ranges of IPS and MPT for cement-based mixtures according to their application. This project aims to investigate the impact of mobility parameters on both the fresh and hardened state characteristics of PPM-proportioned mortars with high limestone filler contents. The use of mortar mixtures was intended to reduce the number of variables in the system, such as interfacial transition zones associated with coarse aggregate, and thus provide a clearer assessment of the role of mobility parameters. In particular, the study evaluates the relationship of IPS and MPT with time-dependent rheological behaviour, compressive strength, and porosity in low-carbon mixtures with reduced cement content.

2. Literature Review

2.1. Proportioning Sustainable Mixtures

While worldwide use of LF represents 20% of overall PC content [24], previous studies [25,26] suggest that LF replacement ratios inferior to 15% have negligible impact on the fresh and hardened state properties of cement-based mixtures. Conversely, when PC is directly replaced by LF at levels greater than 15 wt%, the impact of the LF addition on the fresh (e.g., slump and flowability [27,28]) and hardened properties (e.g., compressive strength [17,29]) is not well understood, highlighting the importance of further evaluating the dilution and filler effects when high levels of LF are added to cement-based mixtures. Nevertheless, a number of studies [30,31] have demonstrated promising fresh and hardened state performance of cement-based materials proportioned via alternative procedures other than direct replacement, and bearing LF contents greater than 15 wt%. Additionally, different gradations of LFs can be used to improve the system’s sustainability (e.g., replacement fillers, RLFs) and mechanical performance (e.g., performance fillers, PLFs). RLFs with a particle size distribution (PSD) similar to PC are used to reduce the cement content of cement-based mixtures [32,33] with a negligible impact on the fresh state [34], whereas PLFs with a smaller PSD than PC are used to reduce the system’s porosity and increase nucleation sites, thereby enhancing cement hydration [17,35,36]. The proper dosage of RLFs and PLFs must be calculated to achieve optimized sustainable mixtures. Previous studies highlighted the use of PPMs as effective proportioning protocols to incorporate optimal amounts of PLF and RLF in cement-based mixtures. This approach enables the production of sustainable mixtures that exhibit notable improvements in both fresh and hardened properties [20,37].

The PPM science is categorized into two primary branches: discrete and continuous. Discrete models represent a finite number of class sizes with a methodology for calculating the system’s packing density [38], whereas continuous models consider an infinite number of class sizes presenting similar conditions across the entire range [21]. Although both categories of PPM aim ultimately to optimize the system packing density (Equation (1)), some authors have suggested that real aggregate blends are better represented by continuous models [10,20,39]:

$$P_d = {\displaystyle \frac{V_s}{V}} ,$$

(1)

where Pd is the packing density, and V and Vs are the total volume and the volume of solids, respectively.

Funk and Dinger [21] developed one of the most recent continuous PPM methods, the modified Andreasen model (also known as Alfred’s model), in 1980, as shown in Equation (2). This method was demonstrated to be effective in proportioning sustainable cement-based mixtures [40]:

$$CPFT = {\displaystyle \frac{D^q - D_s^q}{D_L^q - D_S^q}} ,$$

(2)

where D is the particle size in question, CPFT is the cumulative percent finer than D, D_L and D_S are the largest and smallest particle size in the system, respectively, and q is a distribution factor.

Based on computational analysis, it was determined that Alfred’s model exhibits its highest packing efficiency at a distribution factor of 0.37, yielding a dry porosity of 2.8% based on the model proposed by Westman and Hugill [41]. Consequently, studies have concluded that a q-factor of 0.37 is the optimal parameter for proportioning sustainable mixtures with PC reductions of up to 60% when RLF and PLF are incorporated in the mixture. This optimization maintains the mechanical properties of the material, albeit with a reduction in fresh state performance [10,11]. In contrast, a q-factor of 0.22 is recommended to enhance the flowability of cement-based materials by increasing the proportion of powder content and reducing the amount of coarse aggregates, which applies to Self Consolidating Concrete (SCC) [42,43]. As such, lowering the q-factor can improve the fresh state performance of sustainable mixtures. Müllera et al. [18], for example, evaluated sustainable cement-based mixtures proportioned with a q-factor of 0.34, 111 kg/m³ of cement (on average), and 216 kg/m³ of quartz fillers. Through the addition of quartz fillers and a slightly reduced q-factor (just below 0.37), these mixtures exhibited negligible differences in packing density. However, the fresh state performance displayed clear improvements when compared to pure PC mixtures developed with a q-factor of 0.37. To emphasize the importance of using inert fillers to produce sustainable mixtures, a reduction in PC content of only 15 wt% was achieved when proportioning cement-based mixtures using Alfred’s model and a 0.37 q-factor [10].

Grazia et al. [11] developed sustainable mixtures through the use of Alfred’s model. Yet, the authors decided to divide the entire PSD into two fractions: the powder’s portion (from the smallest diameter—D_S—to the largest diameter available within the powders selected), and the aggregate’s portion (from the smallest diameter of aggregates to the system’s largest diameter—D_L). With the selection of q-factors of 0.21 and 0.37 for the powder and aggregate portions, respectively, a reduction of 41% in PC was achieved with a lower amount of admixtures and compatible fresh state behaviour when compared to other sustainable concrete mix design methods [18,20,37]. In summary, PPMs are primarily used to decrease the system’s porosity and subsequently increase the engineering properties of cement-based materials [20,42,43,44,45]; nonetheless, additional parameters must be combined with PPMs to improve fresh state issues presented in previous works while developing sustainable mixtures incorporating inert fillers.

2.2. Mobility Parameters

In concentrated suspensions such as cement-based materials, upon the addition of water, particle interactions are critical in determining the material’s rheology (i.e., viscosity, yield stress). Generally, the rheological behaviour [18,20] of suspensions is affected by a number of factors, including the solids volume fraction [46,47], specific surface area (SSA) [48], and particle size distribution of solid particles [49]. In this context, Funk et al. [21] and Bonadia et al. [44] developed the concept of mobility parameters, which are parameters enabling the calculation of distances amongst the solid particles in the system. It has been found that mobility parameters, especially in concentrated suspensions, can be applied to assess the fresh state behaviour of cement-based mixtures. In physical terms, these parameters describe the space available for particle movement within the fresh system and therefore help explain the intensity of particle interactions. In this context, IPS represents the average separation distance between the fine particles suspended in water, whereas MPT represents the maximum thickness of paste surrounding the coarser particles. Mobility parameters are considered according to the forces acting in the system: (a) IPS [21] addresses the surface and capillary forces of particles smaller than 150 μm (Equation (3)), and (b) MPT [22,44] defines the maximum distance amongst particles larger than 150 μm where gravitational forces are critical (Equation (4)).

$$IPS={\displaystyle \frac{2}{VSA}}\left({\displaystyle \frac{1}{V_s}}-{\displaystyle \frac{1}{{1-P}_{of}}} \right),$$

(3)

where IPS is the interparticle separation distance of the fine fraction (μm); VSA is the volumetric surface area of particles with PSD < 150 μm (m²/cm³), obtained by multiplying the BET-measured specific surface area (SSA, m²/g) by the material’s specific gravity (g/cm³); Vs is the volumetric solid fraction of the fine particle system (dimensionless, 0–1); Pof is the dry predicted porosity of the fine fraction (dimensionless), calculated using the Westman and Hugill algorithm [41].

IPS represents the average separation distance between the fine particles suspended in water and is associated mainly with the action of surface and capillary forces. As the IPS decreases, the fine particles become closer to each other, increasing interparticle interactions and resistance to flow.

$$MPT={\displaystyle \frac{2}{VS{A}_c}}\left({\displaystyle \frac{1}{V_{s,c}}}-{\displaystyle \frac{1}{{1-P}_{of,c}}} \right),$$

(4)

where MPT is the maximum paste thickness surrounding the coarse particles (μm); VSAc is the volumetric surface area of particles with PSD > 150 μm (m²/cm³), calculated from BET-measured SSA and specific gravity; Vs,c is the volumetric solid fraction of the coarse particle system (dimensionless, 0–1), where the paste (water + fine powder) acts as the suspending fluid; Pof,c is the dry predicted porosity of the coarse fraction (dimensionless), calculated using the Westman and Hugill algorithm [41].

MPT represents the maximum thickness of paste surrounding the coarser particles and is associated with the lubrication of the granular skeleton. When the MPT decreases, the paste layer available to separate the coarser particles becomes thinner, which may increase collision, frictional interaction, and flow resistance under shear. In physical terms, mobility parameters describe the space available for particle movement within the fresh system and therefore help explain the intensity of particle interactions. Because IPS and MPT reflect particle spacing and lubrication conditions, they can be directly related to rheological parameters such as torque, viscosity, and shear-thinning behaviour. In addition, these parameters influence the spatial arrangement of the solid phases, which affects porosity development and can help interpret hardened state properties such as compressive strength.

Previous studies observed a good correlation between the mobility parameters, slump [23,50,51], and mixing energy [52]. Rebmann et al. [52] used ranges of IPS from 0.11 to 0.16 μm and MPT from 4 to 10 μm to relate the torque required to achieve a given shear rate. Grazia et al. [11] selected ranges of IPS from 0.36 to 0.81 μm and MPT from 0.26 to 1.58 μm to correlate them with the required mixing. Besides the correlation of MPs with fresh state, a few studies [53,54] used MPs to partially understand the hardened state properties of cement-based mixtures. Innocentini et al. [55] noted that the IPS was the key parameter to appraise concrete permeability, with an increase of 0.05 μm in the IPS resulting in a factor of 10 increase in permeability, while changing the MPT from 55 to 115 μm resulted in negligible additional impedance. Bergmann et al. [56] found a strong correlation (R² of 0.97) between particle separation distances of cementitious grains in cement (IPS cement) and hardened state (i.e., f′c) cement pastes, with IPS cement ranging from 0.27 to 0.90 μm. Despite these advances, critical limitations remain in the existing literature. First, previous studies have investigated IPS and MPT over inconsistent ranges, limiting cross-study comparability and the identification of optimal MP thresholds for specific applications [23,52]. Second, no study has systematically evaluated the time-dependent rheological response—including viscosity, torque, and shear-thinning behaviour—of PPM-proportioned mortar mixtures as a function of controlled MP ranges. Third, the relationship between MPs and hardened state properties such as compressive strength and porosity remains insufficiently explored for mortar systems with high LF contents. These gaps motivate the present investigation.

2.3. Sustainability Evaluation of Cement-Based Mixtures

The environmental impact of cement-based mixtures on CO₂ emissions is primarily attributed to the decarbonization process of limestone during PC production. In order to comprehend and quantitatively assess this impact, a parameter known as Global Warming Potential (GWP) can be computed using Equation (5). GWP evaluates greenhouse gas contributions across the entire lifecycle of a material, and quantifies these contributions in terms of mass equivalent carbon dioxide (CO₂) values, accounting for location and availability factors [40,57,58].

The GWP of cement-based mixtures is equal to the total greenhouse emissions (using CO₂ as a baseline) per unit volume of material [59,60].

Table 1. Typical CO₂eq for conventional ingredients of cement-based mixtures (in kg CO₂eq/kg).

Ingredient	Cement	Limestone Filler	Fine Aggregate	Coarse Aggregate	Water	Admixtures
GWP (kg CO2eq/kg of material, gi)	0.9	0.032	0.0029	0.0062	0.00034	0.72
Reference	[57]	[62]	[59,62]	[62]	[59,62]	[62]

can be used as a reference for the carbon dioxide (CO₂) equivalent of each ingredient [59,60,61].

$$GWP={\sum }_{i=1}^n{m}_i·g_i,$$

(5)

where m_i is the mass of ingredient (i) per unit volume of the materials, and g_i is the e-CO₂ per unit mass of the ingredient (i).

Additionally, to assess the efficiency of cement-based mixtures (John et al. [9] and De Brito et al. [60]), the amount of GWP to produce a relative desired property (typically compressive strength, f′c) is called the CO₂ intensity index (ci_cs, calculated using Equation (6)):

$${ci}_{cs}={\displaystyle \frac{GWP}{{f^{\prime }}_c}},$$

(6)

where ci_cs is the CO₂ intensity index (kg/m⁻³·MPa⁻¹), GWP is the global warming potential of the cement-based material (CO₂eq/kg), and f′c is the concrete compressive strength (MPa).

3. Scope of the Work

The primary goal of this study is to investigate the influence of mobility parameters on sustainable mortars developed with PPMs and high levels of LF in both fresh and hardened states. The choice of mortars as the material for investigation aims to reduce the number of variables in the system to better understand the influence of MPs. Figure 1 summarizes the experimental design adopted in this study. First, twelve concrete mixtures were proportioned through Alfred’s model and four distinct levels of cement content (i.e., 324, 250, 200 and 150 kg/m³). The investigation was then conducted on the identical volumetric mortar fractions of these twelve mixtures. Two main areas of investigation were defined: 1) optimization of physical packing without and with limestone fillers at constant water-to-cement ratio (w/c), and 2) comparison between mixtures employing decreasing cement contents (i.e., 250, 200 and 150 kg/m³) at constant slump flow (200 mm via admixture adjustment) and varying mobility parameters (IPS (0.51, 0.46, 0.42 ± 0.1 μm) and MPT (0.37, 0.34, 0.32 ± 0.1 μm)). The selected IPS and MPT ranges were based on previous studies [10,11] and were chosen to represent high, medium, and low mobility conditions within the investigated mortar systems. These ranges were obtained through the selection of compatible free-water contents and were used to enable comparison among mixtures under controlled fresh state conditions. The selected cement contents were chosen to represent different levels of cement reduction, from the reference mixture to lower-carbon concrete with high limestone filler incorporation. The calculated mobility parameter ranges were based on previous studies and were used to obtain different free-water demand levels. The fresh state behaviour was appraised through slump flow and rheological measurements (i.e., yield stress and viscosity). The hardened state behaviour was evaluated via compressive strength, Archimedes porosity, electrical resistivity, and dynamic modulus of elasticity. The time-dependent rheological characteristics of the mortars were then compared to the selected range of mobility parameters, and a new equation was proposed to predict the compressive strength of sustainable mortar mixtures based on the MPs.

4. Materials and Methods

4.1. Characterization of Materials

To produce sustainable mortars, two inert limestone fillers presenting distinct PSDs were used. Using laser diffraction data from the Hydro 2000S, the PSDs of the two fillers (ASTM 1797 Type A) along with the PC (CSA A3000 Type GU) were obtained. Moreover, a natural fine aggregate was selected and classified according to ASTM C33 [63]. The particle size distribution of all materials is shown in Figure 2. The SSA of the cement, limestone fillers, and sand fractions were determined using the Brunauer–Emmett–Teller method (BET) and physically adsorbed N2.

Table 2. Material characterization data.

Material	Specific Gravity (g/cm3)	SSA (m2/g)	VSA (m2/cm3)
Cement	3.17	1.00	3.17
Replacement filler	2.66	0.90	2.39
Performance filler	2.6	3.70	9.62
Fine aggregate	2.74	0.92	2.52

displays the physical characterization (i.e., specific gravity, SSA, and VSA) of the materials used. A lignosulfonic acid salt-based mid-range water reducer (MR) and a plasticizing polycarboxylate-based superplasticizer (SP) were used to improve the fresh state of mixtures. The combination of the two admixtures was used to increase homogeneity without overdosing the mixtures [61,64].

4.2. Mix Design Procedure

The mix-proportioning approach used in this study is based on previous research that implemented two separate curves (i.e., one for powders and one for aggregate particles) using distinct q-factors with the goal of producing sustainable mixtures with suitable fresh state performance [11,65]. The mixture design was established to investigate the influence of mobility parameters under controlled fresh state conditions. In this context, the IPS and MPT were used to compare mixtures with similar initial consistency while varying cement content and limestone filler incorporation. A total of twelve concrete mixtures were designed using the Alfred’s model with q_powders = 0.34 for PSD < 150 μm and q_aggregate = 0.31 for PSD > 150 μm, providing systems with a dry porosity of 3.0 ± 0.1% calculated using the Westman and Hugill model [40]. These q-factors and the separation between powder and aggregate fractions were adopted from the particle packing-based methodology used as the basis of the present study. Additionally, the twelve mixtures were further split and assessed into two phases. Phase 1 evaluated the impact of limestone fillers on the production of eco-efficient concrete by reducing cement content from 324 kg/m³ to 250 kg/m³ with a constant w/c (i.e., 0.60, 0.56, 0.52). The three water-to-cement ratios represent the criteria for producing conventional concrete mixtures ranging from 25 to 30 MPa in compression. In Phase 2, cement content was further substituted by replacement limestone fillers to achieve the targeted eco-efficiency (i.e., concrete with cement contents of 250, 200 and 150 kg/m³) with compatible free water contents (148, 137 and 129 kg/m³). The free water contents were selected to achieve suitable mobility parameters based on previous studies [10,11], resulting in IPSs (0.51, 0.46, 0.42 ± 0.1 μm) and MPTs (0.37, 0.34, 0.32 ± 0.1 μm). Although the amount of replacement fillers varied depending on the target PC content, the performance filler was kept constant at 13%, which was determined via the least squares method to achieve the designed q-factor (0.34) curve. The replacement filler was incorporated mainly to reduce cement content and carbon intensity, whereas the performance filler was used to improve packing density. Therefore, the mix design combined cement reduction with packing optimization to produce low-carbon mortar mixtures with suitable fresh and hardened properties. As per conventional mix-design methodologies, the air content of the mixtures was maintained at a 2% constant for all mixtures, and the admixture contents were added based on the total mass of powders (i.e., cement and fillers). To eliminate the additional interference of coarse particle transition zones, the investigation was based on volumetrically identical mortar fractions of these concrete mixtures. The final mixture designs for the mortar fractions are shown in

Table 3. Mix design for the eco-efficient mortar mixtures evaluated.

	Mix Name	Cement kg/m3	Filler—P kg/m3	Filler—R kg/m3	Sand kg/m3	Water kg/m3	w/c	MR %	SP %	IPS μm	MPT μm
Phase 1	320-0.60H	463	-	-	1501	278	0.60	-	-	0.92	0.54
	324-0.56M	471	-	-	1529	264	0.56	0.08	0.08	0.85	0.48
	328-0.52L	481	-	-	1559	250	0.52	0.20	0.15	0.79	0.45
Phase 1 and 2	250-0.60H	370	60	46	1615	224	0.60	0.20	0.50	0.54	0.40
	250-0.56M	373	61	51	1647	209	0.56	0.40	0.60	0.49	0.38
	250-0.52L	375	62	55	1673	197	0.52	0.90	1.00	0.45	0.36
Phase 2	200-0.74H	297	60	109	1623	220	0.74	0.30	0.50	0.51	0.40
	200-0.69M	299	61	115	1656	205	0.69	0.80	0.40	0.46	0.37
	200-0.64L	300	62	119	1681	193	0.64	1.00	1.00	0.42	0.35
	150-0.97H	222	60	172	1639	214	0.97	0.30	0.50	0.49	0.38
	150-0.89M	223	61	178	1671	199	0.89	0.60	0.70	0.44	0.36
	150-0.84L	224	62	182	1695	188	0.84	1.00	1.00	0.40	0.34

Note: Filler—P and Filler—R refer to performance and replacement limestone fillers, respectively; MR—mid-range water reducer; SP—superplasticizer; interparticle separation distance—IPS; maximum paste thickness—MPT.

and are named according to the following convention: cement content of the overall cement-based material (i.e., 324, 250, 200, 150 kg/m³), w/c, and mobility parameter level (i.e., high, medium, low, represented by H, M, and L, respectively). For example, the mixture 320-0.60H contains 320 kg/m³ of cement, has a w/c of 0.60, and has the highest degree of mobility parameter for the given cement content. With the ranges of IPS and MPT maintained, the initial mortar slump flow of all mixtures was fixed at 200 ± 15 mm. Moreover, the higher the mobility parameters (MPT and IPS), the lower the quantity of admixtures required for the target slump. For example, 320-0.60H required no admixture, whereas 324-0.56M and 328-0.52L required 0.16% and 0.35%, respectively.

4.3. Fabrication and Testing Methods

Twenty litres of mortar were fabricated for each of the twelve mixtures evaluated. Before adding water, limestone fillers were mixed with PC and added to the sand; then, water was introduced and all the materials were mixed in a pan mixer. Following the fresh state tests, twelve 100 × 200 mm cylinders were fabricated for each mixture to evaluate the hardened state behaviour. The cylinders were demoulded after 24 h of fabrication, and the specimens were moist cured at a relative humidity of 95% ± 5% and at a temperature of 22 °C ± 2 °C until testing.

4.4. Fresh State Assessment

Using a thermometer and litmus strips (0.5 pH), the fresh mixture temperature (0.1 °C) and pH were determined at the end of a 10 min mixing period for each mortar mixture. The fresh state properties of the mixtures were evaluated using two methods (i.e., slump flow and rheology). The slump flow was measured using a mortar slump cone (ASTM C1810 [66], 50 mm upper diameter, 100 mm bottom diameter and 150 mm height). The conical sample mould was slowly elevated with a constant speed over 5 s, and the final diameter was measured with the aid of a circular measuring template. Additionally, the rheological appraisal (i.e., yield stress and viscosity) was performed in a rotation speed-controlled IBB planetary rheometer with two rotation radii (i.e., inner rotation radius of 40 mm, outer rotation radius of 68 mm) [67,68,69] and an H-impeller with a diameter of 132 mm and a height of 92 mm. A modified mortar bowl with a 233 mm height and 262 mm diameter (Figure 3a) was fabricated with the width of evaluated mortar flow comparable to that of the original cement-based mixture relative to the maximum coarse aggregate size [69]. Although the IBB output values (i.e., G—torque in N·, and H—viscosity in N·m·s) are not presented in fundamental units, previous studies [69,70,71] have noted that they correlate linearly to the fundamental units (e.g., τ—shear stress in Pa, and η—viscosity in Pa·s). Therefore, in the present study, torque was used as a comparative rheological parameter to describe the resistance of the mixtures to flow under the applied shear conditions. The torque measured at a low shear rate was used to represent the initial resistance to movement, whereas the torque measured at a high shear rate was used to compare the flow resistance of the mixtures under more developed shearing conditions.

In addition to the yield stress (i.e., required torque to initiate flow), the rheological profile data were evaluated with a two-cycle process to evaluate the material’s viscosity change as a function of torque [72]: first, an increase in shear rate up to approximately 0.7 s⁻¹, followed by a decrease period at the same, stepwise (Figure 3b). The resulting curves represent the comparative change in shear stress with respect to shear rate, and the evaluation of viscosity is performed. The rheological parameters were evaluated from the second descending cycle because, at this stage, the mixtures were in a more comparable sheared condition, allowing a more consistent assessment of yield-related torque and viscosity among the investigated mortars [73].

After the initial consistency and rheological profile was assessed, a time-dependent evaluation was performed. The yield stress (related to slump flow) was appraised again at 15, 30, 60, 90 and 120 min, and Equation (7) was used to compare the slump flow loss in a relative fashion. Additionally, the rheological parameters of the mixtures (i.e., yield stress and viscosity from rheometer) were subjected to the same tests after 15 and 30 min.

$$Slump Loss={\displaystyle \frac{{(D}_l-{ d}_t)}{\left(D_l-b\right)}}\% ,$$

(7)

where DI is the initial mortar slump flow, dt is the mortar slump flow measured at time t, and b is the base diameter of the cone (100 mm).

4.5. Hardened State Evaluation

The surface electrical resistivity and the ultrasonic pulse velocity (UPV) of three cylinders were tested at 28 days, while the compressive strength was tested at 7, 14 and 28 days as per ASTM C39 [74]. The surface electrical resistivity (ER) was measured using a device based on the Wenner probe as per AASHTO T358-15 [75]. Four probes located in a straight line (and equally spaced) apply a current from the outer probes to the mortar sample surface, and the potential drop is measured by the interior probes.

The UPV was measured as per ASTM C597 [76]. Equation (8) relates the speed of the pulse through the hardened mortar sample to the dynamic modulus of elasticity (E):

$$E = {\left({\displaystyle \frac{t}{L}}\right)}^2\rho \ast {\displaystyle \frac{\left(1 + \mu \right)\left(1 - 2\mu \right)}{\left(1 - \mu \right)}},$$

(8)

where t is the ultrasonic pulse travel time through the length of the cylinder, L is the length of the cylinder, ρ is the density of the mortar, and μ is the dynamic Poisson’s ratio (0.2 mm/mm).

The apparent porosity (AP) was obtained using Archimedes’ method [77] at 28 days. At each age of analysis, a cylinder (100 mm diameter × 200 mm height) was cut into three sections. The sections were dried at 60 °C until they reached a constant mass (m_d). The samples were then submerged (in water) in a vacuum chamber. After 24 h, the samples were removed from the vacuum chamber and the immersed mass (m_i) and the wet mass (m_w) were measured. Equation (9) was used to calculate the apparent porosity of each mixture:

$$AP \left(\%\right) = \left({\displaystyle \frac{m_w - m_i}{m_w - m_d}}\right) \ast 100\%,$$

(9)

where mw is the wet mass, mi is the immersed mass, and md is the dry mass.

5. Results

5.1. Slump, Slump Loss, pH and Fresh Density

Fresh state results, including pH, temperature, fresh density, and air content are presented in

Table 4. Fresh state properties of the various mixtures.

	Mix Name	pH	Fresh Density (kg/m3)	Batch Temp (°C)	MR (%)	SP (%)	Mortar Slump Flow (mm)
Phase 1	320-0.60H	12.5	2262	22.1	-	-	208
	324-0.56M	12.5	2263	21.7	0.08	0.08	185
	328-0.52L	13.0	2260	23.2	0.20	0.15	195
Phase 1and 2	250-0.60H	12.5	2292	20.5	0.20	0.50	195
	250-0.56M	12.5	2317	20.8	0.40	0.60	187
	250-0.52L	12.5	2320	21.2	0.90	1.00	197
Phase 2	200-0.74H	12.5	2286	20.9	0.30	0.50	190
	200-0.69M	12.5	2319	20.7	0.80	0.40	197
	200-0.64L	12.5	2330	22.5	1.00	1.00	190
	150-0.97H	12.5	2258	23.7	0.30	0.50	192
	150-0.89M	12.0	2283	20.8	0.60	0.70	198
	150-0.84L	12.0	2317	21.7	1.00	1.00	185

. As temperature may affect most of the fresh state properties, the batch temperature was measured and was within 22.1 ± 1.6 °C for all mixtures appraised, regardless of the cement content selected. Moreover, most mixtures (nine out of 12) presented a pH of 12.5. Only 328-0.52L displayed a 0.5 increase in pH, and the two mixtures with the lowest cement content and w/c (150-0.97H and 150-0.84M) yielded a pH of 12. These results demonstrate that the mixtures developed with a 38% reduction in cement content (i.e., 200 kg/m³ of cement) yielded similar alkalinity to those of the control mixtures developed with approximately 325 kg/m³ of cement. Regarding the fresh density, the change in cement content did not affect this parameter significantly, as all the mixtures appraised ranged between 2258 kg/m³ and 2330 kg/m³, which are expected density values.

Although mixtures developed with 250 kg/m³ presented the same w/c as the control mixtures (~325 kg/m³), a greater amount of admixture (1% on average) was required to achieve the target slump due to their lower mobility parameters. Furthermore, even though the amount of LF content changed within Phase 2 mixtures, a similar dosage of admixtures was required for mixtures developed with similar mobility parameter grouping to achieve the target slump flow.

Figure 4a,b present the slump loss of Phase 1 mixtures (324 kg/m³ and 250 kg/m³ of cement) and Phase 2 mixtures (250 kg/m³, 200 kg/m³, and 150 kg/m³). Figure 4a shows that mixtures with 324 kg/m³ displayed lower slump loss than those with 250 kg/m³ after 15 min. Additionally, 320-0.60H and 324-0.56M achieved 100% slump loss only after 120 min, whereas 328-0.52L and all 250 kg/m³ mixtures reached 100% slump loss after 60 min. In the mixtures with lower cement contents (Figure 4b), all mixtures lost 100% consistency by 60 min, while 250-0.56M and 150-0.84L showed complete slump loss at 15 min. Furthermore, mixtures with higher mobility parameters (black lines) demonstrated the lowest slump flow loss over time (average 64%). Conversely, in 30 min, mixtures with middle mobility parameters (250-0.56M, 200-0.69M, 150-0.89M) displayed the greatest slump flow loss (average 93%), while low mobility parameter mixtures averaged 84%. Overall, decreasing cement content (250–150 kg/m³) did not significantly affect slump flow loss at 30 min, with variability remaining below 20% across cement levels.

5.2. Rheological Behaviour

The rheological profiles, based on the data points from second descending cycles, are presented in Figure 5a and Figure 5b for Phase 1 and Phase 2 mixtures, respectively. Mixtures without LF (cement content 320 kg/m³) presented an almost linear relationship between the rotation rate and the torque. In contrast, non-linear behaviour (e.g., decreasing shear stress at higher shear rates) became evident with increasing eco-efficiency (i.e., mixtures with cement content of 250–150 kg/m³). To facilitate the comparison, Figure 5c,d present torque values at low and high shear rates for all investigated mixtures. As the initial slump flow was kept constant, a similar yield stress was observed across mixtures, ranging from 1.3 N·m to 2.4 N·m (Figure 5c). However, at a higher shear rate regime (0.7 s⁻¹), the mixtures exhibited distinct final torques (Figure 5d). Moreover, although the same w/c was used, mixtures with 324 kg/m³ of cement displayed lower variability in final torque (3.8–7.4 N·m) compared to mixtures with 250 kg/m³ (9.1–17.5 N·m). In Phase 2, the final torque generally increased with the mobility parameter level (i.e., H, M, L). Moreover, mixtures with higher cement content (250 kg/m³) yielded lower final torque than those with 200 and 150 kg/m³, with the exception of mixture 250-0.56M (15.2 N·m), which exceeded 200-0.69M (13.1 N·m).

The initial torque and final torque measured at 0, 15, and 30 min are presented in Figure 6. Mixtures without LF displayed a marginal increase (on average 0.3 N·m) in initial torque between 0, 15 and 30 min. Although mixtures with 250 kg/m³ of cement were prepared with the same w/c as those with 324 kg/m³, 250-0.56M and 250-0.60H exhibited substantial increases in initial torque after 30 min (56% and 106%, reaching 3.6 and 3.0 N·m, respectively). In Phase 2, mixtures with the highest cement content (250 kg/m³) showed the greatest increase in initial torque at 15 and 30 min. Although Figure 6a indicates a clear rise in initial torque at higher shear rates (0.7 s⁻¹), Figure 6b shows only marginal changes after 30 min. Similar to the trends observed for initial torque, control mixtures (320 kg/m³) exhibited slight increases of 0.5 N·m at high shear rates. Nevertheless, when comparing the final torque after 30 min for mixtures with the same w/c, those with 250 kg/m³ resulted in more than twice the torque of mixtures without LF. In Phase 2, eco-efficient mixtures (250, 200, and 150 kg/m³) did not exhibit a consistent overall trend; however, within each cement content, mixtures with lower mobility parameter levels (H → M → L) generally showed lower final torque values, except for mixture 150-0.89M at 30 min.

5.3. Compressive Strength

Figure 7 displays the compressive strength of the mortar mixtures evaluated. In general, all mixtures achieved 82% (SD 6%) and 91% (SD 3%) of the final compressive strength after 7 and 14 days, respectively. After 28 days, mixtures with 250 kg/m³ of cement (containing 22% less cement than the 320 kg/m³ control mixture at the same w/c), reached, on average, 4.3 MPa (12%) higher strength than the control, demonstrating the promising ability to produce eco-efficient mixtures. This physical enhancement is further illustrated by the comparison between 328-0.52L and 250-0.56M, which achieved similar 28-day strengths (~40 MPa) despite the increase in w/c (from 0.52 to 0.56) and reduction in cement content (from 328 to 250 kg/m³). For mixtures with 200 and 150 kg/m³ of cement, the 28-day strength ranged from 27.9 to 37.0 MPa and 13.4 to 24.1 MPa, respectively; these values were significantly lower than those of the 324 kg/m³ and 250 kg/m³ mixtures due to the higher w/c (from 0.74 to 0.64 and from 0.97 to 0.84, respectively). Moreover, the eco-efficiency potential of Phase 2 mixtures is evident: 250-0.6H and 200-0.64L yielded similar compressive strengths (~37 MPa) despite a 16% reduction in cement content and a 0.04 increase in w/c. The strength enhancement observed in mixtures with 250 kg/m³ of cement relative to the control is attributed to the combined filler and packing effects of LF: the performance filler reduces system porosity by filling interstitial voids, while the replacement filler maintains the packing density while reducing cement content. At higher LF dosages (200 and 150 kg/m³), the dilution effect becomes dominant—the physical distancing of cement particles reduces hydration product formation, leading to lower compressive strength despite maintained packing density. This strength–dilution trade-off is consistent with findings reported in the literature for PPM-designed mixtures with high LF contents [33,56].

5.4. Capillary Porosity, Surface Electrical Resistivity, and Modulus of Elasticity

To further understand the microstructure of the mixtures studied, capillary porosity and surface ER tests were performed (Figure 8). Since porosity is directly proportional to paste microstructural quality, similar trends were observed between porosity and ER, where lower porosity corresponded to higher ER. For instance, 328-0.52L yielded a porosity of 10.5% and an ER of 5.1 kΩ·cm, while 320-0.60H achieved lower porosity (9.2%) and higher ER (7.8 kΩ·cm). Demonstrating the efficiency of the eco-friendly mixtures, 250-0.52M and 250-0.60H achieved, on average, 18% lower porosity and 14.5% higher ER than mixtures with 324 kg/m³ at the same w/c, due to their enhanced microstructure. The only exception was observed in the comparison of low mobility parameter mixtures (328-0.52L and 250-0.52L), which exhibited similar porosity and ER. In Phase 2, as expected due to the higher w/c of mixtures with low cement content, porosity increased and ER decreased as a function of w/c.

Similar to ER, the dynamic modulus of elasticity (MoE) results from Phase 1 (320 and 250 kg/m³) followed the same trend, where lower porosity corresponded to higher MoE. Mixtures with the same w/c (cement content from 328 to 250 kg/m³) presented comparable MoE values, ranging from 35 to 38 GPa. In Phase 2, the MoE of mixtures across the H, M and L mobility parameter groups appeared to increase (34.9, 38.0 and 39.6 GPa, respectively) independent of the microstructure. As such, for these mixtures with higher LF, MoE showed a direct relationship with w/p rather than w/c, where higher w/p resulted in lower MoE. The porosity reduction observed in 250 kg/m³ mixtures relative to the control is attributed to the pore refinement effect of the performance limestone filler, whose smaller PSD fills capillary pores and reduces the system’s connected porosity. This microstructural improvement is reflected in the higher ER values, since electrical resistivity is strongly governed by pore connectivity and pore solution continuity [56]. In Phase 2, the progressive increase in w/c associated with higher LF dosages counteracts the pore refinement effect, resulting in higher capillary porosity and increased pore connectivity, which reduces the electrical resistivity of the system.

6. Discussion

6.1. Rheological Models for Describing Mixture Behaviour

Cement-based materials are usually described using the Bingham model (Equation (10)), where a linear relationship exists between the shear stress and shear rate [78]. Previous studies [27,78,79] show that eco-efficient cement-based materials developed with PPMs often exhibit non-linear shear behaviour, which is better represented by the Modified Bingham model (Equation (11)) [80] and/or the Herschel–Bulkley model (Equation (12)) [56], as summarized in

Table 5. Compared rheological models.

Name	Equation	Equation Reference
Bingham	$\tau ={\tau }_0+k_B\dot{\gamma }$	(10)
Modified Bingham	$\tau ={\tau }_0+{\mu }_p\dot{\gamma }+c{\dot{\gamma }}^2$	(11)
Herschel–Bulkley	$\tau ={\tau }_0+k_{HB}{\dot{\gamma }}^n$	(12)

. It is worth highlighting that the IBB rheometer outputs are not in fundamental units; hence, for comparison purposes in this paper, τ, τ₀, and $\dot{\gamma }$ are reported as torque, initial torque, and rotation rate units, respectively.

Here, τ is the shear stress, τ₀ is the yield stress, k_B is the viscosity constant of Bingham and $\dot{\gamma }$ is the shear rate, μ_p is the modified Bingham viscosity, c is the modified Bingham constant, k_HB is the viscosity constant of the Herschel–Bulkley model, and n is flow behaviour factor (n < 1 for suspensions presenting shear-thinning behaviour and n > 1 for shear thickening suspensions) [79,81].

Figure 9a,b present the three rheological models compared to experimental data of mixtures developed with cement contents of 320 and 250 kg/m³, respectively. Similar trends were observed for mixtures with cement contents of 200 and 150 kg/m³; however, their graphs are not presented here due to space limitations. The suitability of the rheological models was evaluated by comparing their fitting errors, expressed in terms of minimum mean squared error (MMSE), for all investigated mixtures (

Table 6. Calculated parameters and minimum mean squared errors (MMSEs) for the rheological models at 0 s.

	Mix Name	τ0	kB	MMSE	τ0	μp	c	MMSE	τ0	n	kHB	MMSE
		N·m	N·m·s		N·m	N·m·s			N·m		N·m·s
Phase 1	320-0.60H	1.7	3.1	0.49	1.9	2.7	0.0	0.12	1.9	2.7	1.0	0.12
	324-0.56M	2.4	4.7	0.85	2.8	4.0	0.0	0.11	2.9	4.1	1.1	0.09
	328-0.52L	3.0	6.4	1.26	3.5	5.8	0.5	0.11	3.4	5.4	0.9	0.09
Phase 1 and 2	250-0.60H	2.7	8.9	1.04	2.4	11.5	3.5	0.70	3.2	8.2	1.0	0.25
	250-0.56M	4.3	16.2	2.31	3.3	25.2	12.6	1.82	5.0	14.2	0.9	0.31
	250-0.52L	3.2	21.7	2.37	2.1	33.2	16.6	1.36	2.6	19.7	0.8	0.30
Phase 2	200-0.74H	2.7	11.7	1.35	2.1	16.9	7.2	1.08	3.1	10.6	1.0	0.32
	200-0.69M	2.7	15.3	1.50	2.0	21.9	9.3	1.02	2.4	14.1	0.8	0.44
	200-0.64L	3.3	22.9	2.56	2.1	36.0	18.9	1.20	1.8	21.2	0.7	0.22
	150-0.97H	3.6	13.1	2.57	2.3	25.0	17.0	1.65	2.2	11.8	0.6	0.53
	150-0.89M	3.4	24.7	2.78	2.0	38.5	20.0	1.49	2.3	22.5	0.7	0.24
	150-0.84L	3.9	26.5	3.18	2.9	38.5	18.0	1.64	3.4	23.8	0.8	0.52

). As shown, the Herschel–Bulkley model provides the best fit for the rheological behaviour of mortar mixtures developed with PPMs and high amounts of LF. The rheological behaviour of the twelve mixtures appraised at 0 min resulted in minimum mean squared errors (MMSEs) of 1.86, 0.40 and 0.29 for Bingham, Modified Bingham, and Herschel–Bulkley, respectively.

As expected, the Bingham model presented the highest MMSE, as the mixtures display a non-linear shear stress shear rate trend. Although both non-linear models are able to predict the observed behaviour with low average MMSEs, analysis of individual mixtures showed that the Herschel–Bulkley (HB) model consistently yielded lower MMSE values than the Modified Bingham model. Mixtures with an average cement content of 324 kg/m³ exhibited an average MMSE of 0.11 using the Herschel–Bulkley model, compared with 0.87 for the Modified Bingham and Bingham models. In contrast, Phase 2 mixtures, with reduced cement contents, showed an average MMSE of 0.35 using the Herschel–Bulkley model (compared with MMSE values of 0.7 and 2.1 for Modified Bingham and Bingham, respectively). Moreover, the flow behaviour factor n (from Equation (12)) decreased to 0.92, 0.89, and 0.64 for cement contents of 250, 200 and 150 kg/m³ respectively, indicating a progressive increase in shear-thinning behaviour with increasing limestone filler content. Therefore, increasing limestone content results in a progressive increase in shear-thinning behaviour. These results indicate that the Herschel–Bulkley model provided the most suitable representation of the rheological response of the investigated mixtures, as it better captured the non-linear and shear-thinning behaviour observed in the eco-efficient mortars. Additionally, the analysis indicates that the Herschel–Bulkley model consistently provides the best representation of the rheological behaviour of the studied mixtures at 0, 15, and 30 min. This indicates that the torque response of the investigated mixtures was governed by non-linear flow behaviour, which became more pronounced as cement content decreased and limestone filler content increased.

The true viscosity profile (Figure 10) was derived from the HB model at 0 min, as this model was found to provide the best representation of the rheological behaviour of the investigated mixtures. The viscosity profile results demonstrate the need to appraise the viscosity of the mixtures at appropriate shear rates. Mixtures with 324 kg/m³ present approximately the same viscosity regardless of the shear rate applied; however, mixtures presenting lower mobility parameters (grey ones) had a higher viscosity at 0.1 s⁻¹ (low shear rate) than at 0.7 s⁻¹ (high shear rate). At 0.7 s⁻¹, mixtures with 320 kg/m³ yielded an average viscosity of 4.0 N·m·s, while mixtures with the same w/c and lower cement content (250 kg/m³) yielded a viscosity more than twice as high. Analyzing Figure 10b, mixtures with same mobility parameters present similar viscosity at 0.7 s⁻¹, being 8.8, 14.8 and 17.6 N·m·s for H, M and L, respectively. Mixture 150-0.89M was an exception as it presented viscosity similar to mixtures with lower mobility parameters. This appears to indicate that 150-0.89M achieved a limestone filler threshold of 51.8%, at which the viscosity started to increase. Similar results were observed by [34,82,83,84], where the viscosity increased when 20% or more of filler was added; however, the present results suggest that, for the mixtures investigated, a more pronounced rheological response occurred at a substantially higher limestone filler content.

6.2. Analyzing the Impact of Mobility Parameters on the Rheological Behaviour of Mortars

The HB flow behaviour factor (n), which describes the degree of shear-thinning behaviour of the mixture, showed a consistent correlation with the mobility parameters (MPT and IPS) regardless of cement content (Figure 11a). It is worth noting that 150-0.97H (red triangle in Figure 11), which displays an MPT of 0.35 μm, may be considered as an outlier to the trend. Without it, the linear relationship between flow behaviour and mobility parameters presents a R² of 0.82 and MSE of 0.27. This increase in shear-thinning behaviour was also observed by Yahia [78] and Varhen et al. [84] in cementitious mortars, when a w/p above 0.33 was selected, which agrees with this study where the w/p ranged from 0.40 to 0.47. However, in the present study, this behaviour was further quantified for low-carbon mortar mixtures with different cement reduction levels, and the results show that the rheological response was more strongly governed by mobility parameters than by cement content or w/p within the investigated range.

Figure 11b shows the relationship between mobility parameters and initial and final torque at 0.1 and 0.7 s⁻¹, respectively. Regardless of the w/c and w/p, all mixtures appraised present a strong correlation (R² of 0.92) with IPS and MPT at a high shear rate (0.7 s⁻¹). In contrast, an almost flat behaviour was found for the yield stress (torque at shear rate of 0.1 s⁻¹), which is expected since the slump flow of all mixtures was kept constant.

Figure 11c shows the true viscosity at 0.1 and 0.7 s⁻¹ as a function of MPT and IPS. Mixtures in low and high (0.1 and 0.7 s⁻¹) shear rate regimes demonstrated a notable correlation to MPT and IPS with a R² of 0.83 and 0.92, respectively. It should also be noted that these behaviours were largely independent of the w/p. Although Menezes et al. [23] correlated mobility parameters with slump results, the current study is more closely aligned with Rebmann et al. [52], who successfully predicted final torque and maximum apparent viscosity using mobility parameters. Additionally, the influence of filler content (i.e., both percentage and type) is implicitly captured by the mobility parameters, as the five rheological responses (flow behaviour factor, initial and final torque, and initial and final viscosity) are shown to be proportional to the mobility parameters. These results highlight the contribution of this study in showing that MPT and IPS serve as reliable practical indicators of the rheological behaviour of low-carbon mortar mixtures under both low and high shear rate conditions. It should be noted that the proposed relationships were established for the investigated mixtures and water-to-powder ranges only, and that the specific influence of admixture dosage on the calculated MPT and IPS values was not isolated in the present study.

Admixtures have a predominant impact on the surface force regime, explained by IPS [85], and in the present study, slump flow was kept constant (similar yield stress). Therefore, at low shear rate regimes, the collision force regime of the fine aggregates explained by the MPT would govern. Additionally, previous works [86,87] studied a high range of solids content (i.e., percentage of cement, filler, and aggregates to the total volume) and concluded that above a critical concentration (~65%), mixtures are governed by direct frictional forces (MPT) instead of the hydrodynamic force regime (IPS), agreeing with the current study as the solids content is between 77% and 81%. Moreover, Damineli et al. [88] used distinct amounts of admixtures to obtain maximum dispersion for a given IPS, and it was concluded that since admixtures are not accounted for in the IPS, the mixture viscosity did not present correlation with the IPS. This agrees with the present study, where admixtures were employed to maintain constant dispersion (slump flow), which leads again to the understanding that the initial viscosity of mortar mixtures would relate to the MPT instead of the IPS.

Therefore,

Table 7. Average Herschel–Bulkley parameters for distinct ranges of mobility parameters.

IPS Range	MPT Range	Initial Torque (0.1 s−1)	Initial Viscosity (0.1 s−1)	Final Torque (0.7 s−1)	Final Viscosity (0.7 s−1)	Flow Behaviour Factor (n)
		N·m	N·m·s	N·m	N·m·s
0.49–0.54	0.38–0.40	4.6	12.4	10.4	8.8	0.91
0.44–0.49	0.36–0.38	5.9	24.0	15.9	14.7	0.85
0.40–0.45	0.34–0.36	6.5	28.9	19.1	17.6	0.67

was created to facilitate the understanding of the impact of mobility parameters on the rheological behaviours of the mortars assessed.

6.3. Analyzing the Compressive Strength Through Mobility Parameters

The compressive strength of cement-based materials is traditionally described by Abrams’ law, which establishes an inverse relationship between the water-to-cement ratio and the compressive strength of the hardened material [89,90,91]. Additionally, the compressive strength is inversely proportional to the porosity, where lower porosity corresponds to higher strength. Although mobility parameters were first developed to aid the understanding of fresh state properties [92,93], the concept of the distance between the cement grains [89] (i.e., hydrated product distances) has demonstrated a strong correlation with compressive strength in cement pastes [56]; this is re-examined herein. The w/p concept (employed as a workability condition descriptor) has been mentioned as a possible method to evaluate the reduced spacing for hydrate formation by John et al. [9], and the observed relationship in the second phase (i.e., the phase with fillers) does not show a link between the w/p and compressive strength or porosity. Figure 12a demonstrates no relationship between the compressive strength and the w/p solely. The w/c ratio traditionally observed provided a much better relationship to the compressive strength, although there was a noted variation between the mixtures with and without limestone fillers (Figure 12b). Although the w/c predicts well pure cement and sustainable mixtures, it is hard to predict the compressive strength of high limestone filler replacement sustainable cement-based mixtures without trial batches.

This happens because traditional mortar compressive strength laws (e.g., Abrams’) do not currently account for the physical effects (e.g., dilution effect, nucleation effect, and packing effect) and the inclusion of varying types of fillers (i.e., PLF and RLF) [33]. Recent research into w/c suggests that available cement particulate distance [89] combined with decreased porosity [32] through filler packing may explain the increased mechanical capacity. Following Bergmann et al. [32], an alternative method used to predict the compressive strength of PPM cementitious pastes developed with high limestone filler content is proposed, where the compressive strength displays an exponential relationship to IPS cement·w/p. Similarly to IPS, IPS cement appraises the distance between particles [94], but in this case only cement particles are considered the solids of investigation. In this context, limestone filler acts both by increasing the spacing between cement particles (dilution) and by filling voids (filler), although it does not generate additional C-S-H to improve compressive strength. When this effect is combined with the coarse spacing contribution represented by the MPT [11,92], the relationship with hardened state properties is further improved, as shown in Figure 13. These results suggest that mobility parameter-based spacing concepts are not only relevant to fresh state behaviour, but also help explain the hardened state trends observed in mixtures with high limestone filler contents. Therefore, for low-carbon mortar mixtures with high limestone filler contents, mobility parameter-based spacing concepts provide a more suitable framework to interpret compressive strength than w/p alone. It should be noted that, although limestone filler may also contribute to hydration acceleration by nucleation effects reported in the literature, this mechanism was not isolated in the present study, and the observed behaviour is interpreted mainly in terms of dilution, spacing, and filler effects.

6.4. Global Warming Potential

As previously mentioned, the GWP can be used to assess the embodied energy of a mixture, and is presented in Figure 14a. For comparison purposes, a set of data points from the volumetric mortar fraction of a conventionally designed mixture, adapted from [10], was used as a reference benchmark to assess the reduction in embodied energy of the investigated mixtures. ACI mixtures of 35 MPa and 25 MPa developed with cement contents of 370 and 314 kg·m⁻³ emit 482 and 410 kg CO_2eq·m⁻³, respectively. PPM mixtures developed with a cement content of 472 kg·m⁻³ and without fillers present an average contribution of 429 kg CO_2eq·m⁻³. As expected, the GWP was tied to the cement content, with cement contents of 250, 200, and 150 kg·m⁻³ resulting in an average GWP of 343.8, 279.3, and 213.2 kg CO_2eq·m⁻³, respectively. Compared to the conventional ACI 35 MPa benchmark (482 kg CO_2eq·m⁻³), mixtures with 250, 200, and 150 kg·m⁻³ of cement achieved GWP reductions of approximately 29%, 42%, and 56%, respectively. Compared to the PPM reference without fillers (429 kg CO_2eq·m⁻³), the reductions reached 20%, 35%, and 50% for the same cement content groups.

Figure 14b shows the CO₂ Intensity Index (ci_cs), appraising the aforementioned eco-efficiency of materials. The conventional mixtures designed for 35 MPa and 25 MPa display an efficiency of 12.9 and 16.4 kg·m⁻³·CO_2eq·MPa⁻¹. The PPM mixtures without LF (324 kg·m⁻³ cement) exhibited an average ci_cs value of 12.97 kg·m⁻³·CO_2eq·MPa⁻¹ (STD 1.01), while mixtures designed with 250 and 200 kg·m⁻³ showed an average ci_cs value of 9.22 kg·m⁻³·CO_2eq·MPa⁻¹ (STD 0.97). These mixtures can be verified as more sustainable, with improved efficiency compared to the conventional mixtures. Lastly, two of the mixtures developed with 150 kg·m⁻³ of cement (150-0.89M and 150-0.84L) had an average efficiency (elevated ci_cs) of 9.7 kg·m⁻³·CO_2eq·MPa⁻¹ (COV 0.67). The final mixture (150-0.97H) had a much lower efficiency of 16.5 kg·m⁻³·CO_2eq·MPa⁻¹; however, this is likely due to the low compressive strength and would not be comparable to the aforementioned mixtures. Moreover, mixtures developed with 200 kg·m⁻³ had a ci_cs ranging from 7.6 to 10.0 kg·m⁻³·CO_2eq·MPa⁻¹ for compressive strengths between 37.0 and 27.9 MPa. Therefore, 200-0.69M and 200-0.64L achieved similar compressive strengths to that of the control mixtures (320-0.60H and 324-0.56M, respectively), with a ci_cs 36% lower, highlighting the ability of producing sustainable mixtures with the presented mix-design method. These results demonstrate that the PPM and MP-based approach can produce low-carbon mortar mixtures with high limestone filler contents, delivering conventional compressive strength targets (25–40 MPa) with GWP reductions of 20–56% relative to standard ACI benchmarks, representing a meaningful pathway for the concrete industry towards net-zero carbon production without compromising structural performance. Although this study focused on mortar systems, the mix proportions were derived from full concrete designs, and a companion study extending these findings to concrete mixtures—including coarse aggregate effects—is currently underway.

7. Conclusions

The current research investigated mobility parameters in both the fresh and hardened states of mortar mixtures designed using PPMs and high LF contents. The main findings of this research are presented as follows:

• Replacing cement content with limestone filler enhanced the mortar hardened state properties with similar w/c.
• Five rheological parameters (i.e., flow behaviour parameter, torque at low and high shear rates, and viscosity at low and high shear rates) showed strong correlations with IPS and MPT, confirming that reduced mobility parameters increased the resistance to flow of the investigated highly packed mortars.
• Based on the results obtained, Table 7 is proposed to guide the selection of MP parameters to achieve targeted fresh state outcomes.
• IPS and MPT can be used as practical indicators of the low shear rate viscosity of highly packed mortar systems with similar initial consistency, regardless of cement content.
• The compressive strength trends of mixtures with high limestone filler contents were better interpreted by the combined use of IPS cement, w/p, and MPT than by w/p alone, linking spacing-based concepts to the hardened state behaviour of the investigated mortars.
• The investigated mixtures achieved a 36% reduction in ci_cs and up to a 50% reduction in GWP, confirming the potential of the proposed MP-based framework as a practical tool for low-carbon concrete production. The scale-up to full concrete mixtures is the subject of ongoing research.