Eco-Efficient Mortars for Sustainable Construction: A Comprehensive Approach

Abstract

Cement production is responsible for approximately 7% of global carbon dioxide emissions. Despite our efforts, we have not been able to find a competitive substitute that is both reliable and environmentally friendly. The easiest way to solve the issue is to rationalize resources and try to minimize their use by replacing them with other materials. The current market shortage and reduced initial strength have limited the availability of blends that contain a significant amount of fly ash. Given the current economic, political, and environmental circumstances, it is predicted that a solution may be ternary blends with cement, fly ash, and MTK. Despite being “ancient” materials, there have been no recent global performance assessments. In this context, an investigation was carried out with ternary blend mortars. A significant volume of cement has been replaced with fly ash and metakaolin. The results show that these blends’ performance is promising because they offer a wide range of possibilities for replacing cement, maintaining or even improving its properties. MTK and fly ash’s synergies significantly enhance mechanical performance and durability. Furthermore, the global sustainability analysis shows that ternary blends are 36% more efficient than binary blends of cement and fly ash or metakaolin.

1. Introduction

Carbon dioxide (CO₂) is the most important anthropogenic source of greenhouse gases in the atmosphere. It is responsible for approximately 66% of the radiative forcing by long-lived greenhouse gases (LLGHGs), as well as 82% of the increase in radiative forcing over the last decade [1,2]. The latest data are not encouraging, as the concentration of CO₂ in the atmosphere indicates that records have grown significantly. There are historical records from at least four glacial periods, observing the following: before the pre-industrial era, the concentration of CO₂ was close to 278 ppm; it reached 400 ppm in 2015; in 2020, there were already records of 413 ppm, a growth of 1.5 times compared to the pre-industrial era.

The evolution of these values was mainly attributed to emissions caused by burning fossil fuels, cement production, and, to a lesser extent, due to deforestation and land use changes [1,2,3,4,5,6,7,8]. Cement accounts for about 7% of the global greenhouse gas emissions from all human activity [9]. In addition, approximately half of the CO₂ emissions released in cement production come from limestone calcination.

Furthermore, the environmental issues caused by cement have been worsening. To illustrate a few examples: the speed of global construction and development; the rapid growth of the economies of countries such as China and India; increasingly demanding construction deadlines; the constant waste in the rapid and inefficient conception of materials; the evolution of the world’s population, with a population explosion of 9.7 billion inhabitants, located mainly in urban areas; as well as the short-term perspective of environmental costs [2,9,10,11,12,13]. Identifying key elements is essential for reducing cement dependence. The reduction in cement consumption, careful use of this material, reducing the cement/clinker ratio, and using increasingly efficient blends are some of the measures that can be taken.

Cement is the second most consumed material in the world, surpassed only by water [14,15,16]. It is also an effective vehicle to incorporate high levels of industrial by-products, leading to advantages [17,18,19,20,21,22,23,24,25]. Replacing cement has at least three advantages. The economic impact is due to the use of materials that have lower market values. From a technical standpoint, there is a potential to achieve higher performance characteristics, particularly in terms of durability. Concerning the environmental nature, it is due to the lower embodied energy and the reduction in CO₂ emissions released into the atmosphere [26,27,28]. Therefore, using less cement when providing mineral additions can undoubtedly reduce environmental impacts. For these reasons, the incorporation of additions into concrete, mortars, or pastes is essential, both environmentally and economically [29,30,31,32,33].

Although there are many studies on additions, their characteristics and properties are still not fully explored and consolidated in this context [34,35,36,37,38]. It is essential to highlight at least three current issues. The durability challenges of pozzolanic additions are the first thing to consider. When the amount of it being incorporated is high, the alkaline compounds in the concrete are consumed, leading to a decrease in the pH of the porous system. This can destroy the passivating layer that protects armour from corrosion, compromising its durability [39,40,41,42,43]. Another issue is that the European market is facing a shortage of fly ash (FA) due to the closure of coal-fired power stations. This is one of the most significant additions driving the search for new alternatives [44,45,46,47]. Finally, another challenge is to establish methods for evaluating additions, mainly in the overall assessment, including strength, durability, and sustainability [48,49].

One solution to all these issues could be to develop ternary blends with fly ash (FA) and metakaolin (MTK). By using FA, blends can be produced with less cement. Additionally, there are other benefits that are known, both during the initial stage and during the finished product [50,51]. In order to maximize the environmental benefits, it is important to replace as much cement as possible. However, this may not be feasible due to the disadvantages of FA, such as the slow development of pozzolanic reactions. The expectation is that MTK will reduce or reverse some of these disadvantages, namely, correcting the initial lower strengths. It is expected that one can produce blends with a large volume of mineral additions and with mechanical behaviour similar to that of conventional blends or even improved.

The key to a good result may be in MTK. The stages of the hydration process of MTK blended cement pastes are already understood. However, there is still much to learn [52] because the processes are complex [53]. To put it briefly, MTK’s performance in different binder systems is mainly due to its pozzolanic reaction with calcium hydroxide (CH) [53,54,55,56,57,58]. As a result, additional hydration products are formed, primarily different calcium aluminum silicate hydrates [53,57,59,60,61]. The concentration of the CH solution strongly determines the duration of this pozzolanic reaction, which can last up to 90 days [62]. On the one hand, CH is closely related to the process of cement hydration: the greater the amount of cement or hydration time, the greater the amount of CH. In contrast, MTK reacts with CH until there are no more reagents available [53,58,62,63]. In addition, if FA is present, it also reacts with CH, leading to a faster decrease in CH availability [15,45,46,64,65,66,67]. There are other processes, such as the filler effect [68], filling effect [69,70], or nucleation effect [53]. In addition to the pozzolanic reaction, they have the ability to change the matrix’s characteristics.

The work reviewed clearly shows that MTK is a highly effective pozzolan. All of these processes lead to the refinement of the porous structure. The microstructure of cement paste becomes more uniform and denser, with improved durability and strength gain [15,32,33,52,53,56,63,68,71]. MTK enhances the resistance against water transport and diffusion of harmful ions, which cause degradation of the matrix [15,32,68]. So, it shows a marked decrease in capillary porosity, chloride ingress, gas permeability, and sorptivity [56,62,70,71,72].

It is important to mention that FA also has a pozzolanic behaviour, which means it also reacts to CH. In fact, it is the active compounds of FA—silica and alumina—that react with CH, producing hydrated calcium silicates, hydrated calcium aluminates, or hydrated calcium silicoaluminates. These reactions are slower than those of MTK [64,65].

In summary, the main vectors that can justify the use of ternary blends are as follows:

• Producing blends by using a significant amount of FA. Thus, large volumes of the traditional binder were substituted by this industrial by-product, which otherwise would have to be stored in landfills and may also be a source of pollution;
• Concrete compactness and durability are enhanced by these FA contributions, although eventually, the effect will only be felt in the long term due to the slow pozzolanic reaction. The workability of these additions is also enhanced;
• Producing concrete with a very low W/C ratio, with workability controlled by the use of superplasticizers. This can lead to even more compact, stronger, and durable concrete;
• Introducing MTK to ensure the initial strengths necessary to make this concrete competitive in current construction. In addition, using MTK can contribute to an increase in the strength and durability of concrete due to its high reactivity and fineness.

But, although there is already consolidated knowledge about these materials, as seen above, there have been no recent global performance assessments on ternary blends with MTK and FA. Studies have not centred around sustainability analysis as the main focus. Emphasizing the word “recent” is crucial as both costs and environmental parameters are highly time-sensitive. Studying problems worldwide is also important because the current situation is unique and challenging, as already mentioned above. Today’s researchers may have shifted their attention to more innovative and complex materials and behaviours, which may be the reason for this lack. The traditional materials were neglected.

Numerous studies have been conducted on MTK and FA blends, for instance. But usually, no cost or sustainability analysis is carried out. Additionally, no global analysis is performed [33,71,73,74,75,76,77,78,79]. In addition, certain authors have conducted research on these blends, but using specific cements [80,81,82]. Others studied blends with a greater number of different materials [83,84,85]. There are researchers who analyze costs but do not consider sustainability [85]. Others consider sustainability, but only in binary blends [86,87]. Some authors conducted sustainability analyses on ternary mixtures, but the materials they used were distinct from those used in this research [88]. In addition, other authors performed cost and sustainability analysis but with very different parameters from this investigation and without conducting an overall analysis [89]. Other authors have studied ternary blends with MTK and FA but only to produce geopolymers [90,91,92,93,94]. This is not within the scope of the present work.

Even in recent review articles, the authors have taken a very different approach to ternary blends with MTK and FA. For instance, Homayoon et al. [95] made an interesting and complex analysis of environmental issues and CO₂ emissions. The authors discussed cost and environmental analyses by constituents, the role of MTK in the perspective of the cement industry according to sustainability scenarios, the comparison between CO₂ emission of MTK production and Portland cement, etc. However, they neglected to examine ternary compositions or perform a comprehensive analysis of sustainability, cost, or performance. Also, Yang et al. [96] performed very interesting research on materials within Low-Carbon, High-Performance Cement-Based Composites (LCHPCC). Life cycle assessment analysis and practical validation case studies are among the author’s considerations. However, they studied ternary blends that were different from this work. However, they did not conduct a global analysis of sustainability, cost, and performance. Finally, researchers should refer to a comprehensive and extensive review by Brito and Kurda [97], which provides an overview of concrete sustainability in all possible ways. Despite extensive research, the authors did not examine ternary blends with MTK and FA from a global perspective.

In this context, an experimental program was developed. It encompasses the characterization of the performance of ternary blends, where significant volumes of cement were replaced by FA and MTK. Despite being “ancient” materials, there have been no recent global performance assessments. And this is clearly an added value of this research. In light of the current economic, political, and environmental situations, it is expected that these materials could be one solution to current and future problems. Furthermore, the aim is to demonstrate how these blends can and should be evaluated globally.

Thus, the production of mortars was considered, which represented concrete, since the results can be correlated, as demonstrated in other studies [50,98,99]. These mortars were characterized in four different ways: a simple cement blend (reference), a binary cement blend with MTK, three binary cement blends with FA, and three ternary cement blends with MTK and FA. The binder mass was used to calculate the incorporation of the additions as a replacement for cement. The binary cement blend with MTK was produced using “only” 10% of MTK. The reason for this 10% rate is to achieve the most effective mechanical strength gain without compromising workability. According to some research, the incorporation of MTK should range between 8 and 10% to enhance mechanical strength [53,63,68]. In contrast, it is acknowledged that the greater the MTK, the less workability, as it increases the water requirement. In addition, the workability could be completely compromised by 15% MTK [32,52,53,68,71,100,101]. The large surface area and the irregular surface morphology of MTK result in the large adsorption of water, which thus decreases the fluidity of fresh cement paste due to the reduction in free water distributed in the spaces between the cement and MTK particles [32,53,60,102]. Furthermore, 10% and 15% MTK may be sufficient to control deleterious expansion due to alkali–silica reaction in concrete and showed excellent durability to sulfate attack [32].

The binary cement blends with FA were made with 20, 40, and 60% to ensure a comprehensive range. In order to improve evaluation, the rates of incorporation of ternary blends were the sum of the incorporation of binary blends. That is, the ternary blends were made with 10% MTK and 20, 40, or 60% FA. It is obvious that the sum of binary blends cannot be identical to the ternary blends. The physicochemical effects cannot be directly attributed. Moreover, ternary blends had more additions than binary blends. However, this methodology allowed for a performance comparison between binary and ternary blends. Furthermore, it was useful to assess synergies when using both additions simultaneously.

Subsequently, the parameters of workability, mechanical strength, durability, and cost were assessed. After that, the statistical and sustainability analysis was performed. The MTK’s pozzolan reactions can last up to 90 days of age [62]. As a result, hardened concrete tests will also be conducted at this age.

The sustainability analyses involved using a qualitative evaluation methodology. This includes a multi-criteria decision support method for relative sustainability (MARS-SC). Other authors have adopted this methodology [86,87,103], but it has not been applied to ternary cement blends with MTK and FA. The first step in the assessment involved quantifying, normalizing, and aggregating indicators related to environmental, economic, and functional aspects (mechanical properties and durability) [48,104]. After that, global analyses were performed. This was performed while taking into account all indicators simultaneously, obtaining a big picture of the performance of all blends.

2. Materials and Methods

2.1. Materials

2.1.1. Binders

Portland limestone cement (CEM II/B-L 32.5N), manufactured by a Portuguese company, Cimpor, which includes clinker and limestone filler, was adopted. The cement was produced at the Souselas Plant, located at Rua dos Troviscais 10, 3020-886 Souselas, Portugal. The manufacturer reports that the compressive strength was 16.0 and 32.5 MPa after 7 and 28 days, respectively (

Table 1. Properties of cement.

Constituents [%]	Clinker	65–79	Characteristics	Initial set [min]	75
	Limestone	21–31		Expandability [mm]	≤10
	SO3	≤3.5
	Cl	≤0.10		Comp. str. 7 d [MPa]	16
	Other minorities	≤5		Comp. str. 28 d [MPa]	32.5

) [105].

The FA was obtained from the Pego thermoelectric power plant located in Portugal. These consist mainly of silica (60.87%), alumina (20.40%), and iron (7.82%). There is a total CaO content of 2.72%. The density was 2360 kg/m³, and there were 27.30% of fines exceeding 45 μm. Previous research has already characterized FA, which is summarized in

Table 2. Properties of fly ash.

Chemical composition	Loss on Ignition [%]	7.30	Cl− [%]	0.00
	SiO2 [%]	60.87	Free CaO [%]	0.00
	Al2O3 [%]	20.40	Na2O [%]	0.55
	Fe2O3 [%]	7.82	K2O [%]	1.92
	Total CaO [%]	2.72	P2O5 [%]	1,14
	MgO [%]	1.40	TiO2 [%]	1.29
	SO3 [%]	0.22	Total SiO2 + Al2O3 + Fe2O3 [%]	89.09
Physical properties	Density [kg/m3]	2360	Fineness > 45 µm [%]	27.30
	Blaine’s specific surface [m2/kg]	387.9	Humidity [%]	0.16
			Water demand [%]	0.297

[51,106].

The MTK was obtained by thermal activation of kaolin clay extracted from Barqueiros, located in Barcelos Council, Portugal, and named Mibal-C. The sedimentary deposit comprises kaolin and sand, and it is estimated to have a brute reserve of millions of tons [101]. To obtain MTK, kaolin was calcined to a temperature of about 700 °C, and MTK was already characterized in previous investigations, which are summarized in

Table 3. Properties of metakaolin.

Particule dimension [%]	<30 µm	99 ± 3	Unburnt	Loss on Ignition [%]	12.75
	<10 µm	93 ± 5	Humidity [%]	Initial	32 ± 3
	<5 µm	82 ± 5		Beads	18 ± 2
	<2 µm	68 ± 6		After drying	<2
Chemical composition [%]	SiO2	47.0	After drying parameters	Burnout	0.09
	Al2O3	37.1		Flexion strength (110 °C) [MPa]	2.45 ± 0.49
	Fe2O3	1.3
	K2O	2		After burnout flexion strength [MPa]	13 ± 3
	Na2O	0.2
	MgO	0.15		Water absorption [%]	10 ± 2
	TiO2	0.3	Others	Density [g/cm3]	2.4–2.7
	CaO	0.1		Suspension’s pH	6–9

[107,108].

2.1.2. Aggregates

A river sand with a modulus of fineness of 3.44 was employed [109]. The maximum diameter was set at 4 according to the ASTM criteria [110,111]. Figure 1 shows the results of the granulometric analysis.

2.1.3. Superplasticizer

GLENIUM 77 SCC from BASF was employed as a superplasticizer (SP). It was characterized as a powerful high-range water reducer/superplasticizer (T3.1, T3.2 NP EN 934-2: 2003) [112]. It had a density of 1.05 ± 0.02 g/cm³. The maximum content of chlorides and alkyls was less than 0.10 and 0.35%, respectively. According to the manufacturer, it is recommended to take 1.5% of the binder mass.

2.2. Methods

The experimental program was designed to study mortars. In the composition of the samples, eight mortars were used, with the addition of cement (CEM), fly ash (FA), and metakaolin (MTK) (

Table 4. Studied compositions.

		B	CEM	MTK	FA	S	W/B	SP
Mixt.	Designation	[kg/m3]	[%]	[%]	[%]	[kg/m3]	[-]	[%L]
I	REF	484	100	0	0	1457.9	0.55	0
II	10%MTK	484	90	10	0	1449.1	0.55	1.5
III	20%FA	484	80	0	20	1422.8	0.55	0
IV	40%FA	484	60	0	40	1387.6	0.55	0
V	60%FA	484	40	0	60	1352.5	0.55	0
VI	10%MTK + 20%FA	484	70	10	20	1414.0	0.55	0
VII	10%MTK + 40%FA	484	50	10	40	1378.8	0.55	0.4
VIII	10%MTK + 60%FA	484	30	10	60	1343.6	0.55	1.5

). This table also indicates the total quantities of binder (B = CEM + FA + MTK), sand (S), water-to-bind ratio (W/B), and superplasticizer (SP). It should be noted that the binder mass was used to calculate the incorporation of the additions as a replacement for cement. As a starting point, a reference mortar was made solely with cement (I). Subsequently, mortars were mixed with cement and the addition of 10% MTK (II) and 20%, 40%, and 60% FA (III–V). Finally, ternary blends were produced with cement; 10% MTK; and 20%, 40%, and 60% FA (VI–VIII).

As previously mentioned in the introduction, the use of MTK results in a significant decrease in workability. From the work of another researcher, who used precisely this MTK, it is also known the mixtures become sticky inside the mixer [101]. This point is crucial because the workability after mixing is vastly different from the workability during mixing. Other researchers have also observed that standard workability tests were not able to quantify the influence on the general flow properties of MTK concrete [100]. For these reasons, the primary concern was to ensure the production of ternary mixture VIII (10% MTK + 60% FA) based on the SP content provided by the manufacturer. Since the slump test was 20 cm and it was possible to perform the mixing, it was considered an acceptable solution. Then, it was checked if the VII mixture (10% MTK + 40% FA) supported the same amount of SP. It was concluded that it did not. So, the SP content was determined to be comparable to that of mixture VIII: 0.4%L. Finally, it was determined that binary mixture II (10% MTK) should contain 1.5% SP. This decision was made for two reasons. First is because mixture VIII also had this content. Second, even with a good slump of mixture II (>30 cm), it was hard to mix because the mortar was sticky.

Figure 1 shows the structure of the experimental program.

Figure 2 shows the structure of the experimental program. Mortars were produced using a variable-axis electric mixer, Lisprene-LIS 140, that has a capacity of 140 liters. Immediately afterward, the flow test was carried out, according to EN 1015-3: 2004 [113]. The results of each mixture were determined by the arithmetic mean of two orthogonal readings. Then, the specimens were moulded. After one day, the specimens were demoulded and immersed in water for up to 90 days of age. The containers with the specimens and water remained in a humid chamber with a temperature of 20.8 ± 0.27 °C and a relative humidity of 87.8 ± 3.38%.

Over time, multiple tests were conducted for each mixture. All tests were performed in a laboratory environment: a temperature of 20 ± 3 °C and a relative humidity of 62 ± 7%. To carry out the flexural and compressive strength, EN 196-1:2006 was taken into consideration [114]. The LLOYD Universal Materials Testing Machine LR50K Plus was used, with a preload of 50 N and 45 N/s. To carry out the flexural strength test, three specimens of 40 × 40 × 160 mm³ were taken for each age: 3, 7, 14, 21, 28, and 90 days. By using 6 specimens 40 × 40 × ±80 mm³ from the flexural test, the compressive strength test was performed at the same ages.

Migration tests were conducted in a non-stationary regime to evaluate the indicators of durability, according to E463-2024 [115]. Three cylindrical specimens with 50 mm height and 100 mm diameter were used for each mixture at 90 days of age. After being cut, the specimens were vacuum-placed in a desiccator for 3 h. Afterward, the desiccator was filled with water to cover all the specimens. Then, the vacuum was maintained for another hour (at 0.6 bars) before releasing air into the container. The sample rested for 18 h. Following this, the chloride diffusion test was conducted, which lasted either 24 or 48 h. Silver chloride was used to visualize the attack. The results of each sample were obtained by calculating the arithmetic mean of 7 equidistant readings. The final results were determined by combining the mean of the three samples.

Water absorption by capillary was also carried out using 3 cubic specimens with a 50 mm edge for each mixture at 90 days of age. The test was conducted in compliance with EN 1015-18:2002 [116]. It should be mentioned that only the four first test hours were considered to determine the capillary absorption coefficient. By combining the mean of the three samples, the final results were established.

3. Results

3.1. Workability

The results of the flow test are shown in Figure 3: grey represents reference; orange represents binary blend with MTK; green represents binary blends with FA; blue represents ternary blends with FA and MTK. The results show atypical values because SP was used in some blends. The use of SP is a result of the low workability of blends that contain MTK. This makes mortars drier and less mouldable, trapping water. The low workability during the initial mixing stages is particularly concerning, which can result in a poor evaluation of the use of SP. This is the reason why mixture II (10% MTK) has a result above 30 cm: at an early stage, it is very difficult to mix all the components in the mixer, but then the consistency changes suddenly, making the mortar more fluid.

It can be concluded that the workability of MTK is not critical up to 10% incorporation as a CEM replacement. On the contrary, FA results in mortars with increasing workability. These two additions can be complementary concerning flowability because FA mitigates the disadvantage of using MTK. It is possible that the reason for this is the FA’s spherical shape compared to MTK’s [71].

3.2. Mechanical Strength

The results of the mechanical strength tests over time are shown in Figure 4: flexural strength (left) and compressive strength (right). It has been noted that FA (green lines) can be very detrimental. The higher the incorporation of FA, the lower the strength. This is likely due to the replacement of cement but also due to slower pozzolanic reactions. On the contrary, the MTK (orange line) enhances strength at all ages. It reacts very quickly with calcium hydroxide and has high initial strength. When MTK and FA are combined, the ternary blends (blue lines) match or exceed the reference (black line). This shows the good synergy between these additions: MTK acts in the younger age range, and FA acts in the older age range.

This fact can be clearly seen in Figure 5. This shows the results of compressive strength at 28 and 90 days: a solid fill represents 28 days, and a shadow represents 90 days. At 90 days, the binary blend with MTK (II) has a strength 27% higher than the reference (I). On the other hand, when FA is incorporated between 20 to 60% (III to V), strength decreases by 40 to 72%. However, when these FA are combined with MTK (VI to VIII), the strength reduction is only between 22 and 44%. The greater the compactness, the greater the strength. It is suggested that the MTK produces more compact materials with fewer voids.

These findings are consistent with other research. In the Introduction, it was stated that MTK improved strength gain [15,32,33,52,53,56,63,68,71], which explains the superior outcomes of mixing II. FA has slower reactions than MTK, making it difficult to compete for the CH available [64,65]. In addition, mixtures with FA replace 20 to 60% of the cement. Due to these factors, the strength of binary mixtures is naturally lower than the reference. Moreover, the synergistic effect of ternary mixtures has been observed and justified in the past: a more uniform mix, denser microstructure, and better performance of the hardened concrete [71].

3.3. Chlorides by Migration

The results of the diffusion coefficient of chlorides by migration are shown in Figure 6. It has been observed that all blends with additions (II to VIII) perform better than the reference (I). The best results come from ternary blends with MTK and FA (VI to VIII). Moreover, these blends have a very low diffusion coefficient, 88% less than the reference value. This suggests an excellent synergy between MTK and FA. Furthermore, these blends can be excellent for structures that are highly exposed to aggressive agents.

These findings are consistent with other research. In the Introduction, it was stated that MTK enhances the resistance against water transport and the diffusion of harmful ions, including chlorides [15,32,56,68,70,72]. In ternary mixtures, it was found that MTK performed better than FA [71].

3.4. Water Absorption by Capillarity

The water absorption by capillarity coefficients at 90 days is shown in Figure 7. The binary blend with MTK (II) performs 33% better than the reference (I). Binary blends with FA (III to V) also perform better—between 3 and 48%—than the reference. The best results are achieved with ternary blends that include MTK and FA (VI to VII). The performance is 33 to 73% better than the reference. Note that this test is representative of the fulfillment of large capillary pores. Moreover, it is concluded that reference (I) has higher values than the other blends. This is possible because blends with additions have smaller pores. This supports the results of chloride diffusion (Figure 6), as well as some of the mechanical strength results (Figure 4 and Figure 5). The excellent properties of both FA and MTK make this possible. The additions are pozzolanic, producing additional hydration products that result in reduced porosity. On the other hand, these additions act as a filler, which also contributes to the reduction in porosity.

3.5. Cost Analysis

The cost of the produced mortars corresponds to the purchase price of materials in a store. So, it includes production, transportation for the store, maintenance, and the profit margin of the seller. It does not take into account final disposal or the entire life cycle costs. SP was not included in the calculation as it was embedded in the price of MTK. As can be seen, the price given to MTK is on the safe side.

Despite the recent price instability caused by COVID-19, it is known that the price of CEM tends to be competitive. Competitiveness will be expected for future additions that attract interest and demand. In 41 countries across all continents, the average price of cement is around 0.20 EUR/kg [117]. The price of MTK varies greatly depending on the place of sale: in China, it is about 3.4 EUR/kg, while in Germany, it is about 0.1 EUR/kg [118]. The average value of 0.22 EUR/kg is believed to be safe for the intended analysis. If MTK is a suitable solution for concrete, the price will likely decrease in the future. FA is a scarce product in some parts of the world, mainly in Europe. The global average price is 0.02 EUR/kg [119]. Finally, the price of sand is 0.14 EUR/kg [120].

Each mixture’s costs were analyzed, and the cost/benefit ratio was estimated using the compressive strength at 28 and 90 days. The results are shown in Figure 8. It can be concluded that the use of FA in binary blends (III to V) is considerably detrimental. However, the ternary blends containing MTK and FA (VI to VIII) have a cost–benefit ratio that is comparable to the reference (I). MTK can be used to correct the FA.

3.6. Statistical Analysis

All experimental data were analyzed statistically. The aim was to investigate the statistical differences between different blend “groups”. For this, four “groups” were considered: REF, MTK, binary blends, and ternary blends. IBM SPSS Statistics V24 software was used to perform the statistical tests. In this evaluation, a significance level of 5% was taken into account [121,122,123].

The workability results (Flow) were evaluated using a One-Way ANOVA test and a Ryan–Einot–Gabriel–Welsch Post Hoc test. The Post Hoc was chosen for its neutrality, which is neither too conservative nor too liberal. It was also chosen because Levene’s test showed that there is homogeneity of variance. The chlorides by migration (Cl) and water absorption by capillarity (Cm) results were also evaluated in the same way. The only difference was in the case of chlorides; Levene’s test showed that there was no homogeneity of variance. Therefore, the approach was performed using the non-parametric Welch’s One-Way ANOVA test, with a Games–Howel Post Hoc. Finally, the evaluation of flexural (Rf) and compressive (Rc) strength results was carried out by performing a Two-Way ANOVA test and a Ryan–Einot–Gabriel–Welsch Post Hoc test. In One-Way ANOVA tests, two factors were added: “mixture” (I to VIII) and “groups”. In Two-Way ANOVA tests, two factors were added: “age” (3, 7, 14, 21, 28, and 90 days) and “groups”.

Table 5. Summary of statistical analysis.

Test	Designation	Post Hoc	Description	Differences
Flow	One-Way ANOVA	R-E-G-W	F(3,12) = 12.563; p < 0.001	MTK ≠ Others
Rf	Two-Way ANOVA	R-E-G-W	F(3,114) = 65.955; p < 0.001	All different
Rc	Two-Way ANOVA	R-E-G-W	F(3,255) = 376.619; p < 0.001	All different
Cl	Welch’s One-Way ANOVA	G-H	F(3,19) = 22.0603; p < 0.001	Ternary ≠ Binary
				Ternary ≠ MTK
Cm	One-Way ANOVA	R-E-G-W	F(3,20) = 6.141; p < 0.004	Ternary ≠ Binary
				MTK ≠ Binary
				MTK ≠ REF
			F(a,b) = c; p < d
R-E-G-W Ryan–Einot–Gabriel–Welsch			a and b = degrees of freedom; c = F-Value
G-H Games–Howel			d = p-value

shows a summary of the various statistical analyses. In the last column of the table, the statistical differences observed are indicated (Differences). It is possible to conclude that there are statistical differences between the performance of ternary and binary mixtures: Rf, Rc, Cl, and Cm. It can also be concluded that in flexural (Rf) and compressive (Rc) strength, all four elements of the factor “groups” are statistically different from each other. Finally, it is concluded that the performance of binary mixtures with MTK is statistically different from that of other mixtures. In other words, these statistical analyses confirm the conclusions of the experimental work.

3.7. Sustainability Assessment and Global Vision

A qualitative evaluation methodology was used, called MARS-SC, to analyze sustainability. This consists of evaluating the performance of three indicators that, if properly weighted, make it possible to quantify the sustainability of the studied blends, through a final sustainability note [48]. The first step was to calculate the environmental parameters for each composition, which included the Primary Energy Consumption (PEC) and the Global Warming Potential (GWP). These are two fundamental concepts of building sustainability. They can be defined in a simple way as follows. PEC represents the sum of the energy consumed during the extraction of raw materials, their transport to the processing units, and their processing [124]. The GWP is a number that refers to the amount of global warming caused by a substance. The GWP is the ratio of the warming caused by a substance to the warming caused by a similar mass of CO₂ [125].

Table 6. Environmental parameters.

Materials		PEC	GWP
		[kWh/kg]	[g/kg]
Clinker	[48]	1.194	1000
Lime	[48]	0.0127	32
MTK	[126]	0.82	175
FA	[48]	0.0093	4
S	[16]	0.025	3
W	[16]	0.31	5

was used to illustrate this. The different parameters studied were normalized according to Diaz-Balteiro’s expression [48]. This allows an evaluation of the “best/worst” type, as shown in Equation (1): $\overline{{\mathrm{P}}_{\mathrm{i}}}$ represents the normalized result; ${\mathrm{P}}_{\mathrm{i}}$ represents the result of our tests; ${\mathrm{P}}_{\mathrm{*}\mathrm{i}} \mathrm{a}\mathrm{n}\mathrm{d} {\mathrm{P}}_{\mathrm{i}}^{\mathrm{*}}$ represent the worst and the best results, respectively. For example, for the calculation of Cs, 28 of the 20% FA mixture, the parameters ${\mathrm{P}}_{\mathrm{i}}$, ${\mathrm{P}}_{\mathrm{*}\mathrm{i}}, \mathrm{a}\mathrm{n}\mathrm{d} {\mathrm{P}}_{\mathrm{i}}^{\mathrm{*}}$ are 28.5, 13.4, and 55.3 MPa, respectively (Figure 5). Consequently, $\overline{{\mathrm{P}}_{\mathrm{i}}}$ is 0.36.

$$\overline{{\mathrm{P}}_{\mathrm{i}}}={\displaystyle \frac{{\mathrm{P}}_{\mathrm{i}}-{\mathrm{P}}_{\mathrm{*}\mathrm{i}}}{{\mathrm{P}}_{\mathrm{i}}^{\mathrm{*}}-{\mathrm{P}}_{\mathrm{*}\mathrm{i}}}}\forall \mathrm{i}$$

(1)

$${\mathrm{I}}_{\mathrm{e}\mathrm{n}\mathrm{v}}=0.75\cdot \mathrm{P}\mathrm{E}\mathrm{C}+0.25·\mathrm{G}\mathrm{W}\mathrm{P}$$

(2)

$${\mathrm{I}}_{\mathrm{f}\mathrm{u}\mathrm{n}\mathrm{t}}=0.20·\mathrm{f}\mathrm{l}\mathrm{o}\mathrm{w}+0.25·\mathrm{C}\mathrm{s},28+0.15·\mathrm{C}\mathrm{s},90+0.20·{\mathrm{D}}_{\mathrm{C}\mathrm{L}}+0.20·\mathrm{C}\mathrm{m}$$

(3)

$${\mathrm{I}}_{\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{n}}=1.00·\mathrm{c}\mathrm{o}\mathrm{s}\mathrm{t}$$

(4)

$$\mathrm{S}\mathrm{N}={\mathrm{w}}_{\mathrm{G}1}·{\mathrm{I}}_{\mathrm{e}\mathrm{n}\mathrm{v}}+{\mathrm{w}}_{\mathrm{G}2}·{\mathrm{I}}_{\mathrm{f}\mathrm{u}\mathrm{n}\mathrm{t}}+{{\mathrm{w}}_{\mathrm{G}3}·\mathrm{I}}_{\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{n}}$$

(5)

The radar diagrams of Figure 9 were created using the normalized results. The reference result is shown in black, the binary blends with FA are shown in green, and the ternary blends with MTK and FA are shown in blue. Note that the higher the value on the scale from 0 to 1, the better the performance. The figure clearly shows that ternary blends have a much better overall performance. Compared to the reference, they are only impacted by mechanical strengths and some workability. Compared to binary blends with FA, they are undoubtedly superior.

To make the analysis more quantitative, the results were combined into three main indicators: environmental (Ienv), functional (Ifunt), and economic (Iecon). Each of these indicators was calculated using Equations (2)–(4), with the parameters defined in

Table 7. Weights of sustainability assessment.

Indicator	Parameter	Weights
		Parameter	Indicator
Environmental	PEC	0.75	WG1 = 0.30
	GWP	0.25
Functional	Flow	0.20	WG2 = 0.50
	Cs,28	0.25
	Cs,90	0.15
	D,Cl	0.20
	Cm	0.20
Economic	Cost	1.00	WG3 = 0.20

. Using these indicators, sustainable notes (SN) were calculated, assigning several WGi weights according to the degree of representativeness appropriate for each indicator (Table 7) and Equation (5): WG1, WG2, and WG3 represent the weights of the environmental, functional, and economic indicators, respectively. The higher the sustained notes, the more sustainable the material is. On the left side of Figure 10, the sustainability notes are presented with weights of 30% for the environment, 50% for functionality, and 20% for the economy. It can be concluded that ternary blends (VII to VIII) have a high sustainability grade for these proportions. The binary mixture with 60% FA (V) also has an excellent grade, but not as good as the ternary ones.

As weighting always presents some difficulties, a three-entry chart can finally be made (Figure 10(right)), one for each indicator, where the composition with the best SN for each combination of weights will be highlighted. The conclusion is that the ternary blends (VII and VIII) present in almost all combinations have better SN, which indicates exceptional sustainability behaviour. This validates the other findings mentioned above.

4. Conclusions

Based on the obtained results, MTK and fly ash’s synergies significantly enhance mechanical performance and durability. So, it is feasible to produce eco-efficient blends with high volumes of cement replaced by FA by adding a reduced percentage of MTK. This type of ternary blend allows for better performance with less incorporation of FA. That is, it contributes to mitigating the reduced market availability of FA in some parts of the world.

Furthermore, the findings suggest that at younger ages, MTK could be used as a correcting or regulating factor in the mechanical performance of FA, thus minimizing the significant disadvantages of using large amounts of FA.

From the point of view of durability against chloride attack, ternary blends with MTK and FA perform exceptionally well. Achieving this goal is very challenging with current blends or conventional w/b ratios. For this reason, it seems to be an excellent solution to use in reinforced concrete structures in areas that are highly exposed to this kind of aggressive agent.

The workability is greatly affected by MTK. When combined with FA, the performance is better. However, it is essential to conduct more studies on this subject.

The incorporation of MTK could be advantageous in reducing cement in the blends while also significantly increasing the service life of reinforced concrete structures, thus, promoting a more efficient, more ecological, and more efficient construction.

FA may eventually be replaced by other similar pozzolanic additions as long as it is possible to achieve synergies with MTK.

A multi-criteria decision support method for relative sustainability (MARS-SC) is highly suitable for evaluating ternary cement blends with MTK and FA.

The findings are primarily based on laboratory conditions and may need to be further tested in field or real-world applications to confirm their generalizability.