1 Introduction

At present day, blended cements and mineral admixtures are worldwide used to meet sustainable development requirements [1, 2], and cutting CO2 emissions. Due to its abundance and economical benefit, limestone powder (LP) gains more confidence in cement [3] and concrete industries. According to PCA [4], PLC reduces project carbon footprint by 10% (see greenercement.com). It is a good option for green circular economy [5].

First, carbonate addition features in cementing network are well known today. That is, it enhances cement hydration by creating favorable nucleation sites allowing the growth of portlandite hydrate [6], silicate hydrate (S-C-H) [7], and carboaluminate formation (C3A.CaCO3.11H2O) [8] or hemicarboaluminate (C3A.1/2CaCO3.1/2Ca(OH2) [9] instead of sulfoaluminate phase [10]. Furthermore, many workers support the fact of accelerating effect of limestone on Portland cement hydration [11] by shortening the dormant period and pronouncing the second peak time [12]. Others state a good packing density within LP mixes [13, 14]. Consequently, high mechanical properties at early ages [15, 16] as resulted from hydration of tricalcium silicate hydrate (C3S) has been previously recorded [17]. According to Zhao and Zhang [18], many hydration issues of limestone cement still need more investigations.

Furthermore, PLC including inter-ground limestone may exhibit similar to better mechanical and durability properties than ordinary Portland cement (OPC) [14, 19], especially when finely ground [20, 21].

PLC cement limestone substitution is reported in the range of 10 and 25% [3, 22]. However, it has been found that blended PLC may exhibit higher performances [23, 24], especially in the presence of aluminosilicate or pozzolanic sources (slag, pozzolana, etc.) [25], due to performed hydrate dilution, leading to increased LP substitution rate [26]. Actually, alumina allows additional replacement by limestone without compromising the material properties [27]. For instance, Cetin et al. [28]. state delayed dormant period and decreased total and rate of hydration heat when slag was blended with PLC. Moreover, the influences of slag on hydration and mechanical properties have been well conducted previously [29]. Lea [30] states hydration products of slag cement as tobermorite, ettringite, C3A, and portlandite; Scrivener and Nonat [31] report that alumina slows down C-S–H growth in the dormant stage of hydration; Dudovoy et al. [32]. reveal the role of slag to perform compressive strength in blended cements; and Roy and Idorn [33] found reduced heat rate in slag-based combinations.

Besides, the positive effects of silica fume on cement-based materials is known [34] by reducing porosity [35] and improving the aggregate/paste transition zone. Also, it accelerates C3S and C3A hydration [34, 36]. Indeed, a general consensus exists on the influence of SF on shortening the dormant period and the second major peak timing [37], and reducing the total hydration heat [38]. Kadri et al. state enhanced hydration at early age of LP/SF combination due to LP and at later ages due to the delayed pozzolanic activity of SF [39]. Others report enhanced mechanical properties of Portland cement–limestone–silica fume combinations [40]. Besides, Turkel and Altuntas [41], by investigating self-consolidating repair mortars, confirm elevated mechanical properties of 5–10% limestone powder combined with 20% silica fume, similar to the finding by Guneiyisi et al., working on LP/slag/pozzolan self-consolidating concrete sets [42].

Thus, it seems that composite PLC/mineral admixture combinations may lead to improved hydration and engineering properties. Therefore, the main objective of this study is to investigate the influence of PLC/SF and PLC/LP combinations on hydration, compressive strength, sorptivity, and microstructure of different types of mortar (having equivalent W/C ratio), where variables were mineral admixture’s (SF or LP) type and mortar's type (OMs, SCMs). This by characterizing hydration heat by semi-adiabatic calorimeter, portlandite, and carboaluminate amounts through thermogravimetric analysis, compressive strength, SEM imaging, and water sorptivity.

For now, there are few investigations in the literature on the influence of slag-blended PLC combined with chemical and mineral admixtures, especially in the cases of self-consolidating concretes/mortars. The authors suggest that hydration of such complex combinations is influenced by the chemical influence of superplasticizer and the physical packing of mortars (related to mortar type). Furthermore, relating hydration with macroscopic properties of the studied mortars should lead to better knowledge of the cementitious combinations and provide some contributions on practical applications of composite PLCs.

In short, PLC/SF or PLC/LP combinations may have a performing effect on hydration and physico-chemical properties of the studied mixes.

2 Research significance

LP-based cements and concretes provide a sustainable option to cut carbon emission. Ongoing research reveals the key features to get performed mixtures: finest particle size and appropriate limestone rate. Now we are aware that more mineral admixtures allow theoretically increasing LP rate and greener material. However, blinded cements are very special materials and are not available at any time in cement plants. The specifier hence may recommend more mineral admixtures (MAs) thought concreting to have the aimed engineering properties. Works on MAs added PLC-blended self-consolidating mixtures are scarce in the technical and scientific literatures. The authors believe this study may contribute to more knowledge to understand the mechanism and properties of blended PLC/pozzolans sets and will be very useful to concrete technology.

3 Methodological investigation

3.1 Chemical reactions and cementing materials in phase diagram

In this section, an attempt was made to simplify the chemical process involved in our cementitious combinations. That is, we present here the basilar reactions and establishing ternary diagram encompassing the studied cementing combinations.

According to Brunauer and Copeland [43] and Lea [30], the reactions formulas of the main phases leading to hydrates formation are written as follows:Footnote 1

$$ \mathop {2\left( {{\text{C}}_{3} {\text{S}}} \right) }\limits_{{\text{(TricalciumSilicate)}}} + \mathop {6{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}\limits_{{\text{(Water)}}} \, = \, \mathop {{\text{C}}_{3} .{\text{S}}_{2} .{\text{H}}_{3}\raisebox{-0.5pt}{${\kern-12.5pt^{-\!\!-\!\!-}}$} + 3{\text{Ca}}\left( {{\text{OH}}} \right)_{2} }\limits_{{\text{(Tobermorite) (Portlandite)}}} $$
(1)
$$ \mathop {2\left( {{\text{C}}_{2} {\text{S}}} \right)~}\limits_{{{\text{(DicalciumSilicate)}}}} + {\text{ }}\mathop {4{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}\limits_{{{\text{(Water)}}}} = ~\mathop {{\text{C}}_{3} .{\text{S}}_{2} .{\text{H}}_{3}\raisebox{-0.5pt}{${\kern-12.5pt^{-\!\!-\!\!-}}$} + {\text{Ca}}\left( {{\text{OH}}} \right)_{2} }\limits_{{{\text{(Tobermorite) (Portlandite)}}}} $$
(2)
$$ \mathop {{\text{C}}_{4} {\text{AF }} + {\text{ }}10{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}\limits_{{{\text{(Tetracalcium}}\;{\text{Aluminoferrite)}}}} ~ = ~\mathop {~2{\text{Ca}}\left( {{\text{OH}}} \right)_{2} }\limits_{{({\text{Water}})\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;({\text{Portlandite}})}} + ~\mathop {{\text{C}}_{6} .{\text{A}}.{\text{F}}.{{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}_{{12}} }\limits_{{({\text{Calcium Aluminoferrite Hydrate}})}} , $$
(3)
$$ \mathop {{\text{C}}_{3} {\text{A}}~ + {12{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}} + {\text{ Ca}}\left( {{\text{OH}}} \right)_{2} ~}\limits_{{{\text{(Tricalcium Aluminate)(Water) (Portlandite}})}} = ~\mathop {{\text{C}}_{3} .{\text{A}}.{\text{Ca}}\left( {{\text{OH}}} \right)_{2} .{{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}_{{12}} }\limits_{{{\text{(Tetracalcium Aluminate Hydrate)}}}} $$
(4)
$$ \mathop {{\text{C}}_{3} {\text{A}}}\limits_{{{\text{(Tricalcium Aluminate)}}}} ~~~~~~~ + ~~\mathop {~10{{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}}\limits_{{{\text{(Water)}}}} ~~~ + ~~\mathop {~{\text{Ca}}.\check{S}.{{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}_{2} }\limits_{{{\text{(Gypsum)}}}} = ~~~\mathop {{\text{C}}_{3} .{\text{A}}.{\text{Ca}}\check{S}.{{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}_{{12}} }\limits_{{{\text{(Calcium Monosulfoaluminate )}}}} $$
(5)

After Gerry [44], in the presence of calcium sulfate, the product of hydration is a sulfoaluminate, C3A.3CŠ.H31–32, known as the mineral ettringite. It is a prismatic or acicular crystal shaped with a hexagonal cross section.

$$ \mathop {{\text{C}}_{3} {\text{A}}~~~}\limits_{{{\text{(Tricalcium Aluminate)}}}} ~~~~~~ + ~~~~~~\mathop {10{{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}~}\limits_{{{\text{(Water)}}}} ~~~~~ + ~~~~\mathop {3{\text{C}}.\check{S}.{{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}_{2} ~~}\limits_{{{\text{(Calcium Sulfate)}}}} ~~ = ~~~\mathop {~{\text{C}}_{3} {\text{A}}.3{\text{C}}\check{S}.{{\text{H}}\raisebox{-0.5pt}{${\kern-8.5pt^{-\!\!-\!\!-}}$}}_{{32}} }\limits_{{({\text{Ettringite }})}} . $$
(6)

In this study, mortars include four cementitious combinations (related chemical phase calculations are given in Table 1). Thus, when placed in a ternary diagram (Fig. 1), we can expect the nature of hydration products: With slag PLC (Cb2), more portlandite, AFm, and some AFt; in combination with Cb3, more portlandite and formation of AFt (ettringite) rather than AFm; on Cb4, formation of C–S–H with low S/C ratio, C–A–S–H compounds, and silica gel.

Table 1 Calculations of CaO, SiO2, and AlO3 contents for cementitious combinations (based on Fig. 1)
Full size table
Fig. 1
figure 1

Ternary diagram CaO–SiO2–AlO3 for cementitious combinations

Full size image

Furthermore, the presence of inter-ground limestone is reported to form carboaluminate hydrate forms (as presented in the introduction).

3.2 Materials

For this test program, an ordinary Portland cement OPC (CEM I 52.5 N CE CP2 NF) was used in the reference mortar; inter-ground slag PLC (CEM II/B-M(LL-S) 42.5 N CE CP) cement was used for the five other mortars. The cements meet the requirements of the EN 197-1 European standard [45] (equivalent to ASTM C105); a well-graded 0/4 mm (# 8) fine aggregate was also used, with a silico-calcareous petrographic nature, meeting XP P18-545 standard requirements, and having a finesse modulus of 2.82. Furthermore, a Betocarb HP EB limestone powder, classed as ‘A category’ according to NF P 18-508 standard, was implicated (average grain size da = 20 μm = 7.8 10–4 in.). Besides this, condensed silica fume (Condensil S95 DS) was also used. The main chemical and physical characteristics of the used cementing materials are presented in Table 2. A new generation of modified polycarboxylates superplasticizer has been used (Cimfluid Allegro 3010). Its main characteristics are presented in Table 3.

Table 2 Physical and chemical characteristics of cementing materials (by mass) – data from the producer technical sheets
Full size table
Table 3 The main characteristics of the superplasticizer (from:http://fra.sika.com/fr/solutions-produits.html)
Full size table

3.3 Mixture proportions, fabrication, and curing

Four categories of mortars were considered in this investigation [Ordinary mortars (OMs) and Self-consolidating mortars (SCMs)], as their nomenclature is given in Table 4. We notice that the six mortars have been extracted from the mother concretes published previously by the authors [46], and adjusted then, to have appropriate fluidity for each type. The initial W/C ratio was 0.5, and the effective W/C was optimized as given in Table 4. While Oref mixture was based on OPC, Oplc was based on slag PLC. The ordinary mortars were mineral admixture free and served as references. SplcL and SplcS self-consolidating mortars were based on PLC and, respectively, on limestone powder and silica fume.

Table 4 1m3 Mixture composition of mortars
Full size table

Ordinary mortars having a plastic consistency were fabricated and vibrated as per NF EN 196–1 standard [47], but additional four minutes waiting and one minute mixing at high speed were applied to avoid possible flash set [48, 49]; self-consolidating mortars were designed to have a target mortar spread of 240–260 mm (9.44–10.23 in.) after Sedran [50] and placed in molds without vibration.

For compressive strength, steel molds 40 × 40 × 160 mm (1.57 × 1.57 × 6.29 in.) were utilized. Furthermore, 40 × 40 mm (1.57 × 1.57 in.) cube molds were used for capillary absorption test. Samples were extruded from molds after one day and placed immediately underwater curing conditions until testing dates.

3.4 Test procedure

Specimens were brought out from water, and were air-dried in laboratory conditions for 24 h. After that, compressive strength of mortars was determined by 3, 7, 28, 56, and 91 days according to EN 196-1 standard [47].

Mortars’ capillary absorption test was carried out in accordance to previous works [51]. Therefore, 40 × 40 mm (1.57 × 1.57 in.) cubes were adopted. First, mortar samples were cooled in room temperature for one week period to control their humidity rate. Then, they were oven-dried at 105 °C until a constant mass and placed on non-corrosive support in a 1 cm (0.39 in.) water pan filled. We note that the specimen’s depth in contact with water was 5 mm (0.19 in.) (Fig. 2). For a unidirectional flow, the samples’ lateral area was covered with adhesive aluminum. Finally, by using a precision weighing balance the mass of specimens was recorded to 0, 5, 10, 20, 30, 60, 180, 360, and 1440 min, and the amount of the absorbed water was calculated. The capillary absorption coefficient (k) is obtained by the following formula:

$$\frac{Q}{A} = k\sqrt {t,}$$
(7)

where Q = the amount of the adsorbed water in (cm3); A = the cross section of the mortar sample in contact with water (cm2); t = measuring timing (s); and k = capillary absorption of the tested sample (cm/s1/2). The coefficient (k) was graphically determined by plotting Q/A against \(\sqrt{t}\) and calculated from the resulted slope of the.

Fig. 2
figure 2

Water sorptivity measurement of mortars’ samples

Full size image

Hydration heat test was conducted as per EN 196-9 standard [52], using the semi-adiabatic calorimeter. This method consists of measuring the heat released during cement hydration using a thermally isolated Dewar flask. The test is carried out at 20C° monitored in air-conditioned room. Thus, temperature development during the period test (6 days) is defined as the difference between the temperatures of the tested mortar and inert mortar matured for a period superior than 3 moths. The mortar’s temperature heat is the sum of the accumulated heat in the calorimeter, and the heat dispersed in the environment. At a time t, the hydration heat Q measured by gram of binder is obtained by applying the formula:

$$Q = \frac{C}{Mc}\Delta \theta + \frac{1}{Mc}\mathop \int \limits_{0}^{1} \alpha \Delta \theta {\text{d}}t,$$
(8)

where C is the apparatus total thermal capacity (J/°C), Mc is the cement masse or binder mass (g), \(\Delta \theta\) the difference in heating between mortar sample andthe ambientenvironment (°C), and \(\alpha\) represents the total calorimeter thermal loss coefficient (J/h °C).

The rate of heat evolution is calculated using the following formula:

$$R_{{{\text{heat}}}} = \frac{{\left( {Qi - 1 - Qi} \right)}}{{\left( {ti - 1 - ti} \right)}}.$$
(9)

For thermal analysis, a 100 mg powder specimen was taken for each mortar and then heated in the thermal analyzer (STA 449 F1 Jupiter) with a 20 °C/min rate. The objective of such test is to quantify the hydration of cement mortars at medium and later ages. This by quantifying the portlandite (P) quantity developed through hydration and measuring weight loss due to decarbonation served as indicator of the unreacted limestone amount. Surely data obtained by this test have a great significance to explain the physical properties of the tested mortars (compressive strength and water sorptivity).

Next, portlandite amount and decarbonation weight loss (in 3,28 and 91 days) were calculated according to Ref. [53].

$$p\left( \% \right) = \frac{W550 - W430}{{W1000}} \times 100,$$
(10)
$$C\left( \% \right) = \frac{W1000 - W550}{{W1000}} \times 100,$$
(11)
$$P\left( \% \right) = 4.11.p + 1.68.C,$$
(12)

where p is the weight loss in TG curve at 430–550 °C, C is the weight loss at 550–1000 °C; and P is the portlandite amount in the sample.

To assess the microstructural features, we used fractured specimens from mortars tested under the (JEOL 35CF) SEM, first, placed directly in an evaporator and maintained under high vacuum overnight, and then, gold coated in evaporating apparatus just before getting the observations.

4 Experimental results and discussion

4.1 Compressive strength

Figure 3 shows the compressive strength results of mortar samples until 91 days of hydration age.

Fig. 3
figure 3

Compressive strength of mortar samples

Full size image

On the influence of mortar type, it is obvious that SCMs exhibit equal to better compressive strength than OMs samples. This can be explained by the highest paste volume in SCMs and the beneficial dispersing effect of superplasticizer. We can observe too that 30 ℅ of the limestone powder in SplcL contributes better than 7 ℅ silica fume in SplcS on compressive strength enhancement. It can be due to the good packing of the involved cementitious materials used in SplcL mortar. While the added limestone powder has better physical effect, inter-ground limestone with clinker and slag in slag PLC cement has developed a chemical effect through the formed carboaluminate hydrate and the creation of additional nuclei sites.

By comparing OPC-based Oref and slag PLC-based Oplc, equivalent compressive strength can be seen at 3 and 14 days. However, at 7 and 91 days, PLC mortars exhibited 36% better performance than OPC mortars. On one hand, the 7 days highest values of PLC mortar (27 MPa (3.91 Psi) compared to only 17 MPa (2.46 Psi) for OPC mortar) may be attributed to the chemical influence of inter-grounded LP (relatively accelerated hydration at early age as presented below in hydration heat section). One the other hand, the enhanced later compressive strength of Oplc can be due to the pozzolanic reaction provided by the inter-grounded slag.

For the optimal mixture SplcL, the authors believe that added and blended limestone powder may have a synergistic effect when mixed with pozzolanic species. In that case a good compaction is guaranteed by physical packing and pronounced chemical reaction leading to more silica gel formation.

4.2 Water sorptivity

Water sorptivity test results of all mortar specimens are given in Table 5. As expected, mortar sorptivity in this study follows a similar pattern of compressive strength values for all mortar types. Thereby, it is obvious that Oplc behaves better than Oref and SplcL mixtures was the best. The lower recorded sorptivity in SplcL mortar can be explained by the enhanced microstructure due to the cementitious combination of PLC and limestone powder [54], and the good adherence formed in the paste aggregate bond. Nevertheless, SplcS showed relatively higher sorptivity than Oref and SplcL. As discussed above, it seems that the incorporated amount of silica fume in this combination (7℅) was fewer than the optimum claimed in literature (10–15℅). By other words, a fewer quantity of SF cannot be totally dispersed in the mix leading to less involved materials in the pozzolanic reaction.

Table 5 Sorptivity coefficients of mortar samples
Full size table

4.3 Early age hydration by semi-adiabatic calorimetry

Figures 4 and 5 represent hydration heat development within 72 h and the rate of heat release up to 28 h, respectively. The first noticeable observation is the lowest heat and rate of hydration of PLC-based combinations compared to the OPC reference mortar (40% lower), which meant that the presence of mineral admixtures may lead to lesser heat release, and confirms that clinker amount in the binder is the main covering factor. Furthermore, for all mortar types, limestone-based combinations exhibited lower heat release than silica fume mixes. It can be explained the by the influence of the high SF pozzolanicity compared to the low reactivity of LP [37, 55]. It is evident that heat and rate of hydration in the studied mortars can be classified as OMs > SCMs. Again, this confirms that the amount of involved clinker and binder quantity are the most significant parameters regarding the heat of hydration.

Fig. 4
figure 4

Hydration heat development of mortar samples in the semi-adiabatic calorimeter within 72 h (NF EN 196–9)

Full size image
Fig. 5
figure 5

Rate of heat release versus hydration time for mortar samples

Full size image

Besides, in Fig. 5, by comparing the rate of heat release in the second major peak, it can be seen that Oplc has 43 ℅ lower values than Oref sample. In contrast, the acceleration stage in Oplc mortar starts at only one hour versus 2 h recorded in Oref mortar. This makes a support to the existing findings by other researchers on the influence of carbonate additions on accelerating the rate of hydration at early ages [56, 57]. However, in our case when slag PLC was used, the timing of the end of acceleration stage is not shortened compared to OPC mortar, as reported previously [58, 59]. We recorded equivalent values for both Oref and Oplc mortar (11 h). This fact may be attributed to the influence of slag combined with limestone in PLC fabrication, which, by his retarding effect on hydration caused by its glassy structure [33], may neutralize limestone accelerating effect.

For SCMs mortar samples, delayed dormant period (5 h for SplcL and 6 h for SplcS) and 2nd peak timing (12 h for SplcL and 13 h for SplcS) were registered. Those results controvert some literature findings agreeing the accelerated hydration with limestone or silica fume [48] and meet some scholars' results [60]. In fact, the recorded delayed dormant period and the 2nd peak timing can be explained by the cementitious combinations, including clinker, slag, limestone, or silica fume, and for the use of superplasticizer.

By relating compressive strength and rate of heat release results, one can conclude that enhancing compressive strength can be related to the longer delayed acceleration period. This conclusion makes sense when considering the nature of hydration products formed at this stage. Indeed, during that period the growth of S-C-H needles and silicate-bearing products is accelerated [61, 62].

4.4 Thermogravimetric analysis TGA

Portlandite and decarbonation loss by TG analysis for mortar samples up to 91 days are given in Table 6. With the presence of limestone powder, the test recorded high masse decarbonation loss and portlandite amount for SplcL, RplcL, and Oplc. This fact is well known [19] and explained by the presence of CaO as discussed in our ternary diagram.

Table 6 Portlandite and decarbonation masse loses by TG analysis for mortar samples
Full size table

In SF added mortars (SplcS), the portlandite hydrate seems to be consummated in later ages (28 or 91 days) due to the pozzolanic reaction monitored by silica fume and inter-ground slag. It is important to note that the co-existence of slag and silica fume leads to inconsistent portlandite consumption, as explained by some authors by the competition among mineral admixtures in the exhaustion of the existing portlandite [63].

Figure 6 presents the derivative thermogravimetric curves (DTG) of all mortar samples at 3 days of hydration. Doing this task, we focused on carboaluminate characterization, detected approximately at 180 °C in the DTG curve after some minutes. As found in previous works [64], carboaluminate peak in Oplc specimen was observed, but in SplcL mix a small peak was formed. The authors explain this by the unreacted added limestone due to its coarser particle size. Thus, in this case we believe limestone powder has a physical rather than chemical effect.

Fig. 6
figure 6

Carboaluminate characterization by DTG curves

Full size image

4.5 SEM observations

Figure 7 (1–4) shows SEM observations of the mortars’ fractured specimens (Oref, Oplc, SplcL, and SplcS) at 28 days

Fig. 7
figure 7

a SEM images of mortar fracture specimens of Oref. b SEM images of mortar fracture specimens of Oplc. c SEM images of mortar fracture specimens of SplcL. d SEM images of mortar fracture specimens of SplcS

Full size image

First, for Oref sample, the presence of ettringite needles 5 μm (1.9 × 10–4 in.) was observed. In fact, a poor crystallized reticular network of C–S–H was dominant [65] and occupies the majority of the paste. The presence of some unreacted clinker particles can be seen too. Due to its coarser size (> 10 μm = 3.9 × 10–4 in.), the portlandite hydrate may be partially hidden by smaller particles of inner C–S–H or ettringite. We note that the unreacted phases exist in all stages of hydration and may remain in this state with the narrowness of allowed space [66]. However, when morphological SEM observations are carried out, those unreacted materials are generally covered by C–S–H, ettringite or portlandite.

Second, for Oplc mortar, as expected and detected by TG results, the presence of portlandite due to carbonate excess in the mix was well observed. Besides, more refined C–S–H was observed in the monograph, which supports literature findings approving the growth hypothesis in the presence of carbonate [7]. Furthermore, smaller amounts of ettringite needles were remarked. This can be explained by the preferential formation of carboaluminate hydrate within the deceleration stage of hydration versus ettringite formation [67].

Third, for SplcL sample, the presence of plate-shaped portlandite and refined C–S–H hydrate was observed in the monograph. However, no ettringite was detected, may be to its conversion at this age (28 days) to monosulfate or monocarbonate hydrate. Besides, we can see the denser microstructure that explains mechanical and water sorptivity achieved performance of this mortar.

Finally, for SplcS mortar sample, as detected by TG curves portlandite plates were observed in the specimen, which means that SF pozzolanic reaction at this stage does not consume the totality of portlandite. We interpreted in compressive strength and hydration degree section the lowest performance in this combination (PLC + 7 ℅ SF) by the insufficient amount of silica fume on the mix, which does not lead to more announced pozzolanic reaction. Furthermore, the presence of C–S–H covering unreacted clinker was observed. It is assumed in previous works that with the presence of SF a low form of C–S–H is formed (having a C/S ratio about 0.8) [31], and lower portlandite forms nearing the interfacial area [68]. Well-crystallized ettringite needles were also detected making like bridges between cement grains.

5 Summary and concluding remarks

In this study some physical, hydration, and microstructural investigations were carried out on two types of mortars. Based on the results of the investigation, the following conclusions are drawn.

Slag/limestone combination in PLC composite leads to enhanced properties of mortars both at early and later stages.

  1. 1.

    Self-consolidating mortars exhibited 36% better compressive strength than OPC mortar, but lesser heat release and hydration degree. This meant that the mortar physical structure influences strongly the hydration and the resulted macro-properties.

  2. 2.

    The included limestone in slag PLC leads to good mechanical properties especially in SCMs mixes due to the dispersing effect of the superplasticizer. However, when added to mortar it has a little contribution to enhance neither heat release nor hydration degree.

3. Combining blended and inter-ground limestone in the cementitious system should improve its packing (blended PLC), formation of more carboaluminate phase (inter-ground PLC), and creating more nucleation site as template for S-C-H growth.

4. In SCMs mixes, the inter-ground limestone and slag present in the mix compete to react with the 7 ℅ of silica fume. In this case, a higher amount of silica fume is needed to obtain higher reactivity and more mixture adhesion.

5. The amounts of clinker and superplasticizer are significant parameters influencing the hydration mechanism. Increasing clinker leads to increased heat release and hydration degree but does not enhance the compressive strength. Adding more superplasticizer has beneficial effect on rheological properties and compressive strength, but extended dormant period should be occurred.

The study findings state the beneficial influence of limestone/pozzolanic admixtures combination, in production and use stages, to obtain more enhanced engineering properties of cementitious materials. The added limestone powder with PLC and/or silica fume had a good synergistic effect and led to better performance. Hence, the studied mixtures are greener, stronger, and cheaper.

It is suggested therefore to carry out in-depth investigations on the influence of added limestone to composite PLCs, assessing other amounts of SF substitution and measuring the reactivity of the combined cementitious materials.