1 Introduction

The phase evolution in Portland cement involves very complicated thermochemical reactions occurring during clinkerization which is energy intensive high temperature operation (1400–1450 °C). The raw materials used for Portland cement contains several minor oxides (i.e., MgO, Na2O, TiO2, ZnO, MnO, K2O, P2O5) along with main oxide components such as CaO, Al2O3 SiO2 and Fe2O3 (C, A, S and F); SO3 is also infused from combustion fuel in cement during processing [1]. Mineralizers are sometimes added to raw mix in an effort to reduce the process temperature to achieve cost effective production [2,3,4]. These minor oxides and mineralizers have significant effect on phase formation.

The phases formed during clinkerization process are alite (C3S), belite (C2S), aluminate (C3A) and ferrite based solid solution (C4AF) via reaction among CaO, SiO2, Fe2O3 and Al2O3 [5]. These phases are very much critical for the development of strength of cement during hydration. C3S and C2S are responsible for the development of early and later strength, respectively, whereas C3A has a very important role in controlling setting time. C4AF plays the role of colourant in cement and flux for the reduction of process temperature. The chemical and mineralogical names along with the chemical formula and their range of occurrence in Portland cement are given in Table 1. The major phases formed during clinkerization is normally calculated by using well-known Bogue’s equation (Eqs. 1–4) [5] to ensure the quality of final product.

$${\text{C}}_{{3}} {\text{S}} = {4}.0{71}0 \times {\text{CaO}} - {7}.{6}0{24} \times {\text{SiO}}_{{2}} - {1}.{4297} \times {\text{Fe}}_{{2}} {\text{O}}_{{3}} - {6}.{7187} \times {\text{Al}}_{{2}} {\text{O}}_{{3}} - { 2}.{852} \times {\text{SO}}_{{3}}$$
(1)
$${\text{C}}_{{2}} {\text{S }} = {2}.{867} \times {\text{SiO}}_{{2}} - 0.{7544} \times {\text{C}}_{{3}} {\text{S}}$$
(2)
$${\text{C}}_{{3}} {\text{A}} = {2}.{65}0{4} \times {\text{Al}}_{{2}} {\text{O}}_{{3}} - {1}.{692}0 \times {\text{Fe}}_{{2}} {\text{O}}_{{3}}$$
(3)
$${\text{C}}_{{4}} {\text{AF}} = {3}.0{432} \times {\text{Fe}}_{{2}} {\text{O}}_{{3}}$$
(4)
Table 1 Range of major phases seen in Portland cement clinker
Full size table

It is very important to have control over processing to maintain all the phases in a defined range for achieving desirable properties in final product. The effect of all the major oxides on phase formation is very much critical. The authors have already shown the thermodynamic analysis of effect of major oxides on phase formation in a previous report [6]. The aim of the current work is to carry out a systematic thermodynamic study on the effect of minor oxides and mineralizers on phase formations.

The minor oxides and mineralizers have been reported to have significant effects on cement properties [3, 7, 8]. The controlled presence of these constituents in raw materials is beneficial in aiding phase formation, reaction kinetics, refractory life, fuel efficiency and hydration during application [7, 9,10,11].

As cement clinkerization process requires to form sufficient ionic liquid (25–30%) for desirable phase formation, the temperature requirement is very high (1400–1450 °C) for pure system. This temperature is brought down with the addition of mineralizer (i.e., CaF2, AlF3, MgSiF6, Na2SiF6, CaCl2, ZnO, CaSO4 etc.[12, 13]. Thus, the presence of mineralizer helps in achieving cost effectiveness of the process. Therefore, it is very important to study the interactions of these minor constituents with the major oxides during clinkerization process in order to understand their ultimate effect on phase formation.

A complete thermodynamic study is a prime requirement of this system to have a complete understanding on the phase formation. However, available binary or ternary phase diagrams are not enough to study such complex multicomponent systems (> 5 component) since it cannot consider the effect of the entire component together. Therefore, in the present study computational thermodynamics based CALPHAD (CALculation of PHAse Diagram) [14] method has been extensively used to study the phase evaluation at different thermodynamic conditions. Till date, there is no report available on such a study using CALPHAD.

CALPHAD is advanced computational mathematics based thermodynamic calculation method to calculate or predict phase equilibria [15, 16]. It uses available experimental or theoretical thermodynamic data of lower order systems and extrapolates them to higher order systems using various well established Gibbs free energy models at a particular temperature (T) and pressure (P). Finally, Gibbs free energy minimization technique is applied to available phases to reveal the stable phases at a particular equilibrium condition (T and P) [16]. Nowadays many commercial and open source software such as, FactSage [17], Thermo-Calc [18] and MTDATA [19] are available to perform the calculation.

2 Methodology

In current investigation, all the thermodynamic calculations were carried out using FactSage 7.3 software with FToxid database [17]. To validate the FactSage data first, three different raw mixes were prepared by mixing and grinding of Limestone, Bauxite, Laterite and Lithomarge in proper proportions. Then, the raw mix was fired to produce clinker in a platinum crucible using Carbolite Down-Hearth Furnace at 1450 °C for 1 h followed by air quenching at 1300 °C. The clinkers were analyzed by Bruker S8-TIGER X-ray Fluorescence (XRF) to study the chemical composition. Finally, it was also analyzed by using Bruker’s D8 advanced diffractometer (XRD) with \(Cu-{K}_{\alpha }\) radiation having wavelength (\(\lambda\)) 0.154056 nm. Rietveld analysis was carried out using TOPAS XRD software [20]. The CALPHAD result was then compared with XRD data and Bogue’s calculations [21].

To see the phase evolution, phase evaluation diagram was created using FactSage for the XRF composition in a temperature range 700–1500 °C using composition of Sample-3. The microstructure of the same clinker was also studied using Optical microscopy and BSE-SEM imaging with point EDS and EDS mapping.

To study the effect of individual minor oxides, a reference standard composition of C–A–S–F–M system was considered (given in Table 2) and minor oxides were added in the C–A–S–F–M system with various proportions to study their effect within a particular temperature range. All the minor oxides were varied within a range given in the Table 3. Similarly, the effects of important mineralizers (CaF2, AlF3, MgSiF6, Na2SiF6, CaCl2, ZnO and CaSO4) on phase formation and clinkerization temperature have been calculated using equilibrium module of FactSage software. Finally, OriginLab 2017 [22] software was used to draw all the plots using the data obtained by CALPHAD calculation.

Table 2 Baseline raw mix composition of major oxides used to study effect of minor oxides and mineralizers
Full size table
Table 3 Range of oxides considered to study the effect on phase formation
Full size table

3 Results and discussion

Table 4 shows the chemical analysis result of the prepared clinkers obtained by the XRF-chemical analysis. Table 5 displays the percentage of different phases in three clinker samples obtained by different methods such as, CALPHAD calculation, Bogue’s calculation and XRD-Rietveld method. Fitting parameters of XRD-Rietveld analysis (Rwp and Gof) are also given in Table 5. To calculate the stable phases using CALPHAD, C3S, C2S, CaO and MgO was calculated at the quenching temperature (1300 °C) and C3A and C4AF was calculated at 700 °C since they are not stable at 1300 °C. All the solid solutions of CaO–Al2O3–Fe2O3 that are stable at 700 °C are considered as C4AF. It can be observed that C3S prediction in CALPHAD method is very close to the XRD-Rietveld data both for Sample-1, Sample-2 and Sample-3. Prediction of Bogue equation is also very close to the XRD result. For C2S prediction, CALPHAD result showed good match with the XRD-Rietveld data for all three samples. But, C2S predicted by Bogue’s equation is 4–5% high compared to actual XRD-Rietveld data. C3A predicted by CALPHAD is also very close to the XRD result but, C3A predicted by Bogue’s equation is much higher in Sample-2 and Sample-3 compared to the XRD result. C4AF predicted by the CALPHAD is also very close to the XRD-Rietveld result but C4AF predicted by the CALPHAD is 2–5% less compared to the XRD-Rietveld result. Above result shows that, CALPHAD prediction data is very close to the actual experimental data which is not always true for Bouge’s equation since it does not consider the effects of all the oxides and temperature [5].

Table 4 Chemical analysis (XRF) of clinker synthesized in lab
Full size table
Table 5 Phase analysis result of clinker samples obtained by different method
Full size table

From the above discussion, it can be concluded that the CALPHAD can precisely predict the cement phases at a given condition. In addition, CALPHAD method has a spin off benefit of calculating phase at any equilibrium conditions (temperature, pressure) [15]. So, CALPHAD can be used to study the phase evolution with increasing temperature and to understand the role of different important minor oxides in phase formation.

Figure 1 displays the temperature vs phase fraction diagram generated by CALPHAD method in the temperature range of 700–1500 °C for the composition of Sample-3 given in Table 2. Figure 1 shows during clinkerization temperature (1400–1450 °C), the stable phases are C3S, C2S, unreacted MgO and ionic liquid. During cooling, ionic liquid reduces while C2S and C3S phase increases till 1300 °C. Below 1300 °C, C3S is not stable and it dissociates into C2S and CaO as shown in Fig. 1. Below 1280–1290°C, ionic liquid phase separation occurs and another solid solution phase i.e. C4AF starts forming. Below 1050–1100 °C, C3A crystallizes out from remaining ionic liquid and finally remaining liquid solidifies during cooling depending on the quenching rate.

Fig. 1
figure 1

Temperature vs. phase evolution diagram for Sample-3

Full size image

This diagram is very important to understand the phase formation route during heating in C–A–S–F–M system. It also shows that the quenching is necessary to get C3S phase after cooling as C3S phase is stable above 1300 °C. This graph can also be useful in determining the maximum burning zone temperate of the rotary kiln to operate the same safely within a maximum allowed ionic liquid percentage. For an instance, if the permissible limit of ionic liquid in a rotary kiln is 28 wt%, the maximum furnace temperature should not exceed 1430 °C for the given raw mix composition of Sample-3 as suggested by the Fig. 1. Similarly, maximum permissible calciner temperature can be determined from this graph to avoid ionic liquid phase formation in the calciner.

The Optical microscopy image of the above clinker (sample-3) along with the XRD patter is also shown in Fig. 2. It shows that the microstructure constitutes of three major phases, C3S phase with hexagonal grain, C2S phase with rounded grain and bright C4AF matrix. The result of XRD analysis is already shown in Table 5. To identify the phases in microstructure, EDS mapping and Point EDS was carried out. Elemental distribution in the microstate along with BSE image is shown in Fig. 3. It shows that, bright matrix regions are rich in Al and Fe which indicates that it is C4AF. Periclase are present as MgO cluster which appears to be dark in SEM image. Major regions are populated with Ca and Si which indicates C3S and C2S consist of major proportions. Average compositions of different phase obtained by EDS point analysis are also given in Table 6.

Fig. 2
figure 2

a Optical microscopy image and b XRD-Rietveld analysis image of Clinker Sample-3

Full size image
Fig. 3
figure 3

BSE-SEM images, EDS-mapping and Elemental distribution map of Clinker Sample-3

Full size image
Table 6 Average composition of different phases obtained by EDS point analysis
Full size table

Figure 4 shows the effect of SO3 on the major phase formation of C–A–S–F–M system. SO3 increases the C2S and decreases C3S above 1300 °C as shown in Figs. 4a, b. But, above 1450 °C, C2S decreases suddenly as it dissolves into the ionic liquid. The reduction of C2S above 1450 °C directly depends on the SO3 content in the composition as shown in the Fig. 4. With increasing SO3 in the composition, the rate of dissolution increases and the C2S decreases rapidly. A minute presence of SO3 (∼ 0.5%) can reduce the ionic liquid formation temperature by ∼ 20 °C. At lower temperature, SO3 does not have much effect on ionic liquid formation. However, at very higher temperature (≥ 1450 °C) SO3 significantly increases the ionic liquid content by assisting the dissolution of C2S into the ionic liquid as shown in Fig. 4c. So, if SO3 is high (> 1.5 wt%) in the raw mix, the maximum kiln temperature should be less than 1450 °C to avoid the excess liquid formation in the kiln in order to avoid uncontrolled liquid formation in kiln. To understand the influence of SO3 in increasing C2S and lowering C3S, the effect of SO3 on CaO–SiO2 binary diagram was simulated which is illustrated in Fig. 5. It shows the overlapping of two CaO–SiO2 binary phase diagrams, one is without SO3 (black) and other one is with 2% SO3 (red). Figure 5b is the magnified view of the area of interest of Fig. 5a to present the effect of adding 2% SO3 on C3S and C2S formation in the temperature range 1000–1800 °C. As it demonstrated, with the addition of SO3, the equilibrium line PQ shifts to right P’Q’. Now the effect of shifting the line can be understood better by applying the Liver’s rule in the phase diagram. When, CaO/(CaO + SiO2) = 0.7, applying Liver’s rule, the wt. percent of C3S and C2S without using SO3 is represented by \(\frac{OR}{RS}\) and \(\frac{OS}{RS}\) respectively. For the addition of 2% SO3, the percentages of C3S and C2S are \(\frac{{O}^{^{\prime}}{R}^{^{\prime}}}{{R}^{^{\prime}}{S}^{^{\prime}}}\) and \(\frac{{O}^{^{\prime}}{S}^{^{\prime}}}{{R}^{^{\prime}}{S}^{^{\prime}}}\) respectively. The addition of 2% SO3 resulted decreases of C3S and increase of C2S since, \(\frac{{O}^{^{\prime}}{R}^{^{\prime}}}{{R}^{^{\prime}}{S}^{^{\prime}}}\) < \(\frac{OR}{RS}\) and \(\frac{{O}^{^{\prime}}{S}^{^{\prime}}}{{R}^{^{\prime}}{S}^{^{\prime}}}\) > \(\frac{OS}{RS}\). The effect of SO3 on C3S and C2S obtained by CALPHAD agrees with the finding of W. Gutt et al. [23] and Horkoss et al. [24] as they also have reported that SO3 increases C2S and decreases C3S in the clinker. Li et al. [25] has also reported similar observation on C3S and C2S.

Fig. 4
figure 4

Effect of SO3 on phase formation in C–A–S–F–M system in temperature range of 1100 °C to 1500 °C; a effect on C2S, b effect on C3S and c effect on liquid phase

Full size image
Fig. 5
figure 5

a Effect of addition of 2 wt. % SO3 on the CaO–SiO2 phase diagram. b Magnified view of effect of SO3 addition

Full size image

Figure 6 displays the effect of Na2O on phase evolution of C–A–S–F–M system. It also has a negative effect on decreasing C2S above 1290 °C. With increasing Na2O, C3S increases linearly as shown in Fig. 6b. But, at higher temperature (> 1450 °C), C3S dissolves to form ionic liquid and C3S decreases. The dissolution temperature and amount of dissolution also depends on Na2O content in the ionic liquid as shown in Fig. 6b. With increasing Na2O, the dissolution temperature of C3S decreases. The rate of dissolution also increases with increasing Na2O in cement raw mix. The proportion of ionic liquid phase which forms at low temperature, increases with increasing the proportion of Na2O. Above 1450 °C, ionic liquid content suddenly increases due to dissolution of C3S as described earlier. Figure 7a, b shows the effect of increasing Na2O on CaO–SiO2 binary phase diagram. Black and red colour curve represent binary diagram without Na2O and with 0.5% Na2O. Applying Liver rule, it can be seen that, with addition of 0.5% Na2O, C3S increases and C2S decreases as, \(\frac{{O}^{^{\prime}}{R}^{^{\prime}}}{{R}^{^{\prime}}{S}^{^{\prime}}}\) > \(\frac{OR}{RS}\) and \(\frac{{O}^{^{\prime}}{S}^{^{\prime}}}{{R}^{^{\prime}}{S}^{^{\prime}}}\) < \(\frac{OS}{RS}\). The CALPHAD results also matches with the experimental results as, it been proved earlier by experimental investigations that Na2O increases C3S and decreases C2S [26,27,28].

Fig. 6
figure 6

Effect of Na2O on phase formation in C–A–S–F–M system in temperature range of 1100 °C to 1500 °C; a effect on C2S, b effect on C3S and c effect on liquid phase

Full size image
Fig. 7
figure 7

a Effect of addition of 0.5wt% Na2O on the CaO–SiO2 system, b magnified view of fig. a

Full size image

Figure 8a–c describes the influence of K2O in phase formation of Portland cement. With increasing K2O in C–A–S–F–M system, it has a significant effect in decreasing C2S and increase of ionic liquid as shown in the Fig. 8a and c. Minute addition (0.05 wt%) of K2O reduces the ionic liquid formation temperature by 10 − 15 °C. K2O also has a marginal effect in increasing C3S. The root cause of increasing C3S and decreasing C2S can be seen form the binary diagram of CaO–SiO2 as shown in Fig. 9a and b Black and red line represents without addition of K2O and red line represents with addition of 0.5% K2O. As the Fig. 9b shows the effect of K2O on CaO–SiO2 binary phase diagram is very much similar to the effect of Na2O. It also increases C3S and decreases C2S as \(\frac{{O}^{^{\prime}}{R}^{^{\prime}}}{{R}^{^{\prime}}{S}^{^{\prime}}}\) > \(\frac{OR}{RS}\) and \(\frac{{O}^{^{\prime}}{S}^{^{\prime}}}{{R}^{^{\prime}}{S}^{^{\prime}}}\) < \(\frac{OS}{RS}\). Although the effect of K2O on increasing C3S and decreasing C2S is not as strong as Na2O as, the shift of PQ line to P’Q’ is less in case of 0.5% K2O addition as compared with 0.5% Na2O addition.

Fig. 8
figure 8

Effect of K2O on phase formation in C–A–S–F–M system in temperature range of 1100 °C to 1500 °C; a effect on C2S, b effect on C3S and c effect on liquid phase

Full size image
Fig. 9
figure 9

a Effect of K2O on CaO–SiO2 phase diagram, b magnified view of fig. a

Full size image

Figure 10 depicts the effect of increasing MgO on the formation of C3S, C2S and ionic liquid phase. MgO has a nominal effect on decreasing C2S and increasing C3S. Addition of MgO up to 1 wt% in raw mix can increase C3S content by ~ 2 wt%. It reduces the ionic liquid formation temperature by 20 °C but, it does not have significant role in increasing ionic liquid below 1400 °C. Although, it can assist in increasing ionic liquid above 1400 °C. Figure 10d shows, higher amount of MgO (> 1 wt%) leads to the formation of periclase (free MgO) which is partially dissolves with increasing temperature. If MgO is more than 1.5 wt %, it can’t dissolve completely into the ionic liquid phase even at 1400 °C as a result, free MgO may be present as a form of periclase with other phases in the clinker. Free MgO and CaO are not desirable in the final product as it causes expansion during hydration and generates crack after setting of cement.

Fig. 10
figure 10

Effect of MgO on phase formation in C–A–S–F–M system in temperature range of 1100 °C to 1500 °C; a effect on C2S, b effect on C3S, c effect on liquid phase and d effect on Free MgO formation

Full size image

Figure 11 depicts the effect of TiO2 on various phase formation. It significantly increases C2S and decreases C3S above 1290 °C. At lower temperature, it does not have any effect on the formation of ionic liquid. But, above 1350–1370 °C, it significantly increases ionic liquid with increasing TiO2 as shown in Fig. 11c. Figure 12a, b show the effect of TiO2 on CaO–SiO2 binary phase diagram. It shows that at 2 wt% TiO2, C3S decreases and C2S increases as \(\frac{{O}^{^{\prime}}{R}^{^{\prime}}}{{R}^{^{\prime}}{S}^{^{\prime}}}\) < \(\frac{OR}{RS}\) and \(\frac{{O}^{^{\prime}}{S}^{^{\prime}}}{{R}^{^{\prime}}{S}^{^{\prime}}}\) > \(\frac{OS}{RS}\). The amount of TiO2 in raw mix is also very critical as it has substantial effect on all the important phase formation. Although it helps in clinkerization by the formation of ionic liquid phase but, excess amount of TiO2 in raw mix can be detrimental as it leads to the lowering C3S and also causing coating formation in the kiln because of excessive ionic liquid formation at high temperature. The result of CALPHAD found to be in line with the observations of many researchers as they also have reported that TiO2 increases the burnability of raw mix by increasing the ionic liquid phase and as a result the free lime decreases [28].

Fig. 11
figure 11

Effect of TiO2 on phase formation in C–A–S–F–M system

Full size image
Fig. 12
figure 12

a Effect of addition of 2wt% TiO2 on CaO–SiO2 system and b magnified view of fig. a shows the effect on C3S and C2S formation

Full size image

Figure 13 shows the effect of all the minor oxides on phase formation at 1450 °C. SO3 increases ionic liquid phase up to 1% at 1450 °C as shown here in Fig. 12a. C2S increases linearly with increasing SO3 when it is more than 1% whereas, C3S decreases linearly with increasing SO3. Ionic liquid increases with increasing Na2O. But, the rate of increase of ionic liquid, decreases after 0.4% addition of Na2O. C2S decreases with increasing Na2O up to 0.3% then it saturates as shown in Fig. 13b. But, C3S increases with increasing Na2O up to 0.4% and it reduces with increasing Na2O. Similarly, with increasing K2O, ionic liquid increases and C2S decreases with a different pattern but, C3S increases with increasing K2O as shown in Fig. 13c. That is why alkalies (Na2O and K2O) play a critical role in deciding the strength of Portland cement. This effect is more prominent for Na2O as it is more effective in increasing C3S and decreasing C2S as shown in Fig. 13b, c. Addition of 1 wt% of MgO, can increase the ionic liquid content by 1%, increases C3S content by 3 wt% and decreases C2S by 3%. Beyond that, it does not have much effect on phase evolution at 1400 °C. TiO2 has a major role in increasing ionic liquid and C2S at 1450 °C as shown in Fig. 13e. But, it has a negative role in decreasing C3S at this temperature.

Fig. 13
figure 13

Effect plot of a SO3, b Na2O, c K2O, d MgO and e TiO2 on phase formation of C–A–S–F–M system at 1450 \(^\circ{\rm C}\)

Full size image

Figure 14 shows the effect of addition of different mineralizers on ionic liquid formation of C–A–S–F–M system. It shows that the weight percent of ionic liquid increases with increasing doses of different mineralizer. AlF3 has the maximum effect on increasing ionic liquid compared to other mineralizers whereas, ZnO and CaSO4 have very marginal effect in increasing ionic liquid. Up to 0.5 wt%, both MgSiF6 and Na2SiF6 have similar effect. But, MgSiF6 is more effective compared to Na2SiF6 when their proportion is more than 5 wt%. After Na2SiF6, CaF2 is more effective in increasing ionic liquid compared to CaCl2. The order of the mineralizers based on the ability of assisting ionic liquid formation is: AlF3 > MgSiF6 > Na2SiF6 > CaF2 > CaCl2 > CaSO4 > ZnO. This plot is very useful to compare the effectiveness of different mineralizers and to decide the required dosages of a mineralizer.

Fig. 14
figure 14

Effect of mineralizer on the formation of liquid phase at 1400 °C

Full size image

Figure 15a, b depicts the effect of increasing mineralizers on the formation of C3S and C2S. Thus far, it has been observed that mineralizers also have serious impact on C3S and C2S formation while increasing ionic liquid. AlF3, MgSiF6 and Na2SiF6 have prominent effect in decreasing C3S whereas, CaF2 and CaCl2 have moderate effect as shown in Fig. 15a. Although, in case of ZnO and CaSO4, increases C3S marginally. Figure 15b displays, AlF3, MgSiF6 and Na2SiF6 have significant effect in increasing C2S. Addition of AlF3, MgSiF6 and Na2SiF6 by 0.5 wt% can be effective in increasing C2S by ~ 2 wt%. Above 0.5 wt%, it reduces C2S in the system. CaSO4 and CaCl2 have negligible effect in decreasing C2S. CaF2 has comparatively higher effect on decreasing C2S when it is more than 0.5 wt%.

Fig. 15
figure 15

Effect of mineralizers on formation of a C3S and b C2S at 1400 °C

Full size image

From the above results, it can be observed that, fluorides are more effective in increasing the ionic liquid phase, decreasing C3S and increasing C2S compared to other types of mineralizers. If only fluorides are considered, the order of the effectiveness of fluorides also depends on the total fluorine content in the system. The wt% fluorine in MgSiF6 and AlF3 are maximum, 68.51 wt% and 67.85 wt% respectively. But, added advantage of ‘Al’ which is present in AlF3, makes it marginally more effective compared to MgSiF6. Na2SiF6 is also very effective mineralizer containing 60.61 wt% of fluorine. Finally, CaF2 is the least influential fluoride mineralizer among them having very low fluoride content (48.66 wt%). With increasing ionic liquid, the solubility of Ca2+ in the ionic liquid also increases. So, the available free calcium decreases in the system and reacts with C2S to form C3S. As a result, the C3S decreases with increasing ionic liquid and C2S increases. The variation of C3S and C2S is related to the ionic liquid formation.

Figure 16 shows the effect of all the mineralizers on the reduction of clinkerization temperature. Here, clinkerization temperature is assumed to be the temperature at which 25 wt% ionic liquid forms. It shows, by the addition of 0.5 wt% and 1 wt% of AlF3, clinkerization temperature can be reduced by ~ 150 °C and ~ 250 °C respectively. Both Na2SiF6 and MgSiF6, can reduce clinkerization temperature by ~ 100 °C and ~ 200 °C by the addition of 0.5 wt% and 1 wt% respectively. CaF2 has comparatively higher effect in reducing clinkerization temperature compared to CaCl2 whereas, CaSO4 and ZnO has the least effect. So, fluorides are the most effective mineralizer, followed by chlorides and sulphates to reduce clinkerization temperature. To compare the effects of different mineralizers, linear regression model (Y = A + BX, where A and B are the constant) was developed. Table 7 summarizes the result of the regression analysis. The coefficient of X or the slope of the curve, represents the effectiveness of the mineralizer. AlF3 is the most effective mineralizer as it has lowest value (− 247) of the coefficient and ZnO is the least effective mineralizer having highest value of coefficient of X (− 43). The clinkerization temperature can be calculated easily by using these equations within the range of 0 to 1 wt%.

Fig. 16
figure 16

Effect of mineralizer on clinkerization temperature

Full size image
Table 7 Effect of mineralizer on clinkerization temperature
Full size table

Although AlF3 is one of the most effective mineralizer however, it is very expensive. But, it is possible to combine a proportion of currently practiced mineralizer (such as CaF2) with AlF3 to enhance the overall performance of the mineralizer. Figure 17 shows the result of combined effect of different mineralizers in reducing clinkerization temperature. It shows that both CaF2 and CaCl2 is highly effective in reducing clinkerization temperature with addition of AlF3 up to 0.6%. Above 0.6%, AlF3 is not much effective in reducing clinkerization temperature. But, CaF2 can reduce the clinkerization temperature linearly by replacing CaCl2 up to 1 wt% although the effect is quite minimal compared to AlF3 as shown in Fig. 17.

Fig. 17
figure 17

Combined effect of mineralizers on reduction of clinkerization temperature

Full size image

4 Summary and conclusions

The influence of important minor oxides and mineralizers on Portland cement phase formation has been investigated in detail through current CALPHAD study. It was also found that many of our findings through CALPHAD simulation were in line with the observations reported by many researchers as mentioned earlier. Through this study, the usefulness of available CALPHAD software i.e., FactSage to analyze the cement clinker has been demonstrated as it has the spin off benefit of considering minor constituents and process conditions (T, P) in the calculation to predict the clinker phases more accurately.

This study demonstrates how different minor oxides and mineralizers play vital role in influencing the cement phase formation. Increasing SO3 and TiO2, C2S increases and C3S decreases in the clinker. On the other hand, presence of alkalies (Na2O and K2O) increase C3S and decrease C2S. Although all the minor oxides assist formation of ionic liquid. Most of the minor oxides also critically assist in increasing the burnability of raw mix and thereby benefits in fuel saving. So, the presence of alkalis (Na2O and K2O) are desirable up to a certain limit in order to help clinkerization process as they assist in ionic liquid formation. MgO is one of the most common oxide present in clinker as up to 2 wt%, it helps in clinkerization process by decreasing ionic liquid formation temperature and by increasing C3S content [2]. Excess MgO in the clinker remains unreacted and causes expansion problem during hydration. Hence, the use of MgO is restricted to a maximum limit of 6 wt% to suppress the detrimental effect [5].

The effect of selected mineralizers has also been studied using CALPHAD. Fluoride mineralizers are found to be superior in terms of reducing clinkerization temperature. Among all the fluorides, AlF3 is the most effective mineralizer compared to other fluorides. The order of the effectiveness in reducing clinkerization temperature of the mineralizers are: AlF3 > MgSiF6 > Na2SiF6 > CaF2 > CaCl2 > CaSO4 > ZnO. It was also observed, most of the mineralizers reduce C3S and increase C2S while increasing ionic liquid. So, deciding the dosage of mineralizer is very critical as it has a direct impact on the final properties of cement.

This study also establishes that CALPHAD based phase equilibria calculation approach is a highly effective tool and valuable addition in the understanding and analysis of minor oxides and mineralizers in C–A–S–F–M system.