Analysis of Influencing Factors in the Preparation Process of Solid Waste-Based Ternesite Sulphoaluminate Cement

Abstract

Based on a novel ternesite sulphoaluminate cement (NTSAC), the effects of various influencing factors on the calcination of clinker were studied, including mineral composition of clinker, grinding fineness of raw materials, molding technology of samples, and cooling methods of clinker. The research was carried out by taking the calcination system and mineral content of clinker as evaluation indexes, and using RSM and QXRD as analytical means. The results indicate that the optimal calcination temperature of clinker varies with the design mineral composition, while the holding time remains basically unchanged. Clinker with high CaSO₄ content has a relatively lower calcination temperature. The use of a calcination system of 1175 °C-49 min can control the mineral content error of the cement below 15%. Moreover, the molding pressure, molding methods, grinding fineness of raw materials, and cooling methods of clinker have significant effects on the clinker preparation to varying degrees, with the order of influence from high to low being molding methods, grinding fineness of raw materials, molding pressure, and cooling methods of clinker. Within the range of experimental parameters, the better preparation conditions are compression molding (molding method), 15 MPa (molding pressure), and 20 μm (grinding fineness). The above research conclusions provide reference data for cement preparation in the laboratory, offering useful guidance for developing novel types of cement.

1. Introduction

Cement plays an important role in modern infrastructure construction, and with its excellent bonding and hardening properties, it has become an indispensable key material for constructing various types of buildings. However, the large-scale production of cement has caused problems such as the shortage of mineral resource and environmental pollution [1,2]. For the sustainable development of the cement industry, low-carbon cement composed of low-calcium minerals [3,4,5,6] and prepared from solid waste [7,8,9,10] has become a research hotspot. In the research and development process of low-carbon cement, it often needs to go through the laboratory stage, industrial test stage, and industrial production stage. In each stage, there are complex and diverse factors involved, including the quality of raw materials [11,12,13], molding technology of raw materials [14,15], and calcination conditions of clinker [16,17,18]. These factors are intertwined and play a decisive role in the final properties of low-carbon cement. The importance of the laboratory stage as the primary link is self-evident, it provides important reference data and theoretical support for the subsequent stages, so it is of great significance to investigate the effects of influencing factors on cement calcination in this stage.

Generally, the main purpose of laboratory research is to obtain the optimal calcination system and the theory of mineral formation. For specific raw materials, the factors that affect experimental results mainly include the mineral composition of clinker, grinding fineness of raw materials, molding technology of samples, and cooling methods of clinker. At present, there have been some studies on the factors affecting cement calcination, but they are not systematic and universal. Research on the cooling mode is common in cement preparation. Relevant studies [19,20] believe that slow cooling can easily lead to the transformation of β-C₂S to γ-C₂S in clinker and the increase in the grain size of clinker minerals, so that the cement strength obtained by slow cooling is significantly lower than that obtained by rapid cooling. Some research [21] also shows that the change of cooling mode will have a certain impact on the content of various minerals. The molding of raw materials is an indispensable part of cement preparation in the laboratory. Currently, the molding method adopted in the laboratory mainly includes compression molding, spherical molding, and powder direct use of untreated molding. However, different molding methods may lead to different calcination systems, and existing research [14] has shown that the calcination temperature of powdered materials is lower than that of spherical materials. In addition, for the most commonly used compression molding, although the molding pressures used in various studies [15,22] all fall within the range of 5–20 MPa, they are not identical. This inconsistency leads to variations in the cement calcination system, which fails to provide reliable reference benchmarks for future research. Raw material grinding is an extremely important step in the cement production process, and the purpose of raw material grinding is to make the raw material fineness suitable for the preparation of clinker. At present, the fineness requirement for raw materials used in the preparation of cement in the laboratory is generally 80 μm square hole sieve with a residue of less than 5% or 10%, but there is no unified standard. Of course, some studies [23,24] have explored the effect of grinding technology on clinker calcination and the effect of raw material fineness on burnability. The results suggest that the finer the raw material, the better its combustibility, the lower the content of free calcium oxide in clinker, and the faster the reaction rate. However, none of the aforementioned studies have analyzed the effect of fineness on the actual mineral content of clinker. The proportion of raw materials is an important factor affecting cement calcination, and the correct proportion of raw materials is necessary to prepare the target clinker of the set minerals. Studying the calcination system of cement clinker with different mineral compositions can reflect the adaptability of this type of cement to a specific calcination system. Su [25] prepared high belite sulphoaluminate cement (HBSAC) with different mineral compositions and studied the effect of mineral composition on the calcination process of cement clinker. The results indicate that the optimal calcination system is basically the same for different mineral composition groups. However, the mineral formation process varies from different types of cement, so the rules obtained from the above research are not universal.

In order to further understand the effects of influencing factors on the calcination of cement in the laboratory, and based on the laboratory preparation research of ternesite sulphoaluminate cement (NTSAC), the influence of key factors such as the clinker mineral composition of clinker, grinding fineness of raw materials, molding technology of samples, and cooling methods of clinker on mineral content and calcination system is systematically analyzed. Among them, the mineral content is determined by QXRD analysis [26,27], and the optimal calcination system is determined by RSM method [28]. The NTSAC mentioned here is a novel solid waste-based sulphoaluminate cement developed by our research group. Its calcination system is below 1200 °C, with a holding time of less than 60 min, which is significantly lower than the calcination temperature of ordinary Portland cement at 1350–1450 °C, and sulphoaluminate cement at 1250–1350 °C. The difference in holding time is not significant, usually between 30–60 min. In addition, the successful preparation of NTSAC has improved the utilization rate of solid waste such as petroleum coke ash, reduced the use of natural gypsum and limestone, and since the minerals basically do not contain CaCO₃, there is almost no decomposition of CaCO₃ during the calcination process, which effectively reduces CO₂ emissions and alleviates the greenhouse effect. Its detailed preparation method can be found in reference [26]. This study aims to provide new insights into the preparation of cement in the laboratory and offer valuable references for ensuring consistency and comparability of future research results.

2. Experiment

2.1. Experimental Materials

The preparation of NTSAC was carried out under laboratory conditions using three types of solid waste materials, including petroleum coke ash provided by Sinopec Qingdao Petrochemical Co., Ltd., Qingdao, China, fly ash provided by Henan Jinrun New Materials Co., Ltd., Zhengzhou, China, and calcined bauxite provided by Henan Borun Casting Materials Co., Ltd., Gongyi, China. The main chemical and mineral compositions of the raw materials are shown in

Table 1. Main chemical composition of raw materials.

Raw Materials	CaO/%	Al2O3/%	SiO2/%	SO3/%	Fe2O3/%	TiO2/%	MgO/%	Others/%	LOI/%
Petroleum coke ash	62.36	0.45	0.85	26.09	0.24	0.00	1.07	0.36	8.58
Fly ash	3.44	30.35	52.14	0.79	5.63	1.22	0.73	3.25	2.45
Calcined bauxite	1.07	75.14	14.84	0.18	1.53	4.50	0.48	1.63	0.63

Note: LOI refers to loss on ignition.

and Figure 1, respectively.

2.2. Experimental Design

In order to reveal the effects of influencing factors on clinker calcination, we conducted the following experimental schemes design. Among them, the influence of mineral composition design is mainly evaluated through changes in the optimal calcination system of clinker, while the influence of other factors is evaluated by observing changes in the mineral content of clinker under the same calcination system.

(1)

Design of control group

In this article, the study on the preparation of NTSAC in reference [26] is selected as the control group, denoted as Group CP. The mineral composition design, experimental conditions, optimal calcination system, and clinker mineral composition of Group CP are shown in

Table 2. Key elements of Group CP in the preparation process.

Group Number	Calcination System	Cooling Method	Molding Method	Molding Pressure	Fineness of Raw Material	Mineral Composition				Raw Material Mix Ratio/%
						Design Value/%		Actual Value/%
CP	1175 °C -49 min	Rapid cooling	Compression molding	15 MPa	20 μm	${\mathrm{C}}_5{\mathrm{S}}_2\overline{\mathrm{S}}$	45	${\mathrm{C}}_5{\mathrm{S}}_2\overline{\mathrm{S}}$	47.76	Petroleum coke ash	74.42
						${\mathrm{C}}_4{\mathrm{A}}_3\overline{\mathrm{S}}$	35	${\mathrm{C}}_4{\mathrm{A}}_3\overline{\mathrm{S}}$	37.09	Fly ash	12.35
						CaSO4	15	CaSO4	15.62	Calcined bauxite	12.93
						C4AF	5	C4AF	--

Note: Due to the solid solution of Fe element in C₄A₃$\overline{\mathrm{S}}$, 5% C₄AF was not obtained in clinker.

, and its preparation process is shown in Figure 2 [26].

(2)

Design of variable parameters

Based on the Group CP, 5 parameters were selected, including mineral composition of clinker (MC series), cooling methods of clinker (CM series), molding methods (MM series), molding pressure (MF series), and grinding fineness of raw materials (GF series); a total of 11 groups of experiments in 5 series were designed, as shown in Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8.

Table 3. Design content of clinker minerals in each group.

Group Number		MC-1	CP	MC-2
Mineral composition /%	${\mathrm{C}}_5{\mathrm{S}}_2\overline{\mathrm{S}}$	52.5	45	37.5
	${\mathrm{C}}_4{\mathrm{A}}_3\overline{\mathrm{S}}$	35	35	35
	CaSO4	7.5	15	22.5
	C4AF	5	5	5

Table 3. Design content of clinker minerals in each group.

Group Number		MC-1	CP	MC-2
Mineral composition /%	${\mathrm{C}}_5{\mathrm{S}}_2\overline{\mathrm{S}}$	52.5	45	37.5
	${\mathrm{C}}_4{\mathrm{A}}_3\overline{\mathrm{S}}$	35	35	35
	CaSO4	7.5	15	22.5
	C4AF	5	5	5

Table 4. Raw material mix ratio schemes of MC-1 and MC-2 under different calcination systems.

NO.	Petroleum Coke Ash /%	Fly Ash /%	Calcined Bauxite /%
MC-1 (1150.00 °C-42.12 min)	71.43	15.67	12.90
MC-1 (1162.50 °C-5.00 min)	75.30	14.85	9.86
MC-1 (1190.00 °C-60.00 min)	71.13	15.25	13.62
MC-1 (1190.00 °C-60.00 min)	73.22	15.04	11.74
MC-1 (1190.00 °C-60.00 min)	72.75	14.50	12.75
MC-1 (1207.50 °C-5.00 min)	75.24	14.80	9.96
MC-1 (1207.50 °C-50.37 min)	71.43	15.67	12.90
MC-1 (1207.50 °C-5.00 min)	75.30	14.85	9.86
MC-2 (1150.00 °C-42.12 min)	74.14	9.77	16.09
MC-2 (1162.50 °C-5.00 min)	78.39	9.28	12.33
MC-2 (1190.00 °C-60.00 min)	73.59	9.48	16.93
MC-2 (1190.00 °C-60.00 min)	75.99	9.38	14.64
MC-2 (1190.00 °C-60.00 min)	75.17	9.00	15.83
MC-2 (1207.50 °C-5.00 min)	78.30	9.25	12.45
MC-2 (1207.50 °C-50.37 min)	74.14	9.77	16.09
MC-2 (1207.50 °C-5.00 min)	78.39	9.28	12.33

Table 4. Raw material mix ratio schemes of MC-1 and MC-2 under different calcination systems.

NO.	Petroleum Coke Ash /%	Fly Ash /%	Calcined Bauxite /%
MC-1 (1150.00 °C-42.12 min)	71.43	15.67	12.90
MC-1 (1162.50 °C-5.00 min)	75.30	14.85	9.86
MC-1 (1190.00 °C-60.00 min)	71.13	15.25	13.62
MC-1 (1190.00 °C-60.00 min)	73.22	15.04	11.74
MC-1 (1190.00 °C-60.00 min)	72.75	14.50	12.75
MC-1 (1207.50 °C-5.00 min)	75.24	14.80	9.96
MC-1 (1207.50 °C-50.37 min)	71.43	15.67	12.90
MC-1 (1207.50 °C-5.00 min)	75.30	14.85	9.86
MC-2 (1150.00 °C-42.12 min)	74.14	9.77	16.09
MC-2 (1162.50 °C-5.00 min)	78.39	9.28	12.33
MC-2 (1190.00 °C-60.00 min)	73.59	9.48	16.93
MC-2 (1190.00 °C-60.00 min)	75.99	9.38	14.64
MC-2 (1190.00 °C-60.00 min)	75.17	9.00	15.83
MC-2 (1207.50 °C-5.00 min)	78.30	9.25	12.45
MC-2 (1207.50 °C-50.37 min)	74.14	9.77	16.09
MC-2 (1207.50 °C-5.00 min)	78.39	9.28	12.33

Table 5. Cooling method of each group.

Group Number	CM-1	CP
Cooling method	Self cooling	Rapid cooling

Note: Self cooling is natural cooling at room temperature (25 ± 5 °C), while rapid cooling is rapid cooling under the fan.

Table 5. Cooling method of each group.

Group Number	CM-1	CP
Cooling method	Self cooling	Rapid cooling

Note: Self cooling is natural cooling at room temperature (25 ± 5 °C), while rapid cooling is rapid cooling under the fan.

Table 6. Molding method and sample of each group.

Group Number	CP	MM-1	MM-2
Molding method	Compression molding	Spherical molding	Powder direct use of untreated molding
Sample

Note: Spherical molding is obtained by adding proper water to the raw material, rolling it into a ball with a diameter of 1 cm, and then drying. Compression molding is obtained by using a tablet press to compress tablets at 15 MPa, with a diameter of 14 mm × (3~5) mm. Powder direct use of untreated molding is obtained by mixing raw material powder without processing.

Table 6. Molding method and sample of each group.

Group Number	CP	MM-1	MM-2
Molding method	Compression molding	Spherical molding	Powder direct use of untreated molding
Sample

Note: Spherical molding is obtained by adding proper water to the raw material, rolling it into a ball with a diameter of 1 cm, and then drying. Compression molding is obtained by using a tablet press to compress tablets at 15 MPa, with a diameter of 14 mm × (3~5) mm. Powder direct use of untreated molding is obtained by mixing raw material powder without processing.

Table 7. Molding pressure and sample of each group.

Group Number	MF-1	MF-2	CP	MF-3
Molding pressure/MPa	5	10	15	20
Sample

Note: When the pressure reaches 20 MPa, the raw material tablet is prone to delamination and other phenomena, but it does not affect their calcination.

Table 7. Molding pressure and sample of each group.

Group Number	MF-1	MF-2	CP	MF-3
Molding pressure/MPa	5	10	15	20
Sample

Note: When the pressure reaches 20 MPa, the raw material tablet is prone to delamination and other phenomena, but it does not affect their calcination.

Table 8. Grinding fineness of raw materials of each group.

Group Number	GF-1	CP	GF-2
Fineness/μm	10	20	30

Note: The fineness in the table is characterized by the median particle size D50 in the particle size analysis results. The difference in fineness is obtained by grinding the raw materials at different times, and the corresponding particle size analysis results are detailed in Figure 3.

Table 8. Grinding fineness of raw materials of each group.

Group Number	GF-1	CP	GF-2
Fineness/μm	10	20	30

Note: The fineness in the table is characterized by the median particle size D50 in the particle size analysis results. The difference in fineness is obtained by grinding the raw materials at different times, and the corresponding particle size analysis results are detailed in Figure 3.

2.3. Test Methods

(1) Chemical composition: A Model 1800 X-ray fluorescence spectrometer from Shimadzu, Japan was utilized for the analysis. The testing parameters included a Rh target X-ray tube, with the voltage and current of the tube being 40 kV and 80 mA, a 4 kW thin window, and a scanning speed of 300°/min.

(2) Mineral composition: A D8 Advance X-ray diffractometer from Bruker, Germany was utilized for the analysis. The testing parameters included a Cu target, Kα ray, with the voltage and current of the tube being 40 kV and 40 mA, and the qualitative analysis and quantitative analysis having residence times of 0.05 s and 0.5 s, a scanning range from 5° to 60° for the 2θ angle, and step width of 0.02°. The quantitative analysis (QXRD) of clinker minerals was conducted using FullProf 2020.6 software. The crystallographic data for the relevant minerals applied in Rietveld refinement are presented below: PDF#88-0812 for C₅S₂$\overline{\mathrm{S}}$, and the space group is Pnma; PDF#71-0969 for C₄A₃$\overline{\mathrm{S}}$, and the space group is I-43m; PDF#86-0398 for β-C₂S, and the space group is p21/n; PDF#99-0010 for CaSO₄, and the space group is Amma.

3. Results and Discussion

3.1. The Effect of Mineral Composition on Clinker Calcination

In order to obtain the optimal calcination temperature and holding time for the MC-1 and MC-2 groups, this study referred to the approach of the Group CP in reference [26]. Due to limited space, the verification process of model rationality is not elaborated here, and only the predicted results of the optimal calcination degree based on the numerical function in Design Expert 10.0 software are listed. The variance analysis tables of MC-1 and MC-2 models, as well as the 3D response surfaces and contour plots of MC-1 and MC-2, as shown in Table S1 and Figure S2 of the Supplementary Materials.

Table 9. The optimal calcination system of each group and the mineral composition of its clinker.

Group Number	Data	Calcination Temperature /°C	Holding Time /min	${\mathbf{C}}_5{\mathbf{S}}_2\overline{\mathbf{S}}$ /%	${\mathbf{C}}_4{\mathbf{A}}_3\overline{\mathbf{S}}$ /%	CaSO4 /%
MC-1	Wish value	--	--	55.26	36.84	7.89
	Predictive value	1183.77	47.38	52.72	36.84	7.89
	Actual value	1183.77	47.38	53.34	37.95	8.03
CP	Wish value	--	--	47.37	36.84	15.79
	Predictive value	1175.15	49.27	47.40	38.77	15.75
	Actual value	1175.15	49.27	47.76	37.09	15.62
MC-2	Wish value	--	--	39.47	36.84	23.68
	Predictive value	1157.65	47.91	39.47	39.76	23.68
	Actual value	1157.65	47.91	38.98	38.25	22.33

shows the optimal calcination system, wish values, predicted values, and actual values for each group. From the table, it can be seen that the optimal holding time of the three groups is basically the same, while the optimal calcination temperature decreases with the increase in residual CaSO₄ content inside. The contour lines of the wish functions for each group are shown in Figure 4. It can be seen from the figure that the maximum wish function values for MC-1, CP, and MC-2 are 0.985, 0.981, and 0.959, respectively, which are all close to 1, indicating strong reliability of the predicted values. When the calcination system is 1175 °C-49 min, the desired function values of each group are ≥0.85, indicating that cement with different mineral compositions can be prepared under this calcination system, and the error between the obtained clinker mineral content and the desired value is less than 15%.

3.2. The Effect of Cooling Methods on Clinker Calcination

Quantitative analysis was conducted on clinker under different cooling methods, and the results are shown in Figure 5a. According to the figure, the content of each mineral in the clinker obtained by the two cooling methods is similar, indicating that both cooling methods have little effect on the mineral content in the clinker. This is significantly different from the cooling method of ordinary sulphoaluminate cement and Portland cement. The reason for this is that there is no C₂S in this type of cement clinker, and changing the cooling method will not greatly affect the mineral changes.

In order to determine which cooling method yields the closest mineral composition to the design value, this article analyzes the error between the mineral content of clinker obtained by different cooling methods and the design value, as shown in Figure 5b. The results indicate that the mineral content of clinker under self cooling exhibits a high degree of error, especially for the CaSO₄ content (relative error of 4.62%). The errors between the C₄A₃$\overline{\mathrm{S}}$, C₅S₂$\overline{\mathrm{S}}$, and CaSO₄ content in the clinker obtained by self cooling method and the design value are +0.95%, −2.01%, and +4.62%, while the errors by rapid cooling are +0.67%, +0.82%, and −1.11%, respectively. The average error of mineral content in clinker under self cooling is 2.53%, which is 1.65% higher than that under rapid cooling. Therefore, the optimal cooling method within the experimental range is rapid cooling, but if the accuracy requirement is not high, relative energy-saving self cooling can also be chosen.

3.3. The Effect of Molding Technology of Samples on Clinker Calcination

3.3.1. Molding Method

Quantitative analysis was conducted on the clinker obtained under different molding methods, and the analysis results are shown in Figure 6a. As shown in the figure, different molding methods have a significant impact on the content of C₅S₂$\overline{\mathrm{S}}$ and CaSO₄, and a relatively small impact on the content of C₄A₃$\overline{\mathrm{S}}$. The content of C₅S₂$\overline{\mathrm{S}}$ and C₄A₃$\overline{\mathrm{S}}$ is highest during compression molding, at 47.76% and 37.09%, respectively; the content of C₅S₂$\overline{\mathrm{S}}$ and C4A₃$\overline{\mathrm{S}}$ is the lowest during powder direct use of untreated molding, at 41.30% and 34.26%, respectively. The content of CaSO₄ is highest during powder direct use of untreated molding, reaching 23.44%; the content of CaSO₄ is the lowest during compression molding, at 15.62%. This indicates that among the three molding methods of raw materials, the compression molding reaction is the most thorough, followed by spherical molding, and powder direct use of untreated molding is the worst. The reason for the above results is that the raw material particles formed by the powder direct use of untreated molding have larger gaps and smaller contact areas, which significantly hinder thermal conductivity and affect the uniform heating of the raw material. Compared with powder direct use of untreated molding, spherical forming reduces the surface area and volume ratio of raw materials, thereby reducing the surface energy during sintering and improving heating uniformity. However, the improvement in raw material density by spherical molding is relatively small. In contrast, compression molding improves the density of the raw material and the flatness of the tablet surface, which is more conducive to the improvement of thermal conductivity uniformity and the progress of sintering reaction [29,30,31].

In order to explore which molding method can make the obtained clinker composition closest to the design value, error analysis was conducted on the mineral content and design value of clinker obtained under different methods. The analysis results are shown in Figure 6b. From the figure, it can be seen that regardless of the molding method, the error of CaSO₄ content in clinker is the largest among the three minerals. The error between the C₄A₃$\overline{\mathrm{S}}$, C₅S₂$\overline{\mathrm{S}}$, and CaSO₄ content in the clinker obtained by powder direct use of untreated molding and the design values are −6.99%, −12.81%, +48.43%. The errors by spherical forming are −2.56%, −8.90%, and +26.35%, and the errors by compression forming (15 MPa) are +0.67%, +0.82%, and −1.11%. The average errors of each clinker mineral are 12.60% for spherical molding, 0.87% for compression molding, and 22.75% for powder direct use of untreated molding, respectively. It can be seen that the error between the mineral content of the clinker obtained by raw material compression molding and the design value is relatively small, followed by spherical molding and powder direct use of untreated molding. In addition, the scatter points of the compression molding are closer to the zero-error curve, indicating that compression molding is the optimal molding method of raw materials. However, although compression molding is precise, its production efficiency is limited, making it mainly suitable for small-scale sample preparation. When large-scale sample preparation is required, if the error is within the allowable range, spherical forming becomes a feasible alternative that saves time and labor.

3.3.2. Molding Pressure

Quantitative analysis was conducted on cement clinker obtained under different pressures, and the analysis results are shown in Figure 7a. According to the figure, the content of C₅S₂$\overline{\mathrm{S}}$ and C₄A₃$\overline{\mathrm{S}}$ increases with increasing pressure, while the content of CaSO₄ decreases with increasing pressure. The reason for this is that as pressure rises, the inter particle forces and density are improved, and the raw material particles are more tightly bound together. This not only makes the reaction more thorough, but also forms a more uniform and compact clinker structure [32,33]. In addition, the increase in pressure allows more CaSO₄ to participate in the formation process of minerals, but the difference in mineral content between pressures of 15 MPa and 20 MPa is not significant, indicating that when the pressure reaches a certain level, the promoting effect of pressure on the reaction gradually weakens. At this point, continuing to increase pressure has little effect on mineral content. Zhang’s [15] and Chen’s [22] research found that increasing the molding pressure of raw materials can improve the property of cement, but when the pressure increases to a certain extent, the improvement effect gradually weakens, which is consistent with our research conclusion and confirms the reliability of our research.

In order to explore at which pressure the composition of clinker obtained is closest to the design value, an error analysis was conducted on the mineral content of clinker obtained under different pressures and the design value. The analysis results are shown in Figure 7b. The errors between the C₄A₃$\overline{\mathrm{S}}$, C₅S₂$\overline{\mathrm{S}}$, and CaSO₄ contents in the clinker obtained under a pressure of 5 MPa and the design values are −5.03%, −4.67%, and +13.29%. The errors under 10 MPa pressure are −4.65%, −3.01%, and +7.36%. The errors under 15 MPa pressure are +0.67%, +0.82%, and −1.11%, and the errors under 20 MPa pressure are +1.64%, +1.15%, and −3.88%. The average errors between the content of C₅S₂$\overline{\mathrm{S}}$, C₄A₃$\overline{\mathrm{S}}$, and CaSO₄ in clinker obtained under different pressures and the design values are 7.66% (5 MPa), 5.01% (10 MPa), 0.87% (15 MPa), and 2.22% (20 MPa), respectively. As shown in the figure, when the pressure is 15 MPa, the average error between the mineral content obtained and the design value is the smallest, and its scatter points are also closest to the zero-error curve, indicating that 15 MPa is the optimal molding pressure of raw materials. When there are a large number of samples, under the premise of meeting the error range, the molding pressure can also be appropriately reduced to achieve the goal of saving time and improving efficiency.

3.4. The Effect of Grinding Fineness of Raw Materials on Clinker Calcination

Quantitative analysis was conducted on the clinker obtained under a differing grinding fineness of raw materials, and the analysis results are shown in Figure 8a. According to the figure, the finer the raw material, the higher the content of C₅S₂$\overline{\mathrm{S}}$ and C₄A₃$\overline{\mathrm{S}}$ in the obtained clinker, and the lower the content of CaSO₄. This indicates that the finer the raw material, the larger the contact area between particles, which is more conducive to diffusion and heat transfer between particles, thereby accelerating the solid phase reaction. The research results of Andres [23] and Briki [24] indicated that the finer the fineness of the raw material, the lower the content of free calcium oxide in its clinker, and it can accelerate the solid-phase reaction, which strongly supports our viewpoint.

In order to determine the optimal grinding fineness of raw materials, an error analysis was conducted on the mineral composition and design values of clinker fired with a differing grinding fineness of raw materials. The analysis results are shown in Figure 8b. The error between the content of C₄A₃$\overline{\mathrm{S}}$, C₅S₂$\overline{\mathrm{S}}$, and CaSO₄ in the clinker obtained when the grinding fineness of the raw material is 10 μm and the design values are +1.38%, +4.07%, and −19.02%. The errors when the fineness is 20 μm are +0.67%, +0.82%, −1.11%, and the errors when the fineness is 30 μm are −2.04%, −6.11%, and +23.18%. The average errors between the content of C₅S₂$\overline{\mathrm{S}}$, C₄A₃$\overline{\mathrm{S}}$, and CaSO₄ in clinker obtained under different grinding fineness and the design values are 8.16% (10 μm), 0.87% (20 μm), and 10.44% (30 μm), respectively. When the fineness is 20 μm, the error is minimized and its scatter points are closer to the zero-error curve, indicating that 20 μm is the optimal grinding fineness of raw material. In addition, as the grinding fineness increases, the energy consumption for grinding also increases. Taking into account various factors such as mineral content errors and economic considerations, the fineness requirements can be appropriately relaxed.

4. Conclusions

This research mainly used RSM and QXRD methods to explore the effects of influencing factors on clinker calcination in the laboratory preparation process of NTSAC, including mineral composition of clinker, grinding fineness of raw materials, molding technology of samples, and cooling methods of clinker. The main conclusions drawn from the factors are as follows:

(1) For NTSAC, the optimal calcination temperature decreases with the increase in residual CaSO₄ content inside, and its range is roughly 1157–1184 °C, while the optimal holding time for clinker with different mineral compositions is basically the same, about 48 min. When the average error between the mineral content of the clinker and the design value is not greater than 15%, the different mineral compositions of clinker can be successfully prepared at 1175 °C, 49 min.

(2) The cooling method has a relatively small impact on the calcination of clinker, and the effect by rapid cooling is relatively better. The average error of mineral content in clinker under self cooling is 2.53%, which is 1.65% higher than that under rapid cooling. However, if the precision requirement is not high during production, relative energy-saving self cooling can also be chosen.

(3) The molding method has a significant impact on the calcination of clinker, and the effect by compression molding is relatively better. The average errors of each clinker mineral are 12.60% for spherical molding, 0.87% for compression molding, and 22.75% for powder direct use of untreated molding. However, when preparing a large number of samples, spherical molding can also be used as an alternative preparation method to save time.

(4) The molding pressure has a considerable impact on the calcination of clinker, and the effect with a pressure of 15 MPa is relatively better. The average errors between the content of C₅S₂$\overline{\mathrm{S}}$, C₄A₃$\overline{\mathrm{S}}$, and CaSO₄ in clinker obtained under different pressures and the design values are 7.66% (5 MPa), 5.01% (10 MPa), 0.87% (15 MPa), and 2.22% (20 MPa). However, when preparing a large number of samples, the pressure can also be appropriately reduced to improve work efficiency.

(5) The grinding fineness of raw materials has a considerable impact on the calcination of clinker, and the effect is relatively better when the raw material fineness is 20 μm. The average errors between the content of C₅S₂$\overline{\mathrm{S}}$, C₄A₃$\overline{\mathrm{S}}$, and CaSO₄ in clinker obtained under different grinding fineness and the design values are 8.16% (10 μm), 0.87% (20 μm), and 10.44% (30 μm). However, when considering energy consumption and the economy, the fineness requirements can be appropriately relaxed within the allowable range of errors.

This research mainly analyzes the effects of influencing factors of NTSAC preparation from the perspective of clinker mineral content, and achieved a subtle regulation of factors. Based on this, future research will focus on optimizing clinker performance through micro level and multi factor regulation. In addition, the existence and potential impact of amorphous phases will be considered in subsequent research, further deepening and improving the theoretical research of the influence of various factors on cement preparation.