Introduction

SO2, the primary culprit behind acid rain, poses significant threats to the Earth's ecosystem, human society, and economy, with the steel and electricity sectors being major contributors (Liu et al. 2020; Shi et al. 2017). Given to its severe environmental contamination, SO2 emission is currently under rigorous regulation by the State Environmental Protection Agency, creating flue gas desulphurization an enormously relevant issue. It is worth noting that the implementation of desulfurization technology in coal-fired power plants and the steel industry has contributed to SO2 emissions in 2021 accounting for 12.97% of those in 2012, but desulfurized ash will increase proportionately (Wang et al. 2023; Yi et al. 2023), as illustrated in Fig. 1.

Fig. 1
figure 1

Sulfur dioxide emissions in China (2012–2021)

The prevailing desulfurization technologies in the current scenario can be categorized into wet desulfurization technology, dry desulfurization technology, and semi-dry desulfurization technology based on the type of desulfurization agent employed (Zhai et al. 2017; Mjalli et al. 2014; Li et al. 2022). However, the dry desulfurization method exhibits significantly lower desulfurization efficiency compared to wet and semi-dry techniques, leading most enterprises to abstain from its adoption. Although the wet desulfurization process achieves a desulfurization effectiveness exceeding 90% (Li et al. 2022), it suffers from drawbacks such as a protracted process flow and the generation of substantial waste liquid and desulfurized gypsum, severely impeding its marketability and utilization. As a result, the conventional desulfurization process is being increasingly replaced by the more advantageous semi-dry desulfurization method, which entails lower costs and eliminates the generation of excess waste liquid.

Nonetheless, the implementation of the semi-dry desulfurization method by steel mills and power plants gives rise to a considerable quantity of SFGDA, comprising primarily of CaSO3, CaSO4, f-CaO, Ca(OH)2, CaCl2, and CaCO3. (Liang et al. 2011; Zhang et al. 2012; Li et al. 2019). Due to its intricate chemical composition, SFGDA finds limited application in conventional sectors such as cement and building materials, and is predominantly regarded as solid waste, leading to its extensive storage and landfilling. One the one hand, the activated CaO and CaCl2 present in SFGDA negatively impact on the stability and strength of construction materials. The long-term hydration process involving CaO and H2O generates Ca(OH)2, the Ca(OH)2, which leads to volumetric expansion of buildings and structures. In addition, the presence of deliquescent properties and the release of chloride ions from CaCl2 in SFGDA can cause yellow stains on building walls and corrosion of steel bars, posing a significant threat to building safety.

On the other hand, in accordance with national standards, the mass percentage content of SO3 in ordinary cement or non-fired bricks should not exceed 3.5 wt.%, restricting the usages of desulfurization ash (Koralegedara et al. 2019; Wang et al. 2005). The prospect of secondary pollution arising from the release of SO2 from SFGDA under heated or acidic conditions is deeply concerning. Excessive incorporation of SFGDA into building materials is inappropriate, and a substantial amount of unused SFGDA remains stockpiled, thereby necessitating urgent attention to address the utilization of SFGDA.

Recently, the high-temperature pyrolysis process of decomposing CaSO4 and CaSO3 in SFGDA into CaO and SO2 has garnered significant attention in research. The CaO generated from the pyrolysis of the desulfurization ash is recycled as a calcium source, while the SO2-rich flue gas undergoes dust removal, washing, and catalytic conversion to produce sulfuric acid, enabling comprehensive utilization of Ca and S in SFGDA. However, the high decomposition temperature of pure CaSO4 reaching 1673 K, poses challenges in practical applications due to elevated production costs and energy consumption. To reduce the decomposition temperature of CaSO4, experts have conducted extensive research using various methods to reduce the decomposition temperature of CaSO4. Currently, there is a lack of comprehensive comparison and evaluation of these methods. Therefore, the primary aim of this review is to summarize recent advancements in the low-energy and low-environmental-risk thermal decomposition of SFGDA. This includes examining the fundamental characteristics of SFGDA, the decomposition behavior of SFGDA in inert, oxidizing, and reducing atmospheres, as well as the effects of solid additives. The present review will serve as a valuable reference for future investigations focused on highly efficient pyrolysis and the subsequent utilization of SFGDA, thereby contributing to the progress of sustainable materials and energy production research.

Physicochemical properties of SFGDA

Wei et al. (Wei et al. 2021) reported that the SFGDA generated by the iron and steel industry exhibits a varied color palette, encompassing shades of pink, light gray, and dark red, attributed to the presence of iron oxide compounds originating from the raw materials employed in the smelting process. Figure 2 depicts SFGDA from a steel industry with an inconsistent shape, a smooth surface, and an unconsolidated construction (Liu et al. 2010). What is more, the obscuration of SFGDA in steel plants is 15.05%, which is not conducive to apply untreated (Yu et al. 2018).

Fig. 2
figure 2

SEM image of SFGDA from steel factory. (Liu et al. 2010)

Furthermore, Liu et al. (2010) confirmed that the particle-size distribution of the SFGDA is from 3.42 to 13.77 μm and the median size is 4.18 μm, shown in Table 1. Surface analysis revealed that the SFGDA possesses a substantial specific surface area of approximately 7940 m2∙kg−1, indicating its pronounced chemical reactivity.

Table 1 Particle size characteristic parameters of SFGDA in the steel factory

Table 2 presents the composition of semi-dry flue gas desulfurization ash (SFGDA) obtained from steel plants, which includes components such as CaO, CaSO3, CaSO4, f-CaO, SiO2, Al2O3, Fe2O3 and MgO. and magnesium oxide (MgO). Furthermore, the SFGDA produced by iron and steel enterprises shows an average content of 32.70 wt.% CaO, 16.81 wt.% CaSO3, and 16.91 wt.% CaSO4, highlighting the abundance of calcium and sulfur resources. Predicated on the thermophysical properties of SFGDA, several studies have been initiated with the thermal decomposition for SFGDA.

Table 2 Chemical characteristics of SFGDA from steel plants. (Liu et al. 2010)

Thermal decomposition of SFGDA in inert and oxidizing atmosphere

The elucidation of the thermal decomposition mechanism of carbonate, sulfite, and sulfate compounds is not only intriguing in its own right but also provides valuable insights into the physical and chemical transformations occurring within SFGDA (Yu et al. 2018). Yang et al. (2019a, b) and Zhou et al. (Zhou et al. 2017) discovered that temperature plays a significant role in the decomposition of desulfurization ash under inert or oxidizing conditions. In a nutshell, the thermogravimetric experiment confirmed the existence of five distinct stages during the heating process of SFGDA in an N2 atmosphere. In reaction stage I, thermodynamic analysis revealed a 2 wt.% reduction in SFGDA at 473 K, attributed to the desorption of adsorbed water and the loss of crystalline water from compounds such as CaSO3∙0.5H2O, CaSO4∙2H2O.

Consequently, the transformation of Ca(OH)2 into CaO occurred below 673 K. In stage IV, as depicted in Fig. 3 and summarized in Table 3, the thermal decomposition of CaCO3 occurred within the temperature range of 953–1073 K. Characterized as Fig. 3 and Table 3, the thermal decomposition of CaSO3 in SFGDA commenced around 1083 K and concluded approximately at 1313 K during stage IV. In stage V, the release of SO2 can be attributed to the minor decomposition of CaSO4 at 1473 K, shown in Table 3. What’s more, the decomposition pattern of SFGDA in an N2 atmosphere exhibits similarities to its calcination in an Ar atmosphere.

Fig. 3
figure 3

TGA curves of semi-dry sintering flue gas desulfurized ash in N2 and air. (Zhou et al. 2017)

Table 3 Chemical reaction of SFGDA under different temperature changes

The pyrolysis process of CaSO3 in SFGDA also can be divided into five stages in air. Initially, the SFGDA undergoes the release of both free water and structured water below 473 K. Subsequently, Ca(OH)2 decomposes into CaO and H2O below 673 K. Between 673 and 953 K, the primary reaction observed is the oxidation of CaSO3, resulting in the formation of CaSO4 as the major product. In stage IV, as the temperature range up to 953–1073 K, the CaO and CO2 will be generated by CaCO3 component in the SFGDA. During stage V, CaSO4 decomposes, yielding CaO and SO2, which mirrors the reactions observed in an inert atmosphere. In comparison to the former stages, the generation of CaSO4 from CaSO3 in the SFGDA during the oxidation stage is not conducive to low-temperature and high-efficiency decomposition.

Figure 4 (a) illustrates the decomposition efficiency and f-CaO content of CaSO3 in SFGDA following calcination. At 1073 K, in SFGDA undergoes decomposition, resulting in the production of CO2 and CaO. And then, CaO could rapidly adsorb surrounding SO2 to synthesize CaSO3, which reduces the decomposition efficiency of CaSO3. The calcined product of SFGDA in an argon atmosphere reveals the presence of CaS between 1073 and 1373 K, indicating the decomposition of calcium sulfite and the formation of CaS and CaSO4. Evidently, achieving comprehensive desulfurization of CaSO3 in an inert atmosphere presents a challenge due to the occurrence of Eq. 8.

Fig. 4
figure 4

Effect of temperature on decomposition of SFGDA and phase transition in Ar. (Yang et al. 2019a, b)

$$4{\mathrm{CaSO}}_{3}(\mathrm{s})\to \mathrm{CaS}(\mathrm{s})+3{\mathrm{CaSO}}_{4}(\mathrm{s})$$
(8)

Finally, through thermogravimetric non-isothermal experimental research, it was observed that the initial decomposition temperature of analytically pure anhydrous CaSO4 powder in a high-purity N2 atmosphere is 1433 K, while the final decomposition temperature is 1603 K (Ping et al. 2004). Concomitantly, the results indicate that the initial decomposition temperature is 1506 K and the final reaction temperature is 1678 K under an air stream. It is intriguing to note that the prolonged duration of the initial and final reactions is facilitated by an elevated oxygen concentration. Maintaining a specific oxygen concentration in the reaction gas leads to the decomposition of approximately 92 wt.% of CaSO4. As Eq. 5, achieving a higher desulfurization rate poses significant challenges due to the partial coating of CaO on the unreacted CaSO4, thereby impeding the complete decomposition of CaSO4.

Considering the current predicament of incomplete and high-temperature decomposition of CaSO4 in SFGDA, there is an urgent need to reduce the initial reaction temperature for synthesizing CaO with minimal energy consumption. Such an endeavor would signify a noteworthy advancement in the quest for sustainable and energy-efficient methodologies in the production of materials possessing desirable attributes.

SFGDA decomposed by reducing gas

Given the distinctive attributes of SFGDA, a plethora of applications can be envisaged for enhancing gas decomposition processes in the realm of low-temperature pyrolysis. Especially, reducing gas (Robbins 1966; Tian et al. 2010; Yang et al. 2019a, b; Shen et al. 2008; Diaz-bossio et al. 1987; Zheng et al. 2017; Song et al. 2008; Song et al. 2009; Sunphorka et al. 2019; Zhang et al. 2013; Simei et al. 2017), CO is considered to be of particular efficiency to reduce and decompose CaSO3 and CaSO4 to prepare CaO, CaS and SO2 excellently.

Formation of CaO by reducing gas

In a previous investigation undertaken by Robbins (1966), the adsorption of gas and polymorphic transitions were scrutinized throughout the CaSO4 decomposition process employing reducing gas for CaO synthesis. The production of CaO can be conceptualized as involving dissociation and nucleation phenomena. According to the findings of the adsorption experiments, it was noted that CO does not display notable adsorption onto the CaSO4 surface; instead, it undergoes direct surface reactions. The gaseous CO molecules engage in surface reactions with oxygen molecules present on the CaSO4 surface, resulting in the adsorption and formation of SO2 and CO2 compounds on CaO, as depicted in Eq. 9. And then, CO2 would rapidly desorb into the gas phase, as Eq. 10. The final control step is the direct desorption of SO2 and the formation of CaO nuclei, as Eq. 11. Once an adequate number of CaO nuclei are formed, the decomposition rate increases. As hypothesized by Robbins (1966), the surface of the CaO core offers a rapid pathway for the desorption of CO2 and SO2, facilitating the rapid growth of the nucleus and enhancing the release of SO2. Thus, the desorption rate is heavily influenced by factors such as gas velocity, particle size, and CO concentration.

$$\mathrm{CO}(\mathrm{g})+{\mathrm{CaSO}}_{4}\to \mathrm{CaO}\cdot {\mathrm{SO}}_{2}\cdot {\mathrm{CO}}_{2}$$
(9)
$$\mathrm{CaO}\cdot {\mathrm{SO}}_{2}\cdot {\mathrm{CO}}_{2}\to \mathrm{CaO}\cdot {\mathrm{SO}}_{2}+{\mathrm{CO}}_{2}(\mathrm{g})$$
(10)
$$\mathrm{CaO}\cdot {\mathrm{SO}}_{2}\to \mathrm{CaO}+{\mathrm{SO}}_{2}(\mathrm{g})$$
(11)

Conversely, it was observed that the reductive decomposition rate of CaSO4 diminishes as the gas velocity surrounding the particles increases. According to Robbins’s proposed mechanism for CaSO4 reduction, it is suggested that a portion of the SO2 species undergoes reduction to elemental S on the CaO surface, leading to a reduction in the SO2 concentration at the interface. It is plausible that higher flow rates could contribute to an elevation in the SO2 concentration on the CaO surface, considering the abundance of SO2 in the gas stream. The higher concentration of SO2 on the surface hinders the direct desorption of SO2 from CaSO4 and the nucleation of CaO. Additionally, decreasing the particle size increases the maximum desulfurization rate of CaSO4, while having no significant impact on the initial desulfurization rate. On the other hand, Robbins (1966) and Wheelock (Wheelock 1960) stated that an increase in CO2 concentration slightly decreases the maximum desulfurization rate, but does not noticeably affect the initial rate. Furthermore, it is important to maintain a low concentration of CO in the CO-CaSO4 reaction system, as excessive CO concentration is unfavorable for the formation of CaO. Zhao et al. (Zhao et al. 2016) investigated the relationship between SO2 concentrations, the fraction of released S, and different CO concentrations, specifically 0.10%, 0.25%, 0.50%, and 1.00%, Fig. 5.

Fig. 5
figure 5

Effect of CO concentrations on SO2 concentrations and fraction of S released. (Zhao et al. 2016)

The results demonstrated that as the concentration of CO increases, the release of SO2 occurs earlier, and the amount of sulfur released is directly proportional to the CO concentration. Specifically, under 1.00% CO, more than 21.07% of S was released, approximately double when under 0.25% CO. In a separate study, Okumura et al. (Okumura et al. 2003) investigated the impact of CO concentration (2%, 5%, and 10%) on CaSO4 desulfurization at 1273 K. As Fig. 6(1) shown, it was observed that the initial decomposition reaction rate of CaSO4 increases and the desulfurization rate equilibrium point time of CaSO4 is shortened with an augmentation in the CO concentration. Nevertheless, the desulfurization rate value is diminished by the increase of CO, implicating low CaO productivity (Zheng et al. 2011a, b; Zhang et al. 2012; Kuusik et al. 1985). the conversion of CaSO4 to CaO is contingent upon the concentrations of CO and CO2, with a favorable outcome observed when the CO concentration is low and the CO2 concentration is high. Furthermore, Gruncharov et al. (Gruncharov 1985) and Okumura et al. (2003) have indicated that the initial decomposition rate of CaSO4 decreased when the CO2 concentration was varied from 10 to 30 vol.% while maintaining a constant CO concentration of 2.00 vol.%. A comparison between Fig. 6(1) and Fig. 6(2) elucidates that the conversion of CaSO4 to CaO is contingent upon the concentrations of CO and CO2, with a favorable outcome observed when the CO concentration is low and the CO2 concentration is high.

Fig. 6
figure 6

The effects of CO, CO2, PCO / PCO2 and reaction time on desulfurization efficiency: (1) effect of the concentration of CO; (2) CO2 on the decomposition of CaSO4 at 1273 K; (3) Effect of PCO/PCO2 on the final conversion of CaSO4 to CaO; (4) Effect of reaction time on CaSO4 decomposition. (Okumura et al. 2003)

By integrating the methodologies employed by Oh et al. (1990) and Okumura et al. (2003), the pseudo-equilibrium desulfurization rate (wt.%) and reduction potential PCO/PCO2 are graphed. Figure 6(3) illustrates that as the PCO/PCO2 value decreases, the conversion (X) increases. At low reduction potentials (PCO/PCO2 < 0.10), CaSO4 exhibits near-complete regeneration to CaO. However, discrepancies in the X values are observed, particularly for PCO/PCO2 > 0.2. These discrepancies may arise due to variations in the reaction conditions employed. Moreover, as depicted in Fig. 6(4), the CaO obtained from the decomposition of CaSO4 demonstrates superior performance in terms of SO2 and CO2 re-adsorption compared to the CaO derived from limestone decomposition. At 1273 K, an apparent conversion value of 91% for the decomposition of CaSO4 to CaO was obtained in a 2 vol.% CO and 30 vol.% CO2 atmosphere.

Irrespective of the reduction reaction of CaSO4 in the CO system, CO-N2 system, or CO-CO2-N2 system, the interaction between CO and CaSO4 resulting in the formation of CaO is accompanied by a highly endothermic reaction. (Kuusik et al. 1985; Gruncharov et al. 1985; Ghardashkhani et al. 1991). The desorption of SO2 and the formation of CaO necessitate 66 kcal/g·mol, as indicated by Wheelock's (1960) equilibrium measurements. Whereas, the initial reaction between CO and CaSO4, resulting in the production of CO2 and adsorbed SO2, releases 27 kcal/g·mol. Therefore, the total heat required for the overall reduction and decomposition of CaSO4 and CO is 39 kcal/g·mol.

In practical engineering applications, incorporating an appropriate quantity of O2 into the reaction environment can elevate the reaction temperature through the exothermic interaction between O2 and CO. The introduction of O2 not only adjusts the partial pressure ratio of CO2 to CO in the reaction atmosphere but also modifies the reaction mechanism associated with the reduction of CO and the decomposition of CaSO4 into CaO. Xia et al. (Xia et al. 2022) investigated the impact of O2 on the regeneration process of CaO from CaSO4 through CO reduction. In this investigation, the reduction and decomposition characteristics of CaSO4 were examined and compared under various process conditions in a laboratory-scale fixed-bed reactor utilizing CO-N2/CO-CO2-N2 atmospheres. The results revealed that CaSO4 can be completely decomposed into CaO when the reaction temperature exceeds 1273 K, CO% ≥ 2, and P(CO2)/P(CO) ≥ 8. Furthermore, the addition of an appropriate amount of O2 enhances the yield of CaO in the product. In the O2-CO-N2 atmosphere with O2% of 7 and CO% of 16, CaSO4 can be entirely decomposed into CaO without the need for CO2 supplementation. The physical properties of CaO produced through the reduction and decomposition of CaSO4 exhibit superior characteristics compared to those obtained through the calcination of CaCO3, which aligns with the conclusions of Okumura (2003).

In addition to the aforementioned factors, Wheelock (1960) investigated the influences of temperature, particle size, gas flow rate, and SO2 concentration in the gas stream. He concluded that as the temperature increases, the sulfur removal efficiency improves, but accompanied by higher energy consumption, as shown in Fig. 7(a). Moreover, as shown in Fig. 7(b), the maximum desulfurization rates, ranging from 0.1 to 0.3 lb·s−1·m−2, exhibited an almost linear increase with the increase of mass velocity in the 2%CO-98%N2 system. Additionally, as depicted in Fig. 7(c), the initial desulfurization rate showed little dependence on particle size, whereas the maximum desulfurization rate experienced a sharp decline as particle size increased. Hence, it is most probable that the rate is controlled by internal diffusion of the reaction. Eventually, the highest desulfurization rate was achieved at three distinct levels of CO concentration. In Fig. 7(d), it can be witnessed that at a CO concentration of 2%, the maximum desulfurization rate initially decreased with increasing SO2 concentration from 0 to 0.5%, but subsequently exhibited a slight increase with further increases in SO2 concentration.

Fig. 7
figure 7

Effects of temperature, particle size, flow rate and SO2 concentration in gas stream on CaSO4 desulfurization rate (wt.%). (Wheelock 1960)

At a CO concentration of 3%, the effect was reversed. And a level of 4%, a different effect was still observed. Although the results are scattered, the maximum desulfurization rate appears to increase slightly with increasing SO2 concentration across the entire investigated range. A minor concentration of SO2 in the feed gas exerts a substantial impact on the initial desulfurization rate.

Formation of CaS by reducing gas

Undoubtedly, both reaction temperature and CO concentration have a significant impact on the formation of undesired CaS during the desulfurization process. Wheelock (1960) determined that the decomposition reaction into CaS is negligible at temperatures below 1422 K, but the reaction can be entirely avoided if the temperature reaches 1478 K. In addition, the production of CaS is also suppressed when the PCO/PCO2 value in the system is low. To obtain high-purity CaO, it is essential to comprehend the mechanism of CaS generation and manipulate the reaction conditions to minimize its formation. In contrast to the formation reaction of CaO, CaS can be produced when CaO⋅SO2 continues to react with CO prior to SO2 desorption.

According to the mechanism of CaSO4 decomposition proposed by Robbins (1966), the most stable intermediate in reductive decomposition is SO2 adsorbed on CaO, and the adsorbed SO2 can be reduced to elemental sulfur adsorbed on CaO, Eq. 12.

$$\mathrm{CaO}\cdot{\mathrm{SO}}_2+2\mathrm{CO}\left(\mathrm g\right)\rightarrow\mathrm{CaO}\cdot S+{2\mathrm{CO}}_2\left(\mathrm g\right)$$
(12)

Both Wheelock (1960) and Xia et al. (2022) have demonstrated that at temperatures exceeding 1273 K and with a high concentration of CO2, CO2 can displace S on the surface of CaO and be adsorbed. This process effectively hinders the formation of CaS (as described in Eq. 13) by reducing the available amount of adsorbed sulfur for reaction with CO.

$$\mathrm{CaO}\cdot S+\mathrm{CO}\left(\mathrm g\right)\rightarrow\mathrm{CaS}+{\mathrm{CO}}_2\left(\mathrm g\right)$$
(13)

SFGDA decomposed by solid reductant

Several researchers have investigated the reductive decomposition of CaSO4 using a solid reductant, including coke (Yan et al. 2014; Kale et al. 1992; Van der Merwe et al. 1999; Ma et al. 2010), lignite (Zheng et al. 2013a, b), charcoal (Chaalal et al. 2020), biomass (Rebbling et al. 2016; Zheng et al. 2011a, b), and graphite (Zheng et al. 2013a, b; Böke et al. 2002; Feng et al. 2019; Ma et al. 2012; Zhou et al. 2011). Presently, there is no universally accepted comprehensive theory that explains the mechanism of thermal decomposition of CaSO4. However, it is recognized that both solid–solid and gas–solid reaction mechanisms are involved in the reduction of CaSO4 to CaO by C.

Formation of CaO by solid reductant

Initial period, it was observed that the decomposition temperature of CaSO4 could be decreased in the presence of coke in the reaction mixture. Figure 8 presented by Factsage 8.1 demonstrates that when delta G < 0 kJ∙mol−1, the reduction of CaSO4 by C can occur spontaneously, resulting in the production of CaO, CO2 and SO2, as shown in Eq. 14. The CO generated from the reaction between C and CO2 is participated in the decomposition of CaSO4 to CaO at 1200 K approximately, as shown in Eq. 15.

Fig. 8
figure 8

Delta G − T relationship lines of the major chemical reactions

Hence, based on the findings presented in Fig. 8, it can be inferred that the conversion of CaSO4 to CaO and SO2 through C takes places at a higher temperature compared to the decomposition of CaSO4 to CaS, indicating a solid–solid reaction mechanism, as described in Eq. 16. Consequently, at a temperature of 1453 K, the reaction between CaS and CaSO4 leads to the formation of CaO and SO2, as illustrated in Eq. 17. Moreover, CaSO3 undergoes decomposition to CaS in the presence of C, as demonstrated in Eq. 18.

In accordance with the gas–solid process theory, the initial step involves the oxidation of C to CO, whereupon fully contacts CaSO4 to generate CaO. The reaction between C and CaSO4 do not instantaneously but rather proceeds until completion. Previous studies by Ma et al. (2010) and Su et al. (Su et al. 2019) have highlighted the influence of temperature and C content on the decomposition of CaSO4. Supported by thermodynamic analysis, Ma et al. (2010) determined that the main reaction represented by Eq. 14 could occur within a temperature range of 1273–1423 K. It appears that maintaining a prolonged high temperature and utilizing raw materials with low carbon content, or operating under weak reducing conditions, promotes the decomposition of CaSO4 into CaO. Additionally, Jia et al. (Jia et al. 2016) and Zheng et al. (2011a, b) also investigated the reduction of CaSO4 to CaO using lignite, coal, biomass, and other materials as reducing agents.

To achieve a substantial concentration of SO2 gas, CaSO4, CaS, S, and FeS2 can be counted as an abundant sources of sulfur to enhance sulfuric acid production. The interaction between CaSO4 and CaS or S is recognized as a solid-state reaction mechanism, leading to the generation of SO2 in the gaseous phase. Song et al. (Song et al. 2019) conducted an analysis, affirming the technical feasibility of utilizing FeS2 as a solid reductive agent in the decomposition of CaSO4 for the production of SO2. It is noteworthy that the SO2 capture is not limited to FeS2 and CaSO4, as Ca2Fe2O5 and CaO can also be incorporated into the decomposition slag, thereby enhancing the economic feasibility of the process. The main solid–solid reactions between CaSO4 and FeS2 are as follows: Eq. 20 and Eq. 21. Significantly, the conducive conditions for obtaining low-sulfur solid products are hindered by the occurrence of Eq. 20. To effectively eliminate residual sulfur from the slag, it is imperative to comprehensively consider the maintenance of an appropriate oxygen pressure within the reaction system.

$$4CaS{O}_{4}(s)+2Fe{S}_{2}\to C{a}_{2}F{e}_{2}{O}_{5}(s)+CaO+5S{O}_{2}(g)+CaS(s)$$
(20)
$$7CaS{O}_{4}(s)+2Fe{S}_{2}\to C{a}_{2}F{e}_{2}{O}_{5}(s)+5CaO+9S{O}_{2}(g)$$
(21)

Formation of CaS by solid reductant

Based on the thermodynamic calculations illustrated in Fig. 8, it can be inferred that the reduction of CaSO4 and CaSO3 to CaS can be accomplished using C and CO at temperatures below 1173 K. Li et al. (Li et al. 2015) revealed that the thermal decomposition of CaSO3 to produce CaS employing coal as a reducer, was carried out with a molar ratio of C to CaSO3 at 2.5 (n(C)/n(S) = 2.5), a reaction temperature of 1173 K, a reaction time of 1.5 h and a coal particle size 100 mesh, shown in Fig. 9. Both thermodynamic calculations and experimental results indicate that the formation of CaS during CaSO3 decomposition can be attributed to Eq. 17 and Eq. 18, involving solid–solid and gas–solid reactions. Additionally, according to Eq. 16, CaS is consumed and reacts with the remaining CaSO4 to produce CaO. XRD and SEM analyses conducted by Li (2015) have validated that CaS is the only solid product formed under appropriate reaction conditions. Consequently, in the reaction system where CaSO3 is reduced by coal with an n(C)/n(S) ratio of 2.5, the decomposition of CaSO3 to CaO can be considered almost negligible.

Fig. 9
figure 9

Effects of holding time and coal particle size on CaSO4 decomposition (Jia et al. 2016)

Analogously, Jia et al. (2016) discovered that the presence of coal significantly enhances the high-temperature decomposition of CaSO4, resulting in the decomposition of 87% CaSO4 into CaS, which is on account of the solid–solid reaction between C and CaSO4. Additionally, the duration of calcination and the particle size of coal influence the decomposition of CaSO4. As indicated in Fig. 9, the decomposition of CaSO4 is more favorable when a reducing agent with an extended reaction time and finer particle size is employed. This observation aligns with the formation reaction of CaS and CaO.

Realistically, an elevation in carbon content tends to promote the formation of CaS, which is unfavorable for sulfur removal. Su et al. (2019) conducted an experimental investigation utilizing a tubular resistance furnace to to explore the influence of carbon content on the desulfurization rate at various temperatures. They observed that an increase in the n(C)/n(CaSO4) ratio did not result in an improvement of the desulfurization rate at the same temperature, as depicted in Fig. 10(a). Elevated temperature plays a pivotal role in augmenting the desulfurization rate, as indicated by the reaction mechanism wherein decomposed CaS reacts with CaSO4 to form CaO, and the presence of CO in the gaseous phase accelerates the conversion of CaSO4 to CaO. These two factors are recognized the primary catalysts for improving the desulfurization rate.

Fig. 10
figure 10

Variation curve of desulfurization rate with carbon content at different temperatures and gas phase pristine test results. (Su et al. 2019)

Tian, Zheng, and Zhong et al. (Tian et al. 2010; Diaz-bossio et al. 1985; Simei et al. 2017) have investigated that CaSO4 serves as a promising oxygen carrier, readily reducible to form CaS under high concentrations of CO, while remaining stable in a reducing atmosphere. Within the C-CaSO4 reaction system, the concentration of CO gradually increases as the reaction proceeds, and the potency of the reducing environment is contingent upon the carbon content in the raw material. Furthermore, the conversion rate of CaSO4 and CO into CaS is higher at temperatures below 1273 K compared to their reaction in the solid-state, which results in the production of CaO.

More formidably, the residual of CaS does not contribute to effective sulfur removal in SFGDA, resulting in low CaO production. Therefore, the conversion of CaS to CaO is extremely important to achieve a high desulfurization rate correspondingly. Propitiously, the introduction of O2 into residual CaS deemed feasible for industrial application, leading to additional CaO generation (Anthony et al. 2003; Turkdogan et al. 1974; Song et al. 2007). The oxidation of pure CaS (with a particle size of 45 µm) was investigated by Davies, Yrjas, and Marban et al. (Davies et al. 1994; Yrjas et al. 1997; Marban et al. 1999) via an atmospheric thermobalance. They observed that the predominant amount of SO2 was liberated within the initial 25-min period at temperatures ranging from 1123 to 1323 K. Upon oxidizing CaS with 4.2 vol% O2 for 30 min at 1123 K and 1223 K, the reaction described in Eq. 22 occurred, resulting in CaSO4 conversion rates of approximately 8% and 32%, separately. Meanwhile, the oxidation reaction of CaS and O2 can be represented by Eq. 23, with corresponding CaO conversion rates of approximately 2% and 30%, respectively. Davies et al. (1994) attributed the release of SO2 to the apparent solid–solid reaction between CaS and CaSO4, which was confirmed in their study. Furthermore, they discovered that no solid–solid reaction occurred at 1123 K, whereas at 1323 K, complete conversion to CaO was achieved, indicating the transformation from CaS to CaO. Outstandingly, Yrjas et al. (1997) detected that excessively high concentrations of O2 immediately reacted with CaS, forming impermeable product layers of CaSO4 or CaO, thereby impeding further sulfur release reactions.

$$CaS\left(s\right)+2{O}_{2}\left(g\right)\to CaS{O}_{4}$$
(22)
$$CaS\left(s\right)+3/2{O}_{2}\left(g\right)\to CaO+S{O}_{2}\left(g\right)$$
(23)

Maintaining a moderately low level of O2 concentration and a temperature greater than 1323 K will benefit the conversion of CaS to CaO (Turkdogan et al. 1974; Song et al. 2007; Wang et al. 2020). Thence, optimizing factors such as the reduced particle size of the SFGDA, thorough mixing of C, maintaining a reaction temperature above 1323 K, and utilizing low C content prove more advantageous in preventing the formation of CaS, thus enabling the production of high-purity CaO.

SFGDA decomposed by compound additive

Apart from solid reductant, solid iron oxides, which are non-reducing agents, also play a significant role in facilitating the decomposition of SFGDA. Higuchi et al. (Higuchi et al. 2016) proposed a method for synthesizing calcium ferrite from waste gypsum board (mainly composed of CaSO4·2H2O) by incorporating iron oxides. Through thermodynamic estimations and small-scale heating tests, three distinct synthetic routes for calcium ferrite were identified: (1) calcium ferrite production in an air environment, (2) single-step heat treatment, and (3) two-step heat treatment, as shown in Fig. 11. The results indicated that calcium ferrite is partially synthesized through the decomposition of the gypsum board in an air atmosphere at 1453 K, but achieving complete decomposition of CaSO4 is exceedingly challenging. Equation 24 is considered the primary reaction, indicating that reducing the PO2 or PSO2 contributes to the near-complete formation of calcium ferrite.

Fig. 11
figure 11

The phase diagram of chemical equilibrium in the Ca-Fe-S–O system (Higuchi et al. 2016)

$$xCaS{O}_{4}(s)+yF{e}_{2}{O}_{3}(s)\to xCaO\cdot yF{e}_{2}{O}_{3}(s)+xS{O}_{2}(g)+0.5x{O}_{2}(g)$$
(24)

As a consequence, through the addition of C in a specific proportion, the PO2 decreases, resulting in desulfation rates of over 95% for CaSO4. Subsequently, there would be more calcium ferrite generated than processes if SO2 was removed after reaction and slightly oxidative atmosphere in the reaction system was maintained throughout the reaction period (Zhao et al. 2020; Liu et al. 2021). Although some CaSO4 in waste gypsum board decomposes into CaS, it fortunately undergoes oxidation to yield CaO in the presence of trace amounts of oxygen. Whereafter, a portion of the calcium ferrite is produced through the reaction between CaO and Fe2O3, as described by Eq. 25 and Eq. 26.

$${\mathrm{CaSO}}_{4}(\mathrm{s})+{\mathrm{Fe}}_{2}{\mathrm{O}}_{3}(\mathrm{s})+1/2\mathrm{C}(\mathrm{s})\to \mathrm{CaO}\cdot {\mathrm{Fe}}_{2}{\mathrm{O}}_{3}(\mathrm{s})+{\mathrm{SO}}_{2}(\mathrm{g})+1/2{\mathrm{CO}}_{2}(\mathrm{g})$$
(25)
$$2\mathrm{CaS}(\mathrm{s})+{\mathrm{Fe}}_{2}{\mathrm{O}}_{3}(\mathrm{s})+3{\mathrm{O}}_{2}(\mathrm{g})\to 2\mathrm{CaO}\cdot {\mathrm{Fe}}_{2}{\mathrm{O}}_{3}(\mathrm{s})+2{\mathrm{SO}}_{2}(\mathrm{g})$$
(26)

Actually, Eq. 27 to Eq. 30 demonstrate that carbon would reduce Fe2O3 to Fe and simultaneously convert CaSO4 into CaS and CaO. Subsequently, Fe is oxidized to Fe2O3 by O2, which then interacts with CaSO4 to form calcium ferrite.

Meanwhile, C undergoes additional oxidation and is consumed in the processes of iron oxide reduction. Further research on low-carbon consumption is warranted. Gratifyingly, the burning of the C reaction and the oxidation of iron provide a significant amount of energy to the reaction system.

$$3C(s)+F{e}_{2}{O}_{3}(s)\to 2Fe(s)+3CO(g)$$
(27)
$$3CO(g)+F{e}_{2}{O}_{3}(s)\to 2Fe(s)+3C{O}_{2}(g)$$
(28)
$$3/2C(s)+F{e}_{2}{O}_{3}(s)\to 2Fe(s)+3/2C{O}_{2}(g)$$
(29)
$$CaS{O}_{4}(s)+2Fe(s)+{O}_{2}(g)\to CaO\cdot F{e}_{2}{O}_{3}(s)+S{O}_{2}(g)$$
(30)

To reduce carbon consumption, the two-step heat treatment is considered a reasonable and comprehensive approach. Higuchi et al. (2016) proposes the decomposition of gypsum board with carbon as the first step, leading to the formation of CaO or a small amount of CaS. The second step involves the formation of calcium ferrite from the decomposed material and Fe2O3, resulting in the production of calcium ferrite with varying breakdown-product contents. Equation 31 illustrates this process.

$$zCaO(s)+wF{e}_{2}{O}_{3}(s)\to zCaO\cdot wF{e}_{2}{O}_{3}(s)$$
(31)

Two distinct processes are present: single-step and two-step processes, with the latter having the potential to reduce carbon consumption theoretically, as depicted in Fig. 12. Furthermore, the two-step process is more advantageous for synthesizing calcium ferrite with low sulfur content compared to the one-step process. Notwithstanding, the two-step process is complicated with high expenses and high facility requirements, and so whether it is environmental protection, energy-saving or economy should be considerately thought in future industrial applications. On the contrary, the one-step production of calcium ferrite involves challenging conditions for precise control of process parameters. Nevertheless, Route remains the preferred choice for commercial production due to its straightforward process flow.

Fig. 12
figure 12

Diagram of one-step or two-step synthesis of calcium ferrite (Zhao et al. 2020; Liu et al. 2021)

Based on the study of Higuchi (2016), calcium ferrite also was prepared from desulfurization gypsum, iron ore, blast furnace dust, and graphite by Zhao (2020). As shown in Fig. 13(a), the raw materials for the reaction were initially introduced into a tube furnace in both powder and block forms, followed by calcination at a temperature of 1373 K under an air atmosphere. It was observed that the powder exhibited a superior desulfurization effect compared to the block form. The decomposition and desulfurization process of CaSO4 involve both solid–solid and gas–solid reactions, with the latter exhibiting significantly faster kinetics than the former. The presence of gaps between the powder particles facilitates the entry of reducing gases such as CO, allowing for gas–solid reactions with CaSO4. Consequently, the use of powdered reactants promotes sulfur removal and leads to a higher desulfurization rate.

Fig. 13
figure 13

Curve of desulfurization rate and sulfur content with the ratios of raw materials under different conditions: (a) different morphologies, (b) different atmospheres, (c) different carbon content, (d) different iron oxide content. (Zhao et al. 2020)

It can be seen from the Fig. 13(b) that when the molar ratio of CaO (CaSO4) to Fe2O3 is constant, the desulfurization rate of gypsum in argon atmosphere is significantly higher than that in air atmosphere. The sulfur content in the product is lower than that in air atmosphere under argon atmosphere, illuminated that the desulfurization reaction of gypsum hardly occurs in an oxidizing atmosphere. The higher n(CaO): n(Fe2O3) contributes to the lower the desulfurization rate of gypsum in argon, which shows that the increase of iron oxide promotes the decomposition and desulfurization of gypsum.

As shown in Fig. 13(c), it is evident that maintaining a constant ratio of n(CaO) to n(Fe2O3), a higher carbon content corresponds to a higher desulfurization rate. However, the desulfurization rate shows minimal changes beyond a carbon content of 30%, with the highest achieved rate being approximately 95%. As Fig. 13(d), when the carbon content is fixed and the ratio of n(CaO): n(Fe2O3) is 1:1, the product exhibits the highest amount of calcium ferrite, exceeding 95%. Ultimately, the optimal reaction conditions for the one-step synthesis of pure calcium ferrite involve using powdered raw materials in an inert atmosphere, with a CaO/Fe2O3 ratio of 1:1 and a carbon content of 30%. To mitigate the presence of CaS and minimize its impact on the desulfurization rate, Zhao et al. (2020) controlled the temperature at 1423 K and utilized the following composition: 30.4 wt.% desulfurization gypsum (with CaSO4 as the primary component), 31.7 wt.% iron ore, 11.1 wt.% blast furnace dust, and 26.8 wt.% graphite. Experimental results demonstrated a significant formation of calcium ferrite under these optimal reaction conditions, resulting in a desulfurization rate of 98.9% and suggesting its feasibility as a recyclable sintering additive.

Summary and perspective

The production of extensively purified CaO and the concentration of SO2 through pyrolysis are widely acknowledged as promising strategies that have captivated an increasingly large cohort of researchers in their quest to solve the issue of harmless, bulk, and value-added disposal of SFGDA. Achieving efficient and low-temperature sulfur removal in SFGDA necessitates the establishment of favorable reaction conditions. Currently, investigations are underway to optimize the desulfurization efficacy of semi-dry desulfurization ash by controlling variables such as the reaction atmosphere, temperature, reaction duration, and the type and dosage of additives.

SFGDA demonstrates suboptimal pyrolysis efficiency when subjected to temperatures of 1678 K in an air and 1603 K in an N2, respectively. Furthermore, the presence of CaSO4 significantly prolongs the duration required for complete decomposition of SFGDA, thereby diminishing the desulfurization efficacy of CaSO4 and raising the temperature threshold for sulfur removal. This impedes the large-scale industrial production of CaO and SO2. Therefore, there is a pressing need to identify a low-temperature, high-efficiency, and environmentally friendly desulfurization additive that enables the recycling of chemical substances such as S and Ca.

As discussed previously (Fig. 14 and Table 4), the decomposition temperature of CaSO4 in the SFGDA can be reduced, leading to an improved desulfurization rate under a reducing atmosphere. In the CO-N2 atmosphere, the CaS is more effortlessly formed blew 1273 K than CaO generated from SFGDA. This unfavorable occurrence hinders the reduction of sulfur content in solid products. Relatively speaking, there is almost 100% desulfurization rate or decomposition rate in the CO-CO2-N2 and CO-CO2-SO2-N2 at 1323–1423 K, but the unalterable truth is that the reduction reaction of CaSO4 with CO is a strongly endothermic reaction, requiring more external heating. Therefore, in practical industrial applications, it is high-priority to facilitate endothermic reactions through the exothermic reaction between O2 and CO. What's more, CaS generated from the reduction of CaSO4 can be oxidized by O2 at 1323–1373 K, reducing residual sulfur and improving desulfurization rate. It is worth noting that the complexity of the system can be simplified by eliminating the presence of SO2 or CO2. However, maintaining CO pressure stablely and combustion safely is a tricky issue for industrialization process henceforth.

Fig. 14
figure 14

The synthetic route of CaO and Ca2Fe2O5 with various additives

Table 4 Summary of current research status

It is noteworthy that the endeavor to reduce CaSO4 with CO for the production of CaO is currently at the laboratory stage, primarily due to its high raw material cost and expensive equipment investment. In contrast, solid-phase reducing agents such as coke, graphitic carbon, high-sulfur carbon, pyrite, and CaS are more economically viable options for the reductive decomposition of SFGDA. When the molar ratio of CaSO4 to C is 1:0.8, CaSO4 can be effectively decomposed to yield predominantly CaO at temperatures exceeding 1273 K, culminating in a desulfurization rate of over 95%. Additionally, CaSO4 is reduced to form CaO and CaS calcined at 1273 K during the reaction with FeS2, and Fe2O3 is combined with CaO to synthesis calcium ferrite product reused as a sintering additive. The S in the refractory pyrite and SFGDA is migrated to the gas phase, enabling the separation of the sulfur and calcium iron salts. This extraction and utilization of low-grade pyrite not only demonstrates the effective utilization of resources but also holds promising prospects for future applications.

Analogous to sintering flue gas desulfurization gypsum (the main component is CaSO4), the SFGDA decomposition product CaO is combined with Fe2O3 to synthesize calcium ferrite using a rotary kiln device (as Fig. 15). Nonetheless, the decomposition temperature of CaSO4 is immobile high, and the desulfurization rate is not satisfactory. By introducing a solid carbon source into the system containing CaSO4 and Fe2O3, the optimal decomposition temperature of CaSO4 can be lowered to 1373 K, and the carbon in the reaction system plays the role in providing heat source and reducing agent for the reaction. Within the C-CaSO4-Fe2O3 system, CaSO4 is deviously reduced by carbon to form CaS. Moreover, under normal pressure, CaSO4 can be reduced to CaO in a slightly oxidizing atmosphere at a temperature of 1323 K, and then combined with Fe2O3 to form calcium ferrite. Therefore, in the C-CaSO4-Fe2O3 reaction system, it is crucial to maintain a controlled atmosphere of a mildly reducing atmosphere within the rotary kiln device in order to obtain the calcium ferrite product at a lower temperature.

Fig. 15
figure 15

Industrial flow chart of SFGDA to produce calcium ferrite and sulfuric acid

Notwithstanding, despite the increasing interest in the synergistic preparation of calcium ferrite using calcium sulfate, iron, and carbon-containing solid waste materials, there is a lack of reports on small-scale industrial experiments. What’s more, the research on the reaction mechanism of C-CaSO4-Fe2O3 involves heterogeneous reactions, including gas–solid and solid–solid reactions, circumlocutionarily complicated. Thus far, there has been no clear mechanistic explanation for the C-CaSO4-Fe2O3 reaction system. Furthermore, there is a discrepancy between the results obtained from laboratory experiments and industrial tests, and the scarcity of qualified technical personnel has hindered the industrial application of the semi-dry method. Consequently, further verification and elucidation of the microscopic mechanism underlying the C-CaSO4-Fe2O3 reaction are essential to guide actual industrial production. Indispensablely, it is crucial to prioritize the technical training of relevant operators, as it significantly impacts the industrialization process of this method.

Approximately estimated, in the CO-CaSO4-N2, CO-CO2-CaSO4-N2, and C-CaSO4-N2 reaction systems without oxygen involvement, the reaction temperature is set at 1100℃, the calcination time is 4 h, and the power of the calciner is 2500 W, resulting in an external energy consumption of approximately 3.6 × 104 kJ, equivalent to 1.10 $/ton in terms of power cost. Introducing O2 to facilitate the combustion reaction of CO and coke provides heat for the decomposition of SFGDA, reducing this cost component.

Therefore, coke powder offers a cost advantage as a reducing agent. From a product perspective, if only carbon monoxide or carbon reducing agent is added, and the calcined products of SFGDA are all calcium oxide and sulfuric acid products, the benefits per ton of calcium oxide and sulfuric acid are 72.99 and 58.39 $/ton. In contrast, the calcium ferrite product obtained by adding carbon and iron oxide to the SFGDA brings a profit of 437.96 $/ton simultaneously, enabling the comprehensive utilization of calcium, iron, and sulfur resources. Henceforth, the preparation calcium ferrite from SFGDA produced during sintering holds significant potential for multitudinous, innocuous, and high-value consumption of SFGDA, paving the way for its promising commercial application.

Conclusions

Pyrolysis holds great promise as a technology to decompose SFGDA into CaO and SO2, facilitating the recycling of chemical elements. The purity and properties of the CaO product significantly vary depending on factors such as raw material particle size, composition of the reducing gas, types of additives, calcining temperature, and reaction termination time. In current research endeavors, it is a forefront challenge to achieve high efficiency and low temperature desulfurization for SFGDA to produce CaO at 1273–1373 K within 120 min. Appropriate particle size enhances the desulfurization efficiency of SFGDA, shorten the reaction time and improved the purity of CaO.

(1). SFGDA undergoes decomposition in the presence of reducing gas or a weak oxidizing atmosphere. CaSO4, the main refractory pyrolysis component in SFGDA, can be converted to CaO with a conversion rate of 0.91 in a 2%CO-30%CO2-68%N2 atmosphere. Under low reducing atmosphere (PCO/PCO2≤0.10), CaSO4 is completely decomposed into to CaO, but external heat is required for the decomposition process. Interestingly, the adding of 1.0% SO2 in a CO-CO2-N2 (PCO/PCO2≤0.10) the atmosphere achieves a 100% conversion rate of CaSO4 to CaO, and then external heating is needed to provide the required heat for the reaction. Propitiously, it is feasible to attain the necessary reaction temperature by utilizing the exothermic reaction between O2 and CO, enabling the complete conversion of CaSO4 to CaO in a 7% O2-16%CO-77% N2 atmosphere. Unfortunately, there are currently no reports on the industrial application of SFGDA for CaO production CaO due to the high cost of reducing gas, equipment investment, and security concerns.

(2). The addition of solid reducing agents such as C and FeS2 to SFGDA allows for a decrease in the temperature required for CaO preparation. Besides, these solid additives are more cost-effective compared to reducing gases. While more evidence is required to fully comprehend the precise action mechanism of C on SFGDA, it is recognized that the reaction between C and SFGDA involves solid-solid and gas-solid reactions. When C is added in the ratio of n(CaSO4): n(C) = 1: 0.8, the conversion rate of CaSO4 to CaO reaches 95.16% in an N2 atmosphere above 1273K, resulting in a significant amount of CaO product formation. FeS2 introduced to a reducing agent, the decomposition temperature of CaSO4 in the SFGDA can be reduced to 1093-1273 K under low-pressure conditions. However, the content of CaS in the solid product is relatively high, and when the temperature is higher than 1273 K, more encouraging to the decomposition of CaSO4 to form CaO.

(3) The addition of iron oxide as an auxiliary agent reduces the decomposition temperature of SFGDA, albeit to a small extent. Therefore, carbon-based reducing agents are often added simultaneously to promote the decomposition of CaSO4, reducing the reaction temperature below 1373K. Under the catalysis of composite additives, CaO can be completely generated, which combines iron oxide to produce calcium ferrite, enhancing the application value of the process. Ultimately, the one-step method of decomposing SFGDA and producing calcium ferrite represents an industrial pyrolysis technology for efficient and economical utilization of SFGDA.