Introduction

Sulfate ions are present in natural waters and industrial effluents of the chemical and metallurgical industries (Mulinari and Silva 2008; Silva et al. 2012). In most countries, sulfate concentrations in industrial effluents are set from 250 to 500 mg/l to protect the environment. In humans, prolonged consumption of water containing 500–750 mg/l sulfate causes laxative effects, catharsis, dehydration, and gastrointestinal irritation (Silva et al. 2012). Furthermore, high sulfate concentrations in water cause water salinization (Runtti et al. 2018), which is harmful to aquatic organisms that can only withstand limited ranges of water salinity (Nurmesniemi et al. 2021). Therefore, sulfate removal has been extensively investigated around the world by previous researchers (Bowell, 2004).

Technologies used for sulfate treatment in water include biological treatment (Chang et al. 2000; Kiran et al. 2017; Najib et al. 2017), and physicochemical treatment, which included membrane filtration (Lu et al. 2016; Oyewo et al. 2018), ion exchange (Gu et al. 2004; Călinescu et al. 2016; Arahman et al. 2017), adsorption (Hong et al. 2017), evaporation/crystallization (Tait et al. 2009; de Luna et al. 2017), electro-coagulation (Mamelkina et al. 2017, 2019; Rodrigues et al. 2020; Vepsäläinen and Sillanpää 2020; Nurmesniemi et al. 2021), and chemical precipitation (Benatti et al. 2009). The biological treatment employing sulfate-reducing bacteria is appropriate for low to moderate loadings, but it is hampered by a shortage of organics, high salinity, and H2S generation. Except for chemical precipitation, the physicochemical treatments were expensive and had disposal issues with concentrate or regenerating solution. (Tian et al. 2019). Among the mentioned treatment methods, only chemical precipitation is cost-effective for treating wastewater with high sulfate concentrations (Silva et al. 2012; Guimaraes and Leao 2014; Nurmesniemi et al. 2021). The chemical precipitation method is based on the formation of low soluble sulfate salts (Nairn et al. 1991). Chemical precipitation methods include calcium hydroxide precipitation to produce gypsum (Bowell 2004; Silva et al. 2012; De Godoi et al. 2017; Kefeni et al. 2017), barium salts precipitation to produce barite (BaSO4; pKsp = 9.9) (Bowell 2004; Bologo et al. 2012; Runtti et al. 2016; Kartic et al. 2018), and ettringite precipitation (Bowell 2004). Sulfate removal efficiency in the calcium hydroxide precipitation method is very low due to the relatively high solubility of gypsum (pKsp = 3.70) (Nairn et al. 1991; Dou et al. 2017; Runtti et al. 2018). The calcium hydroxide precipitation method can reduce the high concentration of sulfate to about 1200–2000 mg/l range (Silva et al. 2012; Kinnunen et al. 2017).

Despite the high sulfate removal efficiency by the barium chloride precipitation method, it is rarely used because barium chloride is more expensive than calcium hydroxide (Nairn et al. 1991; Tolonen et al. 2016) and toxicity of barium salts. Among the chemical precipitation methods (Tolonen et al. 2016), ettringite precipitation is a reliable and efficient treatment method (Nairn et al. 1991; Kabdaşl et al. 2015; Tolonen et al. 2016) because of its low solubility (pKsp = 111.6) (Almasri et al. 2015), which requires alkaline pH (Ferreira et al. 2011) for removing high sulfate concentrations (Almasri et al. 2015). The ettringite precipitation method is capable of reducing the sulfate concentration to less than 200 mg/l (Madzivire et al. 2010; Tolonen et al. 2016). In this method, ettringite precipitates by adding calcium hydroxide and aluminum salts to the effluent (Tolonen et al. 2016; Özkök et al. 2019). Among the many phases formed during the reaction of Ca2+, Al3+, and SO42− ions, only the mono-sulfate (3CaO.Al2O3.CaSO4.18H2O) and tri-sulfate (3CaO.Al2O3.3CaSO4.31H2O) are stable in the aqueous medium (Mehta and Kelvin 1966; Nair and Little 2009).

Ettringite is containing parallel columns of Al3+, Ca2+, and OH− (Sapsford et al. 2015). The channels between these parallel columns are composed of H2O and SO42− (Johnson 2004) which are essential for the structure of ettringite (Taylor 1973). The actual content of water molecules in the ettringite structure can vary from 24 to 32 mol per mol of ettringite (Tishmack and Burns 2004).

The three commercially available sulfate removal processes utilizing ettringite precipitation (Ca6Al2(SO4)3(OH)12.26H2O) are as follows:

  • The Savannah Mining-Mintek process (SAVMIN)

  • The Cost-Effective Sulfate Removal process (CESR)

  • The Outotec ettringite process.

The SAVMIN and the CESR are two commercial processes for sulfate concentration reduction using ettringite precipitation. These processes have been developed to reduce sulfate in wastewater effluents with sulfate concentrations above 2000 mg/l. Aluminum trihydroxide with recovery of the aluminum source, and a specific Al-containing chemical reagent obtained from cement products without recovery of the aluminum source, are used to precipitate ettringite in the SAVMIN process and the CESR process, respectively (Usinowicz and Monzyk 2006). The main disadvantage of the CESR method is the non-recovery of aluminum trihydroxide despite its high cost of aluminum salts and more mass of sludge (Liang et al. 2015). It seems that SAVMIN is an improvement of the CESR process; because it recycles the ettringite produced.

The reagent used in the Outotec process is sodium aluminate. The drawback of using this reagent is the high concentrations of sodium generated in the treated water effluent (Liang et al. 2015).

The SAVMIN process is known as the Walhalla Process and involves five steps as follows (Bowell 2004):

Metal hydroxide precipitation

This step involves the precipitation of metals such as magnesium by adding calcium hydroxide at an approximate pH of 12 (Maree et al. 1992). The reactions of metal hydroxide formation are (M: Metals):

$$ {\text{M}}^{{{2} + }} + {\text{ OH}}^{ - } \to {\text{ M}}\left( {{\text{OH}}} \right)^{ + } $$
(1)
$$ {\text{M}}\left( {{\text{OH}}} \right)^{ + } + {\text{ OH}}^{ - } \to {\text{ M}}\left( {{\text{OH}}} \right)_{{2}} $$
(2)
$$ {\text{M}}\left( {{\text{OH}}} \right)_{{2}} + {\text{ OH}}^{ - } \to {\text{ M}}\left( {{\text{OH}}} \right)_{{3}}^{ - } \left( {{\text{Feng 2}}000} \right). $$
(3)

Saturated calcium sulfates can be precipitated with the metal hydroxides and act as crystallization seeds and co-precipitates. Previous researches have shown that Mg2+ ions can react with OH‾ ions to form Mg(OH)42− which preferentially in the presence of SO42− and Al(OH)4‾ form hydrotalcite-type compound (Mg6Al2SO4(OH)16.nH2O) rather than ettringite (Shi et al. 2014; Dou et al. 2017). In the Mg2+ and Ca2+ competition for reaction with Al3+, the formation of a hydrotalcite-type compound by Mg2+ ions with higher consumption of Al3+ and OH‾ inhibits the formation of ettringite compound by Ca2+ ions and consequently inhibits further reduction in sulfate (Fang et al. 2018).

Gypsum precipitation/crystallization

This step involves the crystallization of sulfate in supersaturated solution in the presence of gypsum (Maree et al. 1992). The two steps of metal hydroxides and supersaturated gypsum precipitation incorporate as a preliminary pretreatment (Usinowicz 2006). The solubility of gypsum in water is 0.24 g per 100 ml (Seidell 1940; Lewry and Williamson 1994). The reaction of gypsum precipitation is as follow:

$$ {\text{Ca}}\left( {{\text{OH}}} \right)_{2} + {\text{ SO}}_{4}^{{2 - }} + {\text{ 2H}}_{2} {\text{O }} \to {\text{ CaSO}}_{4} .{\text{2H}}_{2} {\text{O }} + {\text{ 2OH}}^{ - } \;\left( {{\text{Tishmack 2}}004} \right) $$
(4)

One of the advantages of this process is the generation of a solid gypsum by-product, which can be sold to the gypsum industry to reduce process costs (Maree et al. 1992).

Adding gypsum seeds to the supersaturated solution of sulfate ions is a means to provide a sufficiently large growth surface area (Liu and Nancollas 1970; Choi et al. 2019) on which more molecules can assemble. It is energetically more favorable than forming a new nucleus (Singh and Middendorf 2007) to increase the purity, improve the crystal size (Qamar et al. 2012), accelerate the sulfate removal from the solution phase as gypsum, and reduce the induction time (Choi et al. 2019). The induction time depends on the ratio of super-saturation (He et al. 1994). The induction time decreases with an increase in the ratio of super-saturation of the solution (Lancia et al. 1999).

Ettringite precipitation

This step involves the sulfate precipitation in the form of ettringite by addition of aluminum hydroxide (Maree et al. 1992). In the formation of ettringite, a significant reduction in the sulfate concentration in the wastewater was observed at high pH (Smart et al. 2010). Adding Ca(OH)2 increases the Ca2+ and OH‾ ions concentration in the solution. Al3+ ions from Al(OH)3 produce [Al(OH)6]3− species which reacts with Ca2+ to form ettringite (Fang et al. 2018; Luo et al. 2019). Previous studies have shown that the optimal pH range for precipitation of stable ettringite is 11–12.5 (Chrysochoou and Dermatas 2006; Almasri et al. 2015). Sulfate and calcium concentrations reduce significantly with the precipitation of insoluble ettringite (Feng et al. 2000). The sulfate removal process by the ettringite precipitation method is described using the following chemical equilibrium reactions (Fang et al. 2018).

$$ {\text{Al}}\left( {{\text{OH}}} \right)_{{3}} + {\text{ OH}}^{ - } \to {\text{ Al}}\left( {{\text{OH}}} \right)_{{4}}^{ - } $$
(5)
$$ {\text{Al}}\left( {{\text{OH}}} \right)_{{4}}^{ - } + {\text{ 2OH}}^{ - } \to \, \left[ {{\text{Al}}\left( {{\text{OH}}} \right)_{{6}} } \right]^{{{3} - }} $$
(6)
$$ {2}\left[ {{\text{Al}}\left( {{\text{OH}}} \right)_{{6}} } \right]^{{{3} - }} + {\text{ 6Ca}}^{{{2} + }} + {\text{ 24H}}_{{2}} {\text{O }} \to \, \left\{ {{\text{Ca}}_{{6}} \left[ {{\text{Al}}\left( {{\text{OH}}} \right)_{{6}} } \right]_{{2}} .{\text{24H}}_{{2}} {\text{O}}} \right\}^{{{6} + }} $$
(7)
$$ \left\{ {{\text{Ca}}_{6} \left[ {{\text{Al}}\left( {{\text{OH}}} \right)_{6} } \right]_{2} .{\text{24H}}_{2} {\text{O}}} \right\}^{{6 + }} + {\text{ 3SO}}_{4}^{{2 - }} + {\text{ 2H}}_{2} {\text{O }} \to {\text{ Ca}}_{6} {\text{Al}}_{2} \left( {{\text{SO}}_{4} } \right)_{3} \left( {{\text{OH}}} \right)_{{12}} .{\text{26H}}_{2} {\text{O}}\;\left( {{\text{Nevatalo 2}}014;{\text{ Fang 2}}018} \right) $$
(8)

The aging time required for this step depends on the final level of sulfate removal required and the amount of reagent added (Reinsel 1999).

pH Reduction/carbonation

This step involves the carbonation and neutralization of treated water by blowing carbon dioxide and subsequently precipitate calcium carbonate (Maree et al. 1992). The carbonation step is an environment-friendly method since the technology can utilize CO2 generated from industries (Vu et al. 2019). The suitable pH range in this step is between 7.5 and 9 (Ramsay 2001). At this step, by stabilizing the treated water with blowing CO2, the calcium concentration reduces by occurring CaCO3 crystallization according to the following reaction (Seidell, 1940; Bologo et al. 2012).

$$ {\text{CO}}_{{2}} \left( {\text{g}} \right) \, \leftrightarrow {\text{ CO}}_{{2}} \left( {{\text{aq}}} \right) $$
(9)
$$ {\text{CO}}_{{\text{2}}} \left( {{\text{aq}}} \right){\mkern 1mu} {\text{ + }}{\mkern 1mu} {\text{H}}_{{\text{2}}} {\text{O}}{\mkern 1mu} \leftrightarrow {\text{H}}_{{\text{2}}} {\text{CO}}_{{\text{3}}} $$
(10)
$$ {\text{H}}_{{\text{2}}} {\text{CO}}_{{\text{3}}} \leftrightarrow {\mkern 1mu} {\text{H}}^{{\text{ + }}} {\text{ + }}{\mkern 1mu} {\text{HCO}}_{{\text{3}}}^{{\text{ - }}} $$
(11)
$$ {\text{HCO}}_{{\text{3}}}^{{\text{ - }}} {\mkern 1mu} \leftrightarrow {\mkern 1mu} {\text{H}}^{{\text{ + }}} {\text{ + }}_{{}} {\text{CO}}_{{\text{3}}}^{{{\text{2 - }}}} {\text{(Mitchell}}{\mkern 1mu} {\text{2010)}} $$
(12)
$$ {\text{CO}}_{{3}}^{{{2} - }} + {\text{ Ca}}^{{{2} + }} \to {\text{ CaCO}}_{{3}} $$
(13)
$$ {\text{H}}^{ + } + {\text{ OH - }} \to {\text{ H}}_{2} {\text{O}}\;\left( {{\text{Seidell 194}}0;{\text{ Nevatalo 2}}014} \right) $$
(14)

According to the following reaction, the dissolution of calcium carbonate in acidic media causes the production of calcium bicarbonate (Nairn et al. 1991). In aqueous solutions, the equilibrium forming of calcium carbonate and calcium bicarbonate depends on the pH and ionic strength (Butler, 1988).

$$ {\text{CaCO}}_{3} + {\text{ H}}^{ + } \to {\text{ Ca}}^{{2 + }} + {\text{ HCO}}_{3}^{ - } \;({\text{Nairn 1991}}) $$
(15)

Recovery of aluminum hydroxide

The main advantage of this process is that adding aluminum hydroxide to the feed stream does not result in the addition of further ions in the treated water (Ramsay 2001). The proper pH range for ettringite decomposition is between 4 and 8.5 (Ramsay, 2001). The resulting aluminum hydroxide is recovered by multistage rinsing and precipitate-liquid separation and returned to step 3 (Maree et al. 1992).

The ettringite precipitation method for sulfate removal has the potential for commercial application in high sulfate concentrations wastewater treatment in the future (Fang et al. 2018). The operating parameters such as molar ratios of SO42−/ Ca2+ and SO42−/Al3+, and pH value have a significant effect on the sulfate removal process efficiency (Aygun et al. 2018). This study presents the feasibility of complete removing sulfate from the wastewater of Iran Chemical Industries Investment Company, a linear alkyl benzene manufacturer (LAB), using the ettringite precipitation method to reuse it in the industry and protect the environment.

Experimental

Materials

The wastewater effluent was obtained from the Iran Chemical Industries Investment Company (ICIIC) and used as feed flow for the laboratory-scale precipitation of dissolved sulfate by the ettringite precipitation method. H2SO4 (98% purity), Al(OH)3, EDTA, and Ca(OH)2 were purchased from Merck and used directly without any purification. Al(OH)3 after heating aluminum hydroxide for 10 h at 350 °C was used as aluminum source. Ca(OH)2 was used as pH adjustment and calcium source. CO2 gas, which was used as a pH adjustment for treated water after sulfate removal, supplied from the off-gas flow of the hydrogen production unit of Iran Chemical Industries Investment Company. The chemical composition of off-gas flow is given in Table 1.

Table 1 Chemical composition of the ICIIC off-gas flow
Full size table

Analytical methods

All the experiments were conducted at an ambient temperature of about 25 °C. The pH values of the solution were measured by 691 Metrohm pH meter with a glass electrode. Before each measurement, the pH meter was calibrated by different buffer solutions for an accurate pH value.

The concentrations of sulfate ions in solution were determined by comparing the colloidal suspension of barium sulfate, precipitated by adding barium chloride powder as a reagent, with the standard solution using HACH DR 5000™ UV–Vis laboratory spectrophotometer. The concentration of calcium and magnesium ions was determined by complexometric ethylenediaminetetraacetate (EDTA) titration method. The concentration of Al3+ in treated water was determined using the Optima 7300 V ICP-OES Spectrometers.

Methods

The flow diagram of the ettringite precipitation process is illustrated in Fig. 1. The process has five main steps followed by a pH adjustment step. Ettringite precipitation was performed in a glass beaker at room temperature. The properties of industrial wastewater and final treated water are shown in Table 2. The procedure of this experiment was as follows:

Fig. 1
figure 1

Flow diagram of the sulfate removal by ettringite precipitation process

Full size image
Table 2 The properties of industrial wastewater and final treated water
Full size table

Step 1: First, 1000 ml of industrial wastewater was poured in a glass beaker. Then, according to the sulfate concentration in the wastewater, a certain amount of calcium hydroxide (the molar ratio of SO42− to Ca2+ was 1:2) was added into the sample to increase the pH value to the 12–12.5 range and increase the concentration of calcium ions required for the ettringite precipitation step. The slurry solution was stirred with a 50 mm magnetic bar stirrer at a speed of 240 rpm for 30 min and then allowed to settle, and the supernatant was separated. In this step, magnesium ions in the presence of hydroxide ions were precipitated as insoluble magnesium hydroxides, and therefore, the concentration of magnesium ions was reduced. The supernatant was used as the ettringite precipitation step feed (Step 2). The supernatant was analyzed for SO42−, Ca2+, and Mg2+ concentrations.

Step 2: Ettringite precipitation at a pH rang of about 12–12.5 removed sulfate ions. At this step for completion of the sulfate precipitation as ettringite, it is necessary that a certain amount of aluminum hydroxide (the molar ratio of SO42− to Al3+ was 1:9) was added to the solution as a source of aluminum and stirred about 61 h by a stainless-steel stirrer equipped with a three-blade propeller at a constant speed of 240 rpm at ambient temperature. The ettringite precipitate was allowed to settle, and the Sulfate-free supernatant was separated. The supernatant was analyzed for SO42− and Ca2+ concentrations. The ettringite precipitated was used as the feed of the aluminum hydroxide recovery step (Step 3). The supernatant was used as the carbonation step feed (Step 5).

Step 3: At this step, sulfuric acid was added dropwise to the ettringite precipitate until the pH decreased to the range of 4–5 to recover aluminum hydroxide. The recovered aluminum hydroxide was first rinsed thoroughly with deionized water to reduce sulfate concentration. The precipitate in the slurry solution was allowed to settle, and then, the supernatant was separated. The supernatant was analyzed for SO42− and Ca2+ concentrations. The supernatant was used as the gypsum crystallization step feed (Step 4) to reduce its sulfate concentration. The recovered aluminum hydroxide was reused the ettringite precipitation step after several rinsing steps.

Step 4: The cycle of rinsing of recovered aluminum hydroxide and gypsum crystallization of sulfates in the rinsing water is essential to reduce sulfate and calcium concentration. In the crystallization step, the sulfates in the rinsing water were precipitated as gypsum in the presence of powdered gypsum (8 g) as seed and calcium hydroxide as calcium source (the molar ratio of SO42− to Ca2+ is 1:1) at ambient temperature. The sulfate concentration was reduced to the range of 1200–1300 mg/l after thirteen cycles. The gypsum was allowed to settle, and then, the supernatant was separated. The supernatant was analyzed for SO42− and Ca2+ concentrations after each rising and crystallization step.

Step 5: Sulfate-free supernatant separated from the ettringite precipitate was carbonized by blowing carbon dioxide gas via porous bubbler. At this step, by precipitating calcium in the form of calcium carbonate, the pH value decreased to the range of 7–8. The calcium carbonate was allowed to settle, and the supernatant was separated. The supernatant was analyzed for SO42− and Ca2+ concentrations. The supernatant from this step is sulfate-free treated water.

Results and discussion

The results of Table 2 indicate the complete sulfate removal by the ettringite precipitation process. Concentration of sulfate, magnesium, and calcium ions in the wastewater were reduced to zero, 15.81, and less than 20.04 mg/l, respectively. Aluminum ions were recovered as aluminum hydroxide by ettringite decomposition using sulfuric acid and then consecutive rinsing.

As shown in Table 2, the concentration of aluminum remaining in the treated water is 0.041 mg/l, which is much lower than the maximum level for aluminum concentration in drinking water, which is 0.2 mg/l. It can be concluded that the aluminum hydroxide added to the alkaline wastewater is significantly consumed in the formation of ettringite precipitation to remove sulfate.

As shown in Table 3, in the first step, by adding calcium hydroxide with a ratio of SO42− /Ca2+ 1: 2 to the industrial wastewater effluent, the pH value increased from 9.92 to 12.30 and the concentration of magnesium ions was reduced from 63.96 to 15.81 mg/l with an efficiency of 75.28%. By adding calcium hydroxide to the wastewater, the hydroxide ions required to react with the magnesium ions to form insoluble magnesium hydroxides are provided, thereby reducing the concentration of magnesium ions.

Table 3 Results of first step of the ettringite precipitation process by adding calcium hydroxide
Full size table

The concentration of calcium ions increased from 400.8 to 1482.96 mg/l by adding calcium hydroxide. But as expected, due to the high solubility of calcium sulfate, the sulfate concentration remained constant. Calcium hydroxide was added in this step to provide the condition for the precipitation of ettringite, which includes an alkaline medium and high calcium concentration.

In the second step, aluminum hydroxide with a ratio of SO42−: Al3+ 1: 9 was added to the slurry solution to form the ettringite precipitate. As shown in Table 4, one of the important factors in this step is the aging time. The sulfate concentration reduces with the gradual formation of ettringite precipitate over time. The ettringite precipitation was completed using fresh aluminum hydroxide after 61 h, and its supernatant has become sulfate-free.

Table 4 Results of second step of the ettringite precipitation process using fresh aluminum hydroxide
Full size table

The decomposition of ettringite precipitate was performed in an acidic medium. In this step, the pH value was reduced from 11.97 to 4.57 by adding sulfuric acid. The supernatant was transferred to the crystallization step to reduce the sulfate concentration and then to the rinsing step of the recovered precipitate. The recovered aluminum hydroxide after the consecutive rinsing steps was transferred to the ettringite precipitation step.

As shown in Table 5, thirteen consecutive rinsing-crystallization cycles were done to rinse the recovered aluminum hydroxide and to reduce the sulfate concentration of circulating rinsing water to about 1200–1300 mg/l. The cycle time of each rinsing and crystallization step was about 60 and 150 min, respectively. The presence of calcium and sulfate ions is essential for the precipitation of calcium sulfate in the crystallization step. Sulfate and calcium ions in crystallization feed were supplied from the recovered aluminum hydroxide precipitate rinsing step and by addition of calcium hydroxide, respectively. Also, to accelerate the crystallization step of calcium sulfate, gypsum powder as seeds was added to the crystallization solution. If the sulfate concentration did not reduce after the crystallization step, calcium hydroxide was added to increase the calcium concentration and the pH value for gypsum precipitation.

Table 5 Results of consecutive rinsing-crystallization cycles
Full size table

As shown in Table 6, in the fifth step, the pH value was reduced from 11.97 to 7.27 by blowing CO2 to the sulfate-free supernatant isolated from the ettringite precipitation step. The calcium concentration in sulfate-free treated water was reduced from 280.56 to less than 20.04 mg/l by precipitation of calcium carbonate with an efficiency of more than 92.86%.

Table 6 Results of fifth step of the ettringite precipitation process using CO2
Full size table

In this study, the ettringite precipitation process was also performed using recovered aluminum hydroxide after completing rinsing cycles with both sulfate and calcium removal efficiencies of 100%. As shown in Table 7, the ettringite precipitation step was completed using recovered aluminum hydroxide after 46 h, and the separated supernatant has become both sulfate- and calcium-free. The aging time required to complete ettringite precipitation using recovered aluminum hydroxide is much shorter than fresh aluminum hydroxide, which could be due to the higher dissolution rate of recovered aluminum hydroxide compared to fresh aluminum hydroxide. Its higher rate of dissolution enhances the rate of Al(OH)4‾ and [Al(OH)6]3‾ formation as the constituting species of ettringite.

Table 7 Results of steps first, second, and fifth of the ettringite precipitation process with recovered aluminum hydroxide
Full size table

Conclusions

The sulfate removal by ettringite precipitation with Al(OH)3 recovery step is highly feasible and cost-effective in the industrial application of sulfate content wastewater treatment. In this study, the removal of sulfate ions from the wastewater of Iran Chemical Industries Investment Company was evaluated. The sulfate was successfully removed from industries wastewater using Ca(OH)2 and Al(OH)3 through ettringite precipitation process, a calcium aluminum sulfate (Ca6Al2(SO4)3(OH)12.26H2O), under an alkaline medium with an efficiency of 100%. The results obtained in this study can be summarized as follows:

  • In the ettringite precipitation process, sulfate ions in the wastewater are precipitated as calcium sulfate and calcium carbonate using calcium hydroxide and aluminum hydroxide.

  • The metal removal step is conducted to oxide and precipitate metals as metal hydroxides by adding calcium hydroxide. In this study, magnesium ions preferentially converted to magnesium hydroxide under alkaline medium with an efficiency of 75.28%.

  • The ettringite decomposition for recovery of Al(OH)3 was performed under acidic conditions.

  • Sulfate ions in the resulting acidic supernatant of the ettringite decomposition step and the rinsing steps of the aluminum hydroxide recovery were crystallized in the presence of gypsum seeds under alkaline medium.

  • PH value, SO42−: Ca2+ ratio, SO42−: Al3+ ratio, and aging time are important parameters in sulfate removal by the ettringite precipitation process.

    The optimal reaction conditions were determined as follows: Reaction temperature (ambient temperature), pH range of ettringite precipitation (12-12.50), stirrer speed (245 rpm), and the aging time of ettringite precipitation step by fresh aluminum hydroxide (61 h) and by recovered aluminum hydroxide (61 h).

  • Based on the equilibrium of ettringite formation, the stoichiometric ratio of sulfate to calcium hydroxide to aluminum hydroxide (SO42−: Ca2+: Al3+) is 3:6:2. Considering the high sulfate removal efficiency, in this process, an optimized molar ratio of SO42−: Ca2+: Al3+ was obtained and constant at 1:2:9.

  • The precipitation of ettringite is pH-dependent. The optimum pH range is 12 to 12.50, which was kept by adding calcium hydroxide. The role of calcium hydroxide is both as a pH adjuster and a source of calcium ions for ettringite precipitation.

  • The results show that the aging time has a significant effect on the sulfate removal process. Sulfate removal efficiency increased by increasing aging time because the reaction between Al3+, Ca2+, and SO42− ions in an alkaline medium to precipitate ettringite is not so fast.

  • In the carbonation step, blowing CO2 as carbonating agent into the high pH treated water reduced the pH to neutral and removed calcium as CaCO3 with an efficiency of 92.86%.

  • The residual aluminum concentration in treated water is 0.041 mg/l, which is much lower than the maximum level for aluminum concentration in drinking water, which is 0.2 mg/l.