2018 Volume 59 Issue 2 Pages 297-302
This study explored the feasibility of removing chloride ions from an aqueous solution containing a high chloride concentration through the chemical precipitation of insoluble Friedel's salt, which was triggered by the addition of calcium and aluminum compounds. Calcium hydroxide (Ca(OH)2) and sodium aluminate (NaAlO2) were used as the reagents. Key factors for the removal of chloride ions, including the dosage of the reagents, reaction temperature, reaction time, initial pH, and initial chlorine concentration, were investigated.
The results indicated that an optimal chloride removal efficiency of 84.0% was obtained when the molar ratio of Ca:Al:Cl was 10:4:1. The main crystal phases in the formed precipitates, confirmed through X-ray diffraction analysis, were Friedel's salt (Ca2Al(OH)6Cl·2H2O), portlandite (Ca(OH)2), and katoite (Ca3(Al(OH)6)2). In addition, the chloride removal efficiency increased when the reaction temperature, initial pH, and initial chlorine concentration were higher. Chlorine removal could be further improved by conducting a second-stage treatment, regardless of the residual concentrations of chloride, calcium, and aluminum ions after the first-stage treatment.
Wastewaters generated from some industrial processes, such as tanning, canning, pickling, and bottom ash washing, often contain high concentrations of chloride ions1–5). Wastewater containing a high chloride concentration corrodes pipelines and exerts harmful effects on agricultural irrigation and fisheries. Several methods to remove chloride ions from aqueous solutions through either adsorption or ion exchange using a synthetic adsorbent or resin have been researched3,4,6–11). However, these methods are not suitable for treating wastewater containing high chloride concentrations because desorption and regeneration of the adsorbent and resin is frequently required. Only a few studies have focused on the treatment of wastewater containing high chloride concentrations.
Friedel's salt, an insoluble chlorine salt first described by Friedel in 1897, is produced from the reaction of calcium oxide (CaO) and aluminum chloride (AlCl3)12). The chemical formula of Friedel's salt is 3CaO·Al2O3·CaCl2·10H2O or Ca4Al2(OH)12Cl2(H2O)4. Its crystal structure is a type of layered double hydroxide where chloride ions enter the middle of the layered structure of [Ca4Al2(OH)12]2+ to balance the electrical charge. Friedel's salt is also an anion-exchange mineral that has a high adhesion for anion ions and is used as an adsorbent in the field of wastewater treatment13–15). Based on the formation and characteristics of Friedel's salt, we hypothesized that chloride ions in aqueous solutions can react with added calcium and aluminum ions to form the insoluble Friedel's salt, thus enabling removal.
This study explored the feasibility of removing chloride ions from aqueous solutions containing high chloride concentrations through the chemical precipitation of insoluble Friedel's salt, triggered by the addition of calcium (Ca) and aluminum (Al) compounds. Calcium hydroxides (Ca(OH)2) and sodium aluminate (NaAlO2) were used as reagents. Key factors for the removal of chloride ions, including the dosage of reagents, reaction temperature, reaction time, initial pH, and initial chlorine concentration, were investigated. The results were confirmed through an X-ray diffraction (XRD) analysis of the produced precipitates.
An aqueous solution containing a high chloride concentration was used in all experiments. The solution was prepared using calcium chloride (CaCl2) (Ferak Berlin GmbH, Germany). Unless stated otherwise, the chloride ion concentration was set to 2,600 mg/L, which reflected the concentration found in a bottom ash washing wastewater. Calcium hydroxide (Ca(OH)2) (Scharlab, S.L., Spain) was used as the Ca reagent and sodium aluminate (NaAlO2) (Taiwan Organic Chemical, Co. Ltd., Taiwan) was used as the Al reagent. All chemicals were reagent grade. In this study, the dosages of Ca(OH)2 and NaAlO2 were presented as the molar ratio of Ca:Al:Cl. The actual addition amounts of Ca(OH)2 and NaAlO2 were calculated on the basis of the chloride ion concentration, whereas the amount of Ca ions already in the aqueous solution was deducted. In addition, Ca(OH)2 and NaAlO2 were fed as powders.
All of the experiments were conducted as batch experiments. For each, 600 mL of a chlorine solution were poured into a 1,000 mL beaker equipped with a stainless magnetic stirrer. After the addition of Ca(OH)2 and NaAlO2, the solution was mixed at 150 rpm for 1 h. Solid–liquid separation was then performed using vacuum filtration. Filters with a diameter of 55 mm and a pore size of 1 μm were used. The concentrations of residual Ca and Al ions in the liquid fraction were analyzed through inductively coupled plasma-atomic emission spectrometry (ICP-AES) (SPS 7800, Seiko Instruments Inc.). Next, the residual chloride ion concentration in the liquid fraction was determined using Mohr's method, which is a precipitation titration method, where chloride ions in aqueous solutions are titrated with silver nitrate16,17). Subsequently, the chloride removal efficiency was calculated on the basis of the difference between the initial and residual chloride ion concentrations. Finally, the solid fraction (precipitate) was dried and the crystal phases in the precipitate were analyzed using an XRD (DMX-2200, Rigaku).
The chemical reaction of the formation of insoluble Friedel's salt in this study, triggered by the addition of Ca(OH)2 and NaAlO2 to an aqueous solution containing a high chlorine concentration, was as follows:
| \[ \begin{split} & {\rm CaCl_2} + {\rm 3Ca(OH)_2} + {\rm 2NaAlO_2} + {\rm 8H_2 O} \\ & \quad \to {\rm 2Ca_2} {\rm Al(OH)_6 Cl} \cdot {\rm 2H_2 O} + {\rm 2NaOH} \end{split} \] | (1) |
According to the chemical formula of Friedel's salt, the theoretical molar ratio of Ca:Al:Cl is 2:1:1. In this study, the dosages of Ca and Al compounds were set at 1 to 5 times the theoretical molar ratio (i.e., Ca/Cl = 2, 4, 6, 8, 10 and Al/Cl = 1, 2, 3, 4, 5). The effect of the various dosages of Ca(OH)2 and NaAlO2 on the removal of chloride ions is illustrated in Figs. 1 and 2. Specifically, in Fig. 1, the Al/Cl molar ratios are presented on the horizontal axis to demonstrate the effect of the Ca(OH)2 dosage, whereas in Fig. 2, the Ca/Cl molar ratios are presented on the horizontal axis to demonstrate the effect of the NaAlO2 dosage.
Effects of the dosages of Ca(OH)2 and NaAlO2 on the removal of chloride ions; the Al/Cl molar ratios are presented on the horizontal axis (experimental conditions: initial chloride concentration = 2,600 mg/L, reaction temperature = 25℃, reaction time = 1 h).
Effect of the dosages of Ca(OH)2 and NaAlO2 on the removal of chloride ions; the Ca/Cl molar ratios are presented on the horizontal axis (experimental conditions: initial chloride concentration = 2,600 mg/L, reaction temperature = 25℃, reaction time = 1 h).
According to Fig. 1, when the molar ratio of Al/Cl was 1, increasing the Ca(OH)2 dosage barely affected the removal of chloride ions. The chloride removal efficiency was approximately 54% at the most. It is thought that the Al ions in the solution fully reacted under this condition. No extra Al ions could react with chloride ions and the introduced Ca ions. Hence, the chloride removal efficiency could not be further improved, regardless of increases in the Ca(OH)2 dosage. This can also explain the results of chloride removal at the molar ratio of Al/Cl of 2 when Ca/Cl is higher than four. In these cases, the chloride removal efficiency was approximately 70% at the most. By contrast, when the molar ratio of Al/Cl ≥ 3, the effect of the Ca(OH)2 dosage on the removal of chloride ions became significant. Specifically, the chloride removal efficiency increased with the Ca(OH)2 dosage. The optimal chloride removal efficiency was 84.0% when the molar ratio of Ca:Al:Cl = 10:4:1.
As indicated in Fig. 2, when the molar ratio of Ca/Cl was 2, the chloride removal efficiency decreased with an increase in the dosage of NaAlO2. Conversely, when the molar ratio of Ca/Cl ≥ 4, the increase in the NaAlO2 dosage improved the chloride removal efficiency. However, when the molar ratios of Ca:Al:Cl were higher than 4:2:1, 6:2:1, or 8:4:1, the chloride removal efficiency could not be further improved and actually decreased. It is thought that when excess NaAlO2 was added, NaAlO2 and Ca(OH)2 react with each other as follows:
| \[ \begin{split} & {\rm 3Ca(OH)_2} + {\rm 2NaAlO_2} + {\rm 4H_2 O} \\ & \quad \to {\rm Ca_3} {\rm (Al(OH)_6)_2} + {\rm 2NaOH} \end{split} \] | (2) |
These results suggest that higher dosages of Ca(OH)2 and NaAlO2 than those theoretically predicted are required to improve the chloride removal efficiency. It is because the Ca(OH)2 introduced to the aqueous solution cannot completely dissolve and the NaAlO2 introduced can also react with the Ca(OH)2.
The XRD analysis results of the precipitates produced at molar ratios of Ca:Al:Cl = 4:1:1–4:5:1 are shown in Fig. 3. Notably, the main crystal phases in the precipitates detected by XRD were Friedel's salt (Ca2Al(OH)6Cl·2H2O), portlandite (Ca(OH)2), and katoite (Ca3(Al(OH)6)2). The XRD analysis confirmed that adding Ca and Al compounds to a chloride solution triggered a reaction between chloride ions and Ca and Al ions to form the insoluble Friedel's salt and thus enable the removal of chlorine. In addition, the presence of portlandite, owing to the unreacted Ca(OH)2, confirms that the introduced Ca(OH)2 is not entirely contribute to the formation of Friedel's salt. Hence, higher dosages of Ca(OH)2 are required to improve the chloride removal efficiency. Meanwhile, the katoite is the reaction product of NaAlO2 and Ca(OH)2, which confirms that these compounds react with each other as indicated in eq. (2). Furthermore, higher dosages of NaAlO2 reduced the production of Friedel's salt. As indicated in Fig. 3, the intensity of the main characteristic peaks of the Friedel's salt are reduced when Ca:Al:Cl = 4:5:1. Therefore, the chloride removal efficiency is reduced.
XRD analysis of the precipitates that were produced through the addition of Ca(OH)2 and NaAlO2 with molar ratios of Ca:Al:Cl set at 4:1:1–4:5:1 (experimental conditions: initial chloride concentration = 2,600 mg/L, reaction temperature = 25℃, reaction time = 1 h).
The effect of the reaction temperature on the chloride removal efficiency is shown in Fig. 4. The temperature was variously set at 25℃, 45℃, and 60℃, and the molar ratio of Ca:Al:Cl was 4:2:1 and 10:3:1. The results showed that with an increase in reaction temperature, the chloride removal efficiency increased from 69.0% at 25℃ to 80.0% at 60℃ when Ca:Al:Cl was 4:2:1. Similarly, the chloride removal efficiency increased from 81.2% at 25℃ to 87.5% at 60℃ when Ca:Al:Cl was 10:3:1. These results suggest that the removal of chloride ions through the addition of Ca(OH)2 and NaAlO2, which forms Friedel's salt, can be improved by increasing the reaction temperature. Specifically, the solubility of Ca(OH)2 increases with the temperature, enhancing the formation of Friedel's salt.
Effect of the reaction temperature on the chloride removal efficiency (experimental conditions: Ca:Al:Cl = 4:2:1 and 10:3:1, initial chloride concentration = 2,600 mg/L, reaction time = 1 h).
Figure 5 shows the effect of the reaction time on the chloride removal efficiency. The molar ratio of Ca:Al:Cl was alternately set at 4:2:1 and 10:3:1. The results showed that at both ratios, the chloride removal efficiency increased rapidly within the first 20 min and reached equilibrium after 1 h. This indicates that chloride ion is rapidly removed through the formation of Friedel's salt.
Effect of the reaction time on the chloride removal efficiency (experimental conditions: Ca:Al:Cl = 4:2:1 and 10:3:1, initial chloride concentration = 2,600 mg/L, reaction temperature = 25℃).
The XRD analysis results of the precipitates produced at different reaction times at the molar ratios of Ca:Al:Cl = 4:2:1 are presented in Fig. 6. The crystal phases of Friedel's salt, portlandite, and katoite were detected in the precipitate produced at 20 min; this further confirms the quick synthesis reaction of Friedel's salt. In addition, the intensity of the characteristic peaks of portlandite and katoite decreased as the reaction time increased, signifying that the introduced Ca(OH)2 fully reacted and contributed to the formation of Friedel's salt during a longer reaction time. Eventually, the produced katoite, which is the reaction product of NaAlO2 and Ca(OH)2, dissolved again. The results imply that the formation of Friedel's salt as presented in eq. (1) is superior to that of the katoite presented in eq. (2); in short, a reversed eq. (2) reaction occurred.
XRD analysis of the precipitates that were produced at various reaction times (experimental conditions: Ca:Al:Cl = 4:2:1, initial chloride concentration = 2,600 mg/L, reaction temperature = 25℃).
The effect of the initial chloride concentration on the chloride removal efficiency is shown in Fig. 7. The initial concentration was set at 2,600 mg/L, 5,000 mg/L, and 10,000 mg/L, and the molar ratio of Ca:Al:Cl was 4:2:1 and 10:3:1. The results indicated that for both molar ratios, the chloride removal efficiency increased with the initial chloride concentration; a linear relationship was observed between the two. Specifically, when the initial chloride concentration increased from 2,600 to 5,000 mg/L and then to 10,000 mg/L, the chloride removal efficiency increased from 69.0% to 71.5% and then to 76.6%, when Ca:Al:Cl = 4:2:1, and from 81.4% to 84.5% to 90.0%, respectively, when Ca:Al:Cl = 10:3:1.
Effect of the initial chloride concentration on the chloride removal efficiency (experimental conditions: Ca:Al:Cl = 4:2:1 and 10:3:1, reaction temperature = 25℃, reaction time = 1 h).
The XRD analysis results of the precipitates produced from the various initial chloride concentration solutions at the molar ratio of Ca:Al:Cl = 10:3:1 are shown in Fig. 8. The main crystal phases in the precipitates were Friedel's salt, portlandite, and katoite. Notably, higher initial chloride concentrations corresponded with the higher-intensity characteristic peaks of Friedel's salt and the lower-intensity peaks of katoite. This finding suggests that at higher initial chloride concentrations, the reaction among Ca, Al, and Cl (which produces Friedel's salt) is superior to that between Ca and Al (which forms katoite). Hence, Friedel's salt became the major product and the chloride removal efficiency increased, similar to the effect of increased reaction time described in Section 3.3.
XRD analysis of the precipitates that were produced from each initial chloride concentration solution (experimental conditions: Ca:Al:Cl = 10:3:1, reaction temperature = 25℃, reaction time = 1 h).
The effect of initial pH on the chloride removal efficiency is depicted in Fig. 9. The results showed that for all tested molar ratios, the chloride removal efficiency increased with an increase in the initial pH. Specifically, for an initial pH ranging from pH 4 to pH 12, the chloride removal efficiency at the molar ratios of Ca:Al:Cl of 2:1:1, 4:2:1, and 8:4:1 increased from 25.4% to 80.1%, from 33.5% to 84.2%, and from 50.1% to 88.5%, respectively. In short, a higher chloride removal efficiency was obtained at a higher initial pH; this occurs because Friedel's salt is formed and exists stably above pH 1218). Therefore, a higher initial pH is beneficial for chloride removal. However, chloride removal from a solution with a low initial pH can be improved by increasing the dosages of Ca(OH)2 and NaAlO2, as demonstrated by a solution that had an initial pH of 4 where the chloride removal efficiency at a molar ratio of 8:4:1 was higher than that at a ratio of 2:1:1 and 4:2:1. Furthermore, the pH of the chloride solution increased to more than pH 12 after the addition of Ca(OH)2 and NaAlO2. It is because NaOH is also produced when Friedel's salt forms, as shown in eq. (1). This result suggests that Ca(OH)2 and NaAlO2 are suitable reagents for the formation of Friedel's salt and that an additional pH adjustment of the chloride solution to maintain the stability of the produced Friedel's salt is not required. Nevertheless, the solution after chloride removal needs to be adjusted to a neutral pH before being discharged to meet most of the effluent standards on pH values.
Effect of the initial pH on the chloride removal efficiency (experimental conditions: Ca:Al:Cl = 2:1:1, 4:2:1, and 8:4:1; reaction temperature = 25℃; reaction time = 1 h).
To further improve chloride removal, a second-stage treatment for residual chloride ions was conducted. The dosage of the molar ratio of Ca:Al:Cl was set at 10:3:1. The actual amounts of Ca(OH)2 and NaAlO2 added were calculated on the basis of the residual chloride ion concentration, whereas the amount of Ca and Al ions already in the aqueous solution were deducted. The experimental conditions were the same as those for the first-stage treatment (i.e., stirred at 150 rpm for 1 h at 25℃). The experimental results are summarized in Table 1. For comparison, two types of residual chloride solutions after the first-stage treatment were used. The first one was obtained at the molar ratio of Ca:Al:Cl of 10:2:1, with a chloride removal efficiency of 69.8% and a residual chloride ion concentration of 785.0 mg/L; the residual concentrations of Ca and Al were 457.5 mg/L and 31.5 mg/L, respectively. After the second-stage treatment, the residual chloride ion concentration in this solution decreased to 148.6 mg/L and a chlorine removal efficiency of 81.1% was achieved. The overall chlorine removal efficiency was increased to 94.3%. The residual concentrations of Ca and Al after the second-stage treatment were 138.1 mg/L and 14.3 mg/L, respectively. The second solution was created at the molar ratio of Ca:Al:Cl = 2:4:1, with a chloride removal efficiency of 35.2% and a residual chloride ion concentration of 1,686 mg/L; the residual concentrations of Ca and Al were 138.6 mg/L and 1,494 mg/L, respectively. After the second-stage treatment, the residual chloride ion concentration in this solution decreased to 356.7 mg/L and a chloride removal efficiency of 78.8% was achieved. The overall chlorine removal efficiency increased to 86.3%. The residual concentrations of Ca and Al after the second-stage treatment were 296.7 mg/L and 30.6 mg/L, respectively.
| Initial chloride concentration (2600 mg/L) | Ca:Al:Cl | ||
|---|---|---|---|
| 10:2:1 | 2:4:1 | ||
| First-stage treatment |
Residual chloride concentration (mg/L) | 785.0 | 1686.0 |
| Chloride removal efficiency (%) | 69.8 | 35.2 | |
| Second-stage treatment |
Residual chloride concentration (mg/L) | 148.6 | 356.7 |
| Chloride removal efficiency (%) | 81.1 | 78.8 | |
| Overall chloride removal efficiency (%) | 94.3 | 86.3 | |
(Experimental conditions: reaction temperature = 25℃; reaction time = 1 h)
The aforementioned results suggest that a second-stage treatment can further improve chloride removal efficiency, regardless of the residual concentrations of Cl, Ca, and Al ions after the first-stage treatment. In addition, the dosages of Ca and Al reagents substantially affect not only the chloride removal efficiency but also the residual concentrations of Ca and Al ions after chloride removal. Under optimal dosage conditions of Ca and Al reagents obtained in this study, a higher chlorine removal efficiency as well as lower residual concentrations of Ca and Al ions after treatment can be achieved.
3.7 Practical application to real high-chloride wastewaterThe method proposed in this study was applied to a bottom ash washing wastewater generated from a municipal solid waste incinerator in Taiwan. The concentrations of Cl, Ca, and Al ions as well as the pH of the wastewater are shown in Table 2. The wastewater contains chloride ions at 2,600 mg/L with a pH of 12.2. The experimental conditions were the same as those in the fundamental investigations, namely stirring at 150 rpm for 1 h at 25℃. Two molar ratio dosages, Ca:Al:Cl = 4:2:1 and 10:4:1, were tested. The actual amounts of Ca(OH)2 and NaAlO2 added were also calculated on the basis of the chloride ion concentration, and the amounts of Ca and Al ions already in the wastewater were deducted. The wastewater originally contains a high Ca ion concentration (1,077.4 mg/L), and the actual amount of Ca reagent added can be reduced. The results are shown in Table 3. In the case of Ca:Al:Cl = 4:2:1, the chloride ion concentration in this wastewater decreased to 858.0 mg/L, and a chloride removal efficiency of 67.0% was obtained. The residual concentrations of Ca and Al were 172.7 mg/L and 19.8 mg/L, respectively. In the case of Ca:Al:Cl = 10:4:1, the chloride ion concentration decreased to 390.4 mg/L, and a chloride removal efficiency of 85.0% was achieved. The residual concentrations of Ca and Al were 282.0 mg/L and 23.7 mg/L, respectively. The final pH of the wastewater of the two dosage cases both slightly increased to pH 12.7. The results were similar to those shown in the fundamental investigations, which confirmed the feasibility of this method.
| Cl | Ca | Al | pH | |
|---|---|---|---|---|
| 2600 mg/L | 1077.4 mg/L | 0.8 mg/L | 12.2 |
| Ca:Al:Cl | Residual chloride concentration (mg/L) |
Chloride removal efficiency (%) |
Final pH |
|---|---|---|---|
| 4:2:1 | 858.0 | 67.0 | 12.7 |
| 10:4:1 | 390.4 | 85.0 | 12.7 |
(Experimental conditions: reaction temperature = 25℃; reaction time = 1 h)
This study explored the feasibility of removing chloride ions from an aqueous solution containing a high chloride concentration through the chemical precipitation of Friedel's salt, which was triggered by adding Ca(OH)2 and NaAlO2. Specifically, the following results were obtained:
(1) Higher dosages of Ca(OH)2 and NaAlO2 than the theoretical molar ratio of Friedel's salt (i.e., Ca:Al:Cl = 2:1:1) were required for optimal chloride removal efficiency, because of the lower solubility of Ca(OH)2 as well as the reaction between Ca(OH)2 and NaAlO2. Overall, an optimal chloride removal efficiency (84.0%) was obtained when the molar ratio was Ca:Al:Cl = 10:4:1.
(2) The main mineralogical crystal phases in the formed precipitates were Friedel's salt, katoite, and portlandite, which were confirmed through XRD analysis.
(3) The chloride removal efficiency increased with the reaction temperature, initial pH, and initial chlorine concentration of the solution.
(4) The chloride removal efficiency can be further improved by conducting a second-stage treatment on residual chloride ions, regardless of the residual concentrations of Cl, Ca, and Al ions that remain after the first-stage treatment.
The method proposed in this study is effective for the treatment of an aqueous solution containing a high chloride concentration and can be integrated with other chloride treatment methods, such as adsorption and ion exchange, to further remove chloride, if stricter effluent standards need to be met.
This work was financially supported in part by the R.O.C. Ministry of Science and Technology under Grant No. MOST 105-2221-E-027-001 as well as the R.O.C. Ministry of Economic Affair, Water Resources Agency under Grant No. 206F11.
