Note
Effect of Si Ion Concentrations on the Phosphorous Ion Separation from the Solution Containing Ca, Si, and P Ions by Ion-exchange Membrane Electrodialysis
2023 Volume 63 Issue 11 Pages 1923-1926
Details
2023 Volume 63 Issue 11 Pages 1923-1926
Phosphorus is an essential raw material for industry and agriculture, and Japan imports all of it. The phosphorus in steelmaking slag is a promising secondary phosphorus resource comparable to the total amount of phosphorus imported by Japan. Several processes have been proposed to recover phosphorous from steelmaking slags. However, due to the high recovery cost and high environmental load in these processes, phosphorus recovery from steelmaking slag is not yet practiced on an industrial scale. We are developing an electrodialysis process for recovering phosphorus from steelmaking slag extract. Steelmaking slag extract contains not only P anions but also Si anions. This paper has investigated the influence of Si ions on phosphorus separation by electrodialysis. In the case of a solution containing more than 50 ppm of Si content, it was found that Si ions significantly inhibit the transport of phosphorus ions through the membrane.
It is well known that phosphorus is a critical element for agriculture and plays a vital role in industrial fields, such as electronics, automobiles, medicines, foods, plastic, etc., especially in the semiconductor and pharmaceutical industries. However, high-grade phosphate ore deposits are rapidly depleted, and the economically minable deposits may be practically exhausted soon. Japan depends entirely on imported phosphorus. Thus, alternative resources for phosphate ore must be found.1,2,3,4,5) Municipal sewage sludge is one of the expected major secondary sources of phosphorus. The phosphorus extraction from sewage has been extensively studied worldwide and implemented in several places, although still with economic problems or required local government subsidiary aid.6,7,8)
Meanwhile, phosphorus is one of the most hazardous elements for steel products, and much effort has been paid to remove phosphorus from molten steel into steelmaking slag.9,10) Through the iron and steelmaking processes, the phosphorus in the iron ore (0.02 mass%P) is enriched in the steelmaking slag (2 mass% P). Ohtake and Tsuneda investigated the overall domestic material flow of phosphorus in Japan in 2016.11) The amount of overall phosphorus in the steelmaking slag through the dephosphorization processes (114.0 kt-P/y) is almost three times larger than that of the imported phosphate ore (41.0 kt-P/y). Due to the progressing worldwide depletion of economically minable phosphate rock, the import of phosphate rock in Japan is expected to decrease shortly. To meet the decrease of the imported phosphate rock, phosphorous recovery from steelmaking slags will be one of the solutions.
Besides the quantity, steelmaking slag has several qualitative advantages, such as phosphorus content, stable and constant supply, and negligible contamination of toxic heavy metals such as lead, mercury, cadmium, etc., compared with other possible secondary sources of sewage sludge and industrial wastes.
The recovery of phosphorous from steelmaking slags has another attractive point. When phosphorus is removed from the steelmaking slag, much phosphorus-free steelmaking slag can be recharged into the blast furnace, and CaO and iron can be effectively recycled.5) This recycling is a tremendous benefit for the steel industry. Therefore, the phosphorus recovery from the steelmaking slag has attracted increasing attention, and many attempts have been carried out on the laboratory scale.
Two kinds of processes, (1) Pyrometallurgical and (2) Hydrometallurgical process, have been studied for phosphorus recovery from steelmaking slag.12) Pyrometallurgical processes such as carbothermic reduction of the slag at a high temperature have been studied to extract phosphorus in the steelmaking slag. The critical problem for the carbothermic reduction of phosphorous oxide in steelmaking slags is the spontaneous dissolution of the produced phosphorus gas into the reduced iron to form Fe–P alloy. The Fe–P alloy is difficult to use further as a phosphorus or iron resource.
Phosphorus recovery from sewage sludge by a hydrometallurgical process has been extensively studied. However, few studies were carried out on phosphorus recovery from steelmaking slag by the hydrometallurgical process. Du et al.13) successfully recovered the phosphorus in the steelmaking slag and produced crude phosphoric acid using hydro-leaching. However, their recovery consisted of many processing steps, requiring many chemicals and generating many waste liquids. Consequently, the recovery cost and environmental load are high. Thus, phosphorus recovery from steelmaking slag is not yet practiced on an industrial scale.
An electrodialysis using an ion-exchange membrane has attracted attention for recovering valuable metals in industrial wastewater.14) Electrodialysis is a membrane-based process involving the transport of ions through semipermeable membranes using an applied electric field. When an anion exchange membrane is used, only anions can permeate through the membrane. Suppose the straightforward phosphorus ion recovery process can be achieved by applying the electrodialysis process from the slag extract. In that case, the process can significantly reduce the number of processing steps, and the required chemicals and the associated waste liquid can also be reduced. Therefore, it will have many advantages over the conventional method.
Electrodialysis has already been put to practical use, such as producing salt from seawater and recovering metals from industrial liquid wastes.14) However, no research has been conducted to recover phosphorus from steelmaking slag by electrodialysis. In general, the behavior of mono and divalent cations such as Na and Ca in electrodialysis has been well studied; however, that of anions has not been studied much except for monovalent anions such as Cl− ion.14) The slag extract contains several complex anions containing P and Si such as H2PO4−, HPO42− and H3SiO4−. The Si-containing anions may influence the P-containing anions separation in the electrodialysis process. Our electrodialysis experiments were carried out in acidic conditions. The only dissolved Si-containing species within the pH range between 1 and 8 is the silicic acid molecule Si(OH)4. However, these monomers are gradually connected through Si–O–Si branches or the polymerization via Si(OH)4 molecules, eventually becoming amorphous silica colloids. In geothermal plants, the silica colloid sometimes deposits on the inner wall of the boiler, causing severe problems such as lowering heat conduction. In the electrodialysis process, the formed colloids may also cause clogging of the ion exchange membrane and deteriorate the separation process. Therefore, it is essential to know the behavior of Si ions in the electrodialysis process to recover phosphorous from steelmaking slag extract. This study investigated the behavior of phosphorus and silicon anions during electrodialysis as basic research to develop an electrodialysis process for efficiently recovering phosphorus from steelmaking slag.
The electrodialysis cell is made by cutting an acrylic block (70×70×80 cm), as shown schematically in Fig. 1. A cylindrical hole with a diameter of 50 mm and a height of 80 mm from the top surface was drilled in the acrylic block as a solution container, and another cylindrical hole with a diameter of 40 mm on the side was made to connect two cells. A circular groove with a width of 2 mm was formed on the side of one block to hold the O-ring. After setting the O-ring in this groove, an anion exchange membrane was sandwiched between the two blocks, and the two blocks were tightly fixed using bolts and nuts through the small cylindrical holes made in the four corners of the block that are not shown in Fig. 1.
The cell with an anodic electrode was called the anodic cell, and that with a cathodic electrode was named the cathodic cell. Titanium plates (L: 40 mm, W: 30 mm) were inserted into the solutions held in the cells as electrodes, and electricity was applied between them using a constant current device. The solution in the cell was stirred (150 rpm) using a magnetic stirrer to homogenize the solution concentration.
The extract obtained by the extraction of steelmaking generally contains cations of Ca, Al, Mg, Fe, and complex anions of Si and P ions. It is known that the cations can be easily separated from solutions by using cation membranes. Our study aims to investigate the effect of Si ions on the phosphorous ion separation behavior during the electrodialysis process. Thus, as a typical extract, a solution containing the anions of Si, P, and Ca as a cation was selected for simplicity, and the electrodialysis experiments were conducted by varying the Ca, Si, and P concentrations in the range of 2000–500 ppm, 500–50 ppm, and 1000–50 ppm, respectively.
The solution was prepared by dissolving an appropriate amount of H3PO4, Ca(OH)2, and NaSiO3·9H2O chemical-grade reagents (Wako Pure Chemical Industries, Ltd.) into deionized water. pH of the solution was adjusted by adding an appropriate small amount of NaOH or HCl solution. As mentioned, the Si colloid formation gradually occurs in a high Si concentration solution. As a result, the concentration of dissolved Si slowly decreases with time. Therefore, in most experiments, the prepared solution was used for electrodialysis experiments within 10 min after the sample solution preparation. When using a solution that has passed more than 10 min, the concentrations of Si and P in the solution were measured and used as the initial values.
All experiments were carried out at room temperature (around 300 K). The anion exchange membrane is immersed in a phosphoric acid solution for 3 hours before mounting on the cell to saturate the ion exchange membranes with phosphoric acid. A constant current was applied to the anodic and cathodic containers after introducing the solutions with a predetermined concentration in the cells. The simulated slag extract was held in the anodic cell, and the very diluted H3PO4 solution was in the cathodic cell to endorse good electrical conductivity. The applied current of about 0.5 mA was slightly less than the limit current confirmed in the preliminary experiment.
After a predetermined time, the current was stopped, and then about 1 ml of the solution in both cells was sampled from the solution. Each sampled solution was diluted with hydrochloric acid. The Ca, P, and Si concentrations were measured using an inductively coupled plasm, and the pH was measured using a pH meter. The pH at the start of the experiment was set to about 4. The pH changed slightly during the experiment due to the variation of phosphorus and Si concentrations, but it was not controlled. As a preliminary experiment, five commercially available anion-exchange membranes were used for electrodialysis of phosphorus-containing solutions to evaluate the transport efficiency and type AFX (Astom Corp.) showed the highest efficiency. Therefore, AFX was used for all experiments in this study. The specification of AFX is shown in Table 1.
| Electric resistance | Burst strength | Thickness | Recommended temperature | Recommended pH |
|---|---|---|---|---|
| 2.5 Ω·cm2 | >0.25 MPa | 0.17 mm | <40°C | 0–8 |
Preliminary experiments were carried out by varying the Ca concentration between 2000 ppm and 500 ppm, but no significant difference was observed in phosphorus transfer. Therefore, the Ca concentration was fixed at 600 ppm for most of the experiments. Figure 2 shows the results of electrodialysis performed for 6 hours, with the initial concentrations of phosphorus on the anode side and cathode side being 10 ppm and 1000 ppm, respectively. The anion-exchange membrane was immersed in a 1000 ppm phosphorus solution for 3 hours before starting the experiment to soak the membrane with phosphorus ions. In Fig. 2, results (A) without soaking and without current application, (B) using the membrane without soaking under the current application, and (C) using the membrane with soaking under the current application are also shown.
Electrodialysis is affected by various experimental conditions, such as the solution velocity gradient, concentration near the ion-exchange membrane, and the distance between electrodes. Since it is difficult to keep these conditions constant in our experiments, the experiment was repeated three times, and the average value of the concentration was used as the representative value. In the case of A, about 264 ppm of phosphorous decreased in the cathode cell, and 192 ppm of phosphorous increased after 6 hrs of electrodialysis. It is not mass-balanced because some phosphorous ions may remain in the membrane. By applying current, the amount of phosphorus transported is doubled compared with that without the current.
As a typical example, the change in phosphorus concentration with time in each cell due to electrodialysis when the initial phosphorus concentrations in the anode cell and the cathode cell were set to 1000 ppm and 10 ppm, respectively, is shown in Fig. 3.
Phosphorus concentration decreased by about 10% in 1 hour. The recovery efficiency seems to be low, but it can be easily enhanced. For example, in an industrial process for producing salt from seawater using ion exchange membranes, 2000 pairs of cation and anion exchange membranes with an area of 1 to 2 m2 are alternately stacked at 0.5 to 0.75 mm intervals, and seawater is flowing at high speed in this narrow space of 0.5–0.75 mm. Therefore, efficiency is greatly improved when applying a similar-scale industrial system.
Figure 4 shows the result of 6-hour electrodialysis with the initial phosphorus concentration of the solution in the anode chamber set at 10 ppm and the initial phosphorus and Si ion concentrations in the cathode cell at 905 pm and 276 ppm, respectively. Although the initial concentration of P and Si was set to 1000 ppm and 450 ppm when the solution was prepared, since the Si concentration was high, silica colloid formation proceeded immediately after the preparation of the solution, resulting in low values, respectively. The decrease in P concentration in the initial solution is due to the adsorption of phosphorus ions to the silica colloid under formation. The slight increase in Si content after 6 hours is possibly due to some analytical fluctuation.
The P transfer is significantly inhibited in the solution containing originally 450 ppm of Si compared to the case without Si addition. The mass balances of Si and P are not maintained, but this is because Si and P ions may remain in the exchange film.
Figure 5 shows the time change of phosphorus transfer when 50 ppm of Si is added to the cathode solution. The initial phosphorous contents in the cathode and anode cells are 1000 ppm and 10 ppm, respectively.
The phosphorous mass balance shown in Fig. 5 is quite unbalanced. This is possibly due to the silica colloidal formation, as already mentioned. In the 50 ppm Si-contained solution, the amount of P permeation was comparable to when the experiment was performed without Si ions, as shown in Fig. 3. The effect of Si concentration on phosphorus transfer is shown in Fig. 6. When the Si content is 100 ppm or more, phosphorus transfer is inhibited.
In order to investigate the influence of Si ions on the phosphorous transfer, the distribution of ion species in each solution containing Ca, Si, and P was evaluated as a function of pH using the solution analysis software PHREEQC.15) The results are shown in Fig. 7. The Ca and Si contents in the solution are fixed at 600 ppm and 250 ppm, respectively. The temperature of each solution is fixed at 25°C.
Even if the Si concentration increases, the relative distributions of Ca and P ion species, except Si ion, do not change so much. At more than 100 ppm of Si ion in the solution, however, amorphous silica appears, and the amount of amorphous silica increases with the increase of Si content. As shown in Fig. 6, the phosphorous transport is retarded when the solution contains more than 100 ppm Si ions. Amorphous SiO2 is not observed when the Si concentration in the cathode cell solution is 50 ppm, and the retardation did not occur.
The solubility (ppm) of amorphous silica in an aqueous solution16) is presented by
| (1) |
based on this Eq. (1), the solubility of amorphous silica at 298 K is about 117 ppm.
This value is slightly different from the PHREEQC results of 60 ppm, as shown in Fig. 7(a), due to the influence of Ca and P ions. For the solubility of amorphous silica in an aqueous solution, the solubility is known to be almost constant within the pH range between 1 and 8 when the Si concentration is under the saturated condition, as shown in Fig. 7. The only dissolved species in this range is the silicic acid molecule Si(OH)4. Once the silicic acid molecule, Si(OH)4 concentration reaches the saturated concentration, these monomers gradually connect through Si–O–Si branches, or the polymerization via Si(OH)4 molecules proceeds, and eventually, colloid formation occurs.17,18) This fundamental feature will not change even if other ions exist. Thus, it is considered that the main reason for the significant decrease in phosphorous transportation from the cathode to the anode in the case of high Si concentration is possibly the space blockage in the ion-exchange membrane due to the colloid formation of amorphous SiO2.
The colloid formation condition of amorphous silica in an aqueous solution is relatively well-studied.17,18,19,20) However, that in various acid solutions has not been investigated much since the colloid formation is a complex process and depends on several parameters such as the oversaturation of silicon, pH value, temperature, and the ionic strength of the solution. Thus, further work on the colloid formation kinetics is undoubtedly required to establish an effective separation of phosphorous from steelmaking slag extract.
As a concluding remark, this study clarified that the Si concentration in the steelmaking slag extract must be less than 50 ppm when electrodialysis is applied to recover phosphorus from the slag extract. In other words, phosphorus can be easily recovered by using electrodialysis if Si can be removed from the slag extract to 50 ppm or less in advance. The silica colloid formation is also one of the most common problems in the hydrometallurgical processing of silica-containing resources. Silica colloid formation represents a serious drawback in recovering metals from ores and process residues by hydrometallurgical methods because the colloid-containing solutions can no longer be filtered and reduce the leaching kinetics significantly.21,22,23) However, no decisive means to meet the practical SiO2 colloid removal problem have been established. Therefore, to recover phosphorus from the slag extract using electrodialysis, it is necessary to carry out research to prevent the formation of silica colloids and effectively remove silica colloids in parallel.
This work is partially supported by the “ISIJ research promotion grant (2021)”.
