P₂O₅ Solubility Behavior and Resource Utilization of P-Bearing Slag

Abstract

In order to recycle the phosphorus in P-bearing steelmaking slag and make it used as slag phosphate fertilizer, the citric acid solubility of P-bearing steelmaking slag was researched. The research results show that the citric acid solubility of P-bearing steelmaking slag is decreased with the increasing of P₂O₅ and Fe₂O₃ content for the CaO–SiO₂–Fe₂O₃–P₂O₅ slag system. Added CaF₂ into slag can easily form the fluorapatite (Ca₅(PO₄)₂F) that can’t be dissolved in 2% citric acid solution, which makes citric acid solubility of the slag decreased obviously. And added MgO, MnO or Na₂O into slag can prevent the precipitation of β-Ca₃(PO₄)₂ phase with low citric acid solubility, which makes the citric acid solubility of slag increased. While for the CaO–SiO₂–Fe₂O₃–P₂O₅–X (MgO, MnO or Na₂O) slag system, MgO and MnO in slag mainly enters into RO phase, which has no effect on the phosphorus existence form in slag and has little influence on citric acid solubility of the slag. The Na₂O in slag changes the phosphorus existence form in slag, but the generated Na₂Ca₄(PO₄)₂SiO₄ and Na₃PO₄ also have good citric acid solubility, which also has little effect on citric acid solubility. Al₂O₃ and TiO₂ modification only increases the phosphorus content in phosphorus-rich phase, does not change the phosphorus existence form in slag, which makes the citric acid solubility increased slightly. While SiO₂ modification prevents the precipitation of β-Ca₃(PO₄)₂ phase with low citric acid solubility, which makes citric acid solubility of the slag increased significantly. The modified slag without fluoride makes the citric acid solubility increased to 95% and meets the requirement of producing slag phosphate fertilizer.

1. Introduction

With the obviously increasing of high quality ore price and the development of new generation steel-making duplex process in China, high phosphorus content ore utilization in melting is inevitable. Duplex process is using two converters for production, one converter is using for dephosphorization, another one receiving the low phosphorus hot metal from dephosphorization converter is using for decarbonization High phosphorus iron ore resource is rich in China, proved reserves is up to 7.45 billion tons, accounted for above 10% of national iron ore resource reserves. Currently, due to high phosphorus content, the ore can’t be fully used. If blast furnace hot metal with phosphorus content greater than 0.3% was used for dephosphorization by duplex process, which can obtain high phosphorus slag with P₂O₅ content greater than 10%, and also can obtain high quality slag phosphate fertilizer with further processing, settle the environment pollution and create economic benefits. M. Ishikawa in Sumitomo Metal Industries produced the dephosphorization slag by dephosphorization furnace, and then recycled the slag by slag regeneration processing, which can get slag with P₂O₅ content above 10% and the slag can be directly used as slag phosphate fertilizer.^1,2) JFE Steel used low silicon hot metal with 0.1–0.2% phosphorus, blowed CaO and oxygen source for dephosphorization, and produced the high phosphorus slag by dephosphorization furnace in duplex process can be directly used as phosphate fertilizer raw materials.³⁾

Using of slag phosphate fertilizer has hundreds years of history, in the country with rich phosphorus containing iron ore such as France, Germany etc., slag phosphate fertilizer always has big share,⁴⁾ accounts for 13–16% of total phosphate fertilizer.^1,5) However, blast furnace slag and steel slag used as agricultural fertilizer and soil improvement is totally 230000 tons in China 2010, accounts for 0.5% of total blast furnace slag and steel slag using amount,⁶⁾ so the utilization of steel slag for fertilizer is very low.

Up to now, recycling P-bearing slag as slag phosphate fertilizer has attracted widespread attention, in this regard, some scholars majorly researched on phosphorus existence in slag^7,8) and phosphorus enrichment separation behavior of P-bearing slag,^{9,10,11,12,13)} while there are few researches on citric acid solubility (for short P₂O₅ solubility) of P-bearing slag.^14,15,16) Slag phosphate fertilizer can not be dissolved in water, but can be dissolved in 2% citric acid solution. So the slag phosphate fertilizer belongs to citrate-soluble phosphatic fertilizer, and its fertilizer efficiency is determined by P₂O₅ content and P₂O₅ solubility. P₂O₅ solubility means the mass fraction ratio between P₂O₅ content that can be dissolved in 2% citric acid solution and total P₂O₅ in slag, and the P₂O₅ solubility is the important indicator to evaluate if the dephosphorization slag can be used as slag phosphate fertilizer or not. The using of slag phosphate fertilizer provides necessary support for the using of high phosphorus ore, while also saves the mining amount of phosphorus ore resources, and instructs the resource utilization of P-bearing slag.

This paper researches the influence of different basic slag composition (basicity, Fe₂O₃ and P₂O₅), surplus calcium alkalinity and slag composition modification on the P₂O₅ solubility of P-bearing slag, the effect of melting modification by SiO₂, Al₂O₃ and TiO₂ addition and slag without fluoride on the slag P₂O₅ solubility was discussed emphatically, which provide the necessary research basis for the using of slag phosphate fertilizer.

2. Experimental Methods

2.1. Thermal Experimental of P-bearing Slag

Reagent-grade CaO, SiO₂, Fe₂O₃, P₂O₅, MgO, MnCO₃, CaF₂ and Na₂CO₃ were used to produce slag in this experiment. Then, the reagents were mixed in various ratios in order to produce the CaO–SiO₂–Fe₂O₃–P₂O₅ slag system. For investigating the effect of other elements on P₂O₅ solubility, Na₂O, CaF₂, MgO and MnO were added in some cases. The mixing conditions are summarized in Table 1. Figure 1 also shows the composition range of the used slag on the CaO–SiO₂–Fe₂O₃ phase diagram that calculated by FactSage 6.2 for samples A, B, and C respectively. As shown in Fig. 1, all experimental slag are in the dicalcium silicate (C₂S) primary zone, C₂S was precipitated firstly during the cooling process, and so nCa₂SiO₄–Ca₃(PO₄)₂ (for short nC₂S–C₃P) solid solution (mainly Ca₁₅(PO₄)₂(SiO₄)₆ and Ca₅(PO₄)₂SiO₄) was easily formed. Most of the iron oxide would be in the form of FeO in steelmaking condition, but Fe₂O₃ was used as iron oxide in this experiment, the reasons are as follow: 1) Ito et al.¹⁷⁾ has shown that the distribution behavior of P₂O₅ between solid solution and liquid phases was not different, when the iron oxide was changed from FeO to Fe₂O₃. 2) Shimauchi et al.¹³⁾ found that FeO was solved into the solid solution but the content of Fe₂O₃ in solid solution was negligible small. 3) From the preliminary experiments, it was found that phosphorus enrichment behavior between solid solution and liquid phases was not affected, when the iron oxide was changed from FeO to Fe₂O₃.

Table 1. Composition of the synthetic slag samples (mass fraction, %).

No.	CaO	SiO2	Fe2O3	P2O5	MgO	MnO	CaF2	Na2O	P2O5 solubility/%
A-1	45.71	18.29	30	6					92.31
A-2	42.86	17.14	30	10					84.70
A-3	37.14	14.86	30	18					58.47
B-1	49.33	24.67	20	6					96.90
B-2	46.67	23.33	20	10					93.54
B-3	41.33	20.67	20	18					76.05
C-1	50.40	33.60	10	6					98.39
C-2	48.00	32.00	10	10					95.51
C-3	43.20	28.80	10	18					82.86
A-2(5%MgO)	39.29	15.71	30	10	5				79.59
A-2(10%MgO)	35.71	14.29	30	10	10				92.81
A-2(5%MnO)	39.29	15.71	30	10		5			98.28
A-2(10%MnO)	35.71	14.29	30	10		10			96.89
A-2(3%Na2O)	40.71	16.29	30	10				3	96.99
A-2(6%Na2O)	38.57	15.43	30	10				6	96.13
A-2(3%CaF2)	40.71	16.29	30	10			3		17.18
A-2(6%CaF2)	38.57	15.43	30	10			6		11.25

Fig. 1.

Observed slag composition in the ternary phase diagram of the CaO–SiO₂–Fe₂O₃ system.

Some scholars^18,19,20,21) modified the slag by adding some oxides of SiO₂, Al₂O₃ and TiO₂ in the P-bearing slag, and used melting-cooling method to research the phosphorus enrichment and separation, and phosphorus enrichment and separation in the slag was realized, and the separated phosphorus-rich phase was used as phosphate fertilizer. But the research on P₂O₅ solubility (P₂O₅ solubility is an important indicator for the evaluation of slag used for fertilizer) of P-bearing slag after melting modification is less.

Based on the converter slag of one steel plant, reagent grade P₂O₅ was added into original slag and the P₂O₅ contents were all adjusted to 10%, reagent grade SiO₂ (slag basicity R is 3.0, 2.0 and 1.0, and R=(mass%CaO)/(mass%SiO₂), the same below), Al₂O₃ (Al₂O₃ content is 10% and 15% respectively) and TiO₂ (TiO₂ content is 10%) were added to research the effect of modification of SiO₂, Al₂O₃ and TiO₂ on P₂O₅ solubility in slag, the composition of modified slag is listed in Table 2.

Table 2. Chemical composition of the original slag and modified slag (mass fraction, %).

Slag sample	CaO	SiO2	Fe2O3	P2O5	Al2O3	MgO	TiO2	CaF2	MnO	P2O5 solubility/%
Original slag 1	46.72	11.68	19.47	10	1.07	5.46	0.97	0.37	4.26	58.35
No. 1	44.78	14.93	18.66	10	1.03	5.23	0.93	0.35	4.08	60.50
No. 2	41.35	20.68	17.23	10	0.95	4.83	0.86	0.33	3.77	72.43
No. 3	33.63	33.63	14.01	10	0.77	3.93	0.70	0.27	3.07	98.84
Original slag 2	45.72	15.24	17.82	10	1.09	5.00	0.89	0.34	3.90	74.96
No. 4	41.13	13.71	16.03	10	10	4.50	0.80	0.31	3.51	78.91
No. 5	38.46	12.82	15.12	10	15	4.24	0.76	0.29	3.31	81.09
No. 6	41.12	13.71	16.00	10	0.88	4.49	10	0.31	3.50	83.07

The mixed slag (200 g) was placed in an crucible 60 mm dia×100 mm, placed inside a graphite crucible and heated in a MoSi₂ electric resistance furnace up to 1773 K, according to observing, the slag had been melted at 1773 K during the experimental process. In order to make slag fully melted, and the slag was maintained the temperature for 30 minutes. Then the slag was cooled to 1623 K at a cooling rate of 3 K/min, kept at this temperature for 1 hour in order to fully promote the precipitation of nC₂S–C₃P solid solution, then cooled to 1423 K at a cooling rate of 3 K/min, the furnace was then closed and the slag sample was cooled within it (see Fig. 2). After each experiment the slag was ground (less than 300 mesh (48 μm)), mineralogical phases were determined by XRD analysis. Diffraction patterns were measured in a 2θ range of 10–90° using copper K_α radiation of 40 kV and 30 mA, and the scan speed was 5°/min.

Fig. 2.

Experimental conditions for precipitation of C₂S–C₃P solid solution.

2.2. P₂O₅ Solubility Experimental of P-bearing Slag

The quinoline phosphomolybdate gravimetric method provided by national standard GB20412-2006 is used to measure the P₂O₅ solubility of P-bearing slag. 1.0000 g P-bearing slag ground to less than 300 meshes (48 μm) has been prepared, and then it was put in a dry 250 mL volumetric flask. Accurately added 150 mL, 28–30°C 2% citric acid solution into the volumetric flask, kept the temperature in the range of 28–30°C, and put the volumetric flask on oscillator to shake for one hour. After adding water in volumetric flask to the tick mark, the solution was blended and filtered. Certain sample solution was poured into a 500 mL beaker, and 10 mL 1:1 nitric acid solution was added, then water was added into the beaker to dilute to 100 mL. Sample solution was heated to boil. Then 35 mL quimociac was added into the beaker, and reacted with the P₂O₅ in sample solution and precipitated as yellow quinoline phosphomolybdate. And then the beaker was covered with watch glass, heated it again to micro-boil to make the sediment layered, and the beaker was taken out to cool to ambient temperature. The glass filter which has been put in a (180±2)°C drying oven to dry to constant weight was used to filter, then put the precipitate and glass filter in the (180±2)°C drying oven to dry for 45 min. At last, the precipitate and glass filter were taken out to cool and weight. Meanwhile, the blank experimental is also done. The main reaction in experiment is Eq. (1). Figure 3(a) is the picture of extracted quinoline phosphomolybdate precipitation, Fig. 3(b) is the blank experiment for comparison.   

$$\begin{array}{l}{\text{H}}_{\text{3}}{\text{PO}}_{\text{4}}+12{\text{MoO}}_4^{2-}+24{\text{H}}^{+}+3{\text{C}}_{\text{9}}{\text{H}}_{\text{7}}\text{N}\\ ={\left({\text{C}}_{\text{9}}{\text{H}}_{\text{7}}\text{N}\right)}_3{\text{H}}_3{\text{(PO}}_4\cdot 12{\text{MoO}}_3\text{)}\downarrow +12{\text{H}}_{\text{2}}\text{O }\end{array}$$(1)

Fig. 3.

Comparisons between quinoline phosphomolybdate precipitation and blank experiment. (a): quinoline phosphomolybdate precipitation; (b): the blank experiment.

The effective P₂O₅ content dissolved in 2% citric acid solution was calculated, and then the slag P₂O₅ solubility expression is shown in Eq. (2), where (%P₂O₅)_Available, (%P₂O₅)_Total denote P₂O₅ content that can be dissolved in 2% citric acid solution and total P₂O₅ content in slag respectively, (P₂O₅)_Solubility is slag P₂O₅ solubility.   

$${\left({\text{P}}_{\text{2}}{\text{O}}_{\text{5}}\right)}_{\text{Solubility}}=\frac{{\left({\text{\%P}}_2{\text{O}}_5\right)}_{\text{Available}}}{{\left({\text{\%P}}_{\text{2}}{\text{O}}_{\text{5}}\right)}_{\text{Total}}}$$(2)

3. Experimental Results and Discussion

3.1. The Effect of Slag Composition on P₂O₅ Solubility for CaO–SiO₂–P₂O₅–Fe₂O₃ Slag System

Some synthetic slags were used for experiment in Table 1, to research the effect of slag basic composition on P₂O₅ solubility of CaO–SiO₂–P₂O₅–Fe₂O₃ slag system. The synthetic slag composition of experiment is shown in Table 1.

Figure 4 shows the effect of P₂O₅ and Fe₂O₃ content on P₂O₅ solubility in CaO–SiO₂–P₂O₅–Fe₂O₃ slag. As shown in Fig. 4, about the CaO–SiO₂–P₂O₅–Fe₂O₃ slag system, when P₂O₅ content in the slags is same, with the increasing of slag basicity and Fe₂O₃ content in slag, P₂O₅ solubility is decreased, and the slag basicity and Fe₂O₃ content has greater impact on the slag P₂O₅ solubility with higher P₂O₅ content. While for the slag with same basicity and Fe₂O₃ content, with the increasing of P₂O₅ content in slag, P₂O₅ solubility is decreased. According to the slag ionic structure theory:²²⁾ O^2– in slag is decreased with the increasing of P₂O₅ content in slag, and because the polarization between ions, P⁵⁺ and O^2– will form polyhedral structure ${\text{PO}}_4^{3-}$ complex ion, that is the dense group is formed on the base of O^2–, phosphorus in slag is located among the gaps of O^2– dense group, which makes phosphorus in slag not easily dissolved in 2% citric acid solution, thereby the P₂O₅ solubility of the slag is reduced.

Fig. 4.

The effect of P₂O₅ and Fe₂O₃ content on P₂O₅ solubility for CaO–SiO₂–P₂O₅–Fe₂O₃ slag system.

3.2. Effect of Slag Composition Modification on the P₂O₅ Solubility of P-bearing Slag

Figure 5 shows the effect of different slag composition modification on P₂O₅ solubility in CaO–SiO₂–P₂O₅–Fe₂O₃ slag, synthetic slag composition is shown in Table 1. The P-bearing slag is composed of phosphorus-rich phase, matrix phase and RO phase, the phosphorus in slag is mainly in the existence of nC₂S–C₃P solid solution in the phosphorus-rich phase. Matrix phase is mainly composed of nCaO·SiO₂, the matrix phase also contains some iron and phosphorus, and the RO phase mainly consists of iron oxide or iron and magnesium oxide, the iron in the slag mainly exists in this phase. XRD results of modified synthetic slag with different composition are shown in Fig. 6. It can be seen from Fig. 5, when adding MgO, MnO or Na₂O into CaO–SiO₂–P₂O₅–Fe₂O₃ slag system, the slag P₂O₅ solubility is generally more than 90%. And adding MgO, MnO and Na₂O into CaO–SiO₂–P₂O₅–Fe₂O₃ slag system can break the complex net structure formed by Si–O on certain degree, and also can hinder the precipitation of β-Ca₃(PO₄)₂ crystal with low P₂O₅ solubility during the melting-cooling process.¹⁵⁾ Therefore adding appropriate MgO, MnO or Na₂O content in CaO–SiO₂–P₂O₅–Fe₂O₃ slag can improve the slag P₂O₅ solubility.

Fig. 5.

Effect of different slag composition modification on slag P₂O₅ solubility.

Fig. 6.

XRD results of different slag composition modification.

For A-2(5%MnO) and A-2(10%MnO) slag, MnO mainly enters into RO phase which has no effect on phosphorus existence in slag, so which has little effect on P₂O₅ solubility of the two slags. For A-2(5%MgO) and A-2(10%MgO) slag, the principle is the same.

For A-2(3%Na₂O) and A-2(6%Na₂O) slag, owing to adding Na₂O into slag, Ca ions in phosphorus-rich phase (nC₂S–C₃P solid solution) are partly replaced by Na ions, then Na₂Ca₄(PO₄)₂SiO₄ and Na₃PO₄ are formed, so the phosphorus existence form is changed. But these compounds (Na₂Ca₄(PO₄)₂SiO₄ and Na₃PO₄) and phosphorus-rich phase (nC₂S–C₃P solid solution) have almost same good citric acid solubility, so P₂O₅ solubility of the two slags have little difference.

With the adding of CaF₂, fluorapatite phase (Ca₅(PO₄)₃F) exists along with Ca₁₅(PO₄)₂(SiO₄)₆ in phosphorus-rich phase of slag. As shown in Fig. 5, when the CaF₂ content is respectively 3% and 6%, the slag P₂O₅ solubility is decreased to 17.18% and 11.25%. The adding of CaF₂ makes the slag P₂O₅ solubility decreased significantly, because the fluorine ion can easily enter into the α-Ca₃(PO₄)₂ crystal lattice, and fluorapatite is formed, which plays a strong role in the crystal lattice stability. Fluorapatite has low energy state and stable structure, and its generation process is a highly spontaneous process, so the fluorapatite has extremely stable structure and thus can’t be dissolved in 2% citric acid solution. Therefore, in order to improve the slag P₂O₅ solubility, to make the dephosphorizing converter slag produced phosphate fertilizer slag, and to finally achieve resource utilization of dephosphorizing slag, the slag forming route without fluoride should be used, and using flux such as soda, iron scale etc. instead of fluorite for slag forming.

3.3. Effect of Surplus Calcium Alkalinity on P₂O₅ Solubility of P-bearing Slag

The surplus calcium alkalinity of slag has an important effect on P₂O₅ solubility of P-bearing slag. Some thermal experimental slag in Table 1 was used for experiment, to research the effect of surplus calcium alkalinity on P₂O₅ solubility. The surplus calcium alkalinity expression is shown in Eq. (3), where (%CaO)_T, (%MgO)_T, (%SiO₂)_T denote CaO, MgO and SiO₂ mass fraction in slag respectively, ${\left(\text{\%CaO}\right)}_{3\text{CaO}\cdot {\text{P}}_{\text{2}}{\text{O}}_{\text{5}}}$ is CaO mass fraction combined with phosphorus to form 3CaO·P₂O₅,%, R^θ is the surplus calcium alkalinity.   

$${\text{R}}^{\theta }=\frac{\frac{{\left(\text{\%CaO}\right)}_T-{\left(\text{\%CaO}\right)}_{3\text{CaO}\cdot {\text{P}}_{\text{2}}{\text{O}}_{\text{5}}}}{56}+\frac{{\left(\text{\%MgO}\right)}_T}{40}}{\frac{{\left({\text{\%SiO}}_{\text{2}}\right)}_T}{60}}$$(3)

Figure 7 shows the effect of slag surplus calcium alkalinity on citric acid solubility, and the relationship between slag property indicators and slag P₂O₅ solubility is shown in Table 3. For the CaO–SiO₂–P₂O₅–Fe₂O₃ slag system, when the slag basicity is constant, (%CaO)/((%SiO₂)+(%P₂O₅)) and surplus calcium alkalinity are also increased with increasing of (%CaO)/(%P₂O₅) and (%SiO₂)/(%P₂O₅), the slag P₂O₅ solubility is increased correspondingly.

Fig. 7.

The effect of surplus calcium alkalinity on slag P₂O₅ solubility.

Table 3. The relationship between slag property indicators and slag P₂O₅ solubility.

No.	P2O5 solubility/%	(%CaO)/ (%P2O5)	(%SiO2)/ (%P2O5)	(%CaO)/ ((%SiO2)+ (%P2O5))	Surplus calcium alkalinity
A-1	92.31	7.62	3.05	1.88	2.26
A-2	84.70	4.29	1.71	1.58	1.94
A-3	58.47	2.06	0.83	1.13	1.14
B-1	96.90	8.22	4.12	1.61	1.83
B-2	93.54	4.67	2.33	1.40	1.60
B-3	76.05	2.29	1.15	1.07	1.04
C-1	98.39	8.40	5.60	1.27	1.38
C-2	95.51	4.80	3.20	1.14	1.21
C-3	82.86	2.40	1.60	0.92	0.81
A-2(5%MgO)	79.59	3.93	1.57	1.53	2.35
A-2(10%MgO)	92.81	3.57	1.43	1.47	2.84
A-2(5%MnO)	98.28	3.93	1.57	1.53	1.87
A-2(10%MnO)	96.89	3.57	1.43	1.47	1.79
A-2(3%Na2O)	96.99	4.07	1.63	1.55	1.90
A-2(6%Na2O)	96.13	3.86	1.54	1.52	1.86
A-2(3%CaF2)	17.18	4.07	1.63	1.55	1.90
A-2(6%CaF2)	11.25	3.86	1.54	1.52	1.86

As shown in Fig. 7 and Table 3, for the CaO–SiO₂–P₂O₅–Fe₂O₃–X (X represents MgO, MnO, Na₂O or CaF₂) slag system, except the A-2(MgO) slag, (%CaO)/(%P₂O₅) and (%SiO₂)/(%P₂O₅) are increased in other slag system, (%CaO)/((%SiO₂)+(%P₂O₅)) and surplus calcium alkalinity are also increased at the same time, the slag P₂O₅ solubility is increased correspondingly. Surplus calcium alkalinity is not increased with increasing of (%CaO)/(%P₂O₅) and (%SiO₂)/(%P₂O₅) for A-2(MgO) slag, but slag P₂O₅ solubility is also increased with the increasing of surplus calcium alkalinity. So, for the same slag system, the slag P₂O₅ solubility is increased with the increasing of surplus calcium alkalinity.

3.4. Effect of Industrial Slag Modification on Slag P₂O₅ Solubility

In order to fully recycle the phosphorus resource (used for phosphate fertilizer) in slag and further confirm feasibility of slag melting modification and separating phosphorus-rich phase from P-bearing slag by magnetic separation method, SiO₂, Al₂O₃ and TiO₂ were added into converter slag of one steel plant, and the effect of slag melting modification on slag P₂O₅ solubility was researched. The composition of original slag and modified slag is listed in Table 2.

The comparison of P₂O₅ solubility between original and modified slag is shown in Fig. 8. For the SiO₂ modification, slag P₂O₅ solubility is increased with the increasing of SiO₂ content, mainly because the increased SiO₂ in slag can effectively prevent the precipitation of β-Ca₃(PO₄)₂ phase with low P₂O₅ solubility, phosphorus is mainly in the existence of α-Ca₃(PO₄)₂ phase with high P₂O₅ solubility in slag rapid cooling process.²³⁾ So adding SiO₂ into slag can improve the slag P₂O₅ solubility, which is related to domestic and foreign companies are adding sand into slag to improve the SiO₂ content during slag phosphate fertilizer production.²³⁾ However, When the addition of SiO₂ is excess (R<1.5), the precipitation amount of 2CaO·SiO₂ phase decreases significantly even disappears, CaO·SiO₂ phase generates, so the generation of nC₂S–C₃P solid solution is thermodynamically difficult, which is adverse to phosphorus enrichment and recovery in slag. Therefore, the suitable modification basicity of slag should be 1.5–2.0,¹⁹⁾ the XRD results of SiO₂ modification slag are shown in Fig. 9.

Fig. 8.

XRD results of original slag 1 and SiO₂ modified slag.

Fig. 9.

XRD results of original slag 2, Al₂O₃ and TiO₂ modified slag.

For Al₂O₃ and TiO₂ modification, P₂O₅ solubility of modified slag is increased slightly compared to original slag with the increasing of Al₂O₃ and TiO₂ content, which is mainly due to the added Al₂O₃ and TiO₂ reacted with Ca and Si in nC₂S–C₃P solid solution, which respectively promoted the precipitation of gehlenite (Ca₂Al₂SiO₇) and CaTiSiO₅, CaTiO₃ during cooling process, so nC₂S–C₃P solid solution with high phosphorus content was formed, which is coincides with the XRD results in Fig. 10.

Fig. 10.

Effect of industrial slag modification on slag P₂O₅ solubility.

Al₂O₃ and TiO₂ only make the phosphorus content of phosphorus-rich phase increased, does not change the existence form of phosphorus element in slag, so P₂O₅ solubility was slightly increased after Al₂O₃ and TiO₂ melting modification. The reaction mechanism between Al₂O₃, TiO₂ modification and phosphorus-rich phase are shown in expression (4) and (5) respectively, where subscripts (ss), (in slag) and (S) denote the solid solution, modifier in slag and liquid phase slag , respectively. So according to above mentioned description, slag melting modification of SiO₂, Al₂O₃ and TiO₂ can improve slag P₂O₅ solubility, but P₂O₅ solubility was slightly increased after Al₂O₃ and TiO₂ melting modification   

$$\begin{array}{l}{\text{Ca}}_{\text{3}}{\text{(}{\text{PO}}_{\text{4}}\text{)}}_{\text{2}}{\text{-yCa}}_{\text{2}}{\text{SiO}}_{\text{4(ss)}}+{\text{xAl}}_{\text{2}}{\text{O}}_{\text{3(in}\mathrm{\hspace{0.17em} }\text{slag)}}\\ ={\text{xCa}}_{\text{2}}{\text{Al}}_{\text{2}}{\text{SiO}}_{\text{7(S)}}+{\text{Ca}}_{\text{3}}{\text{(}{\text{PO}}_{\text{4}}\text{)}}_{\text{2}}{\text{-(y-x)Ca}}_{\text{2}}{\text{SiO}}_{\text{4(ss)}}\end{array}$$(4)

  

$$\begin{array}{l}{\text{Ca}}_{\text{3}}{\text{(}{\text{PO}}_{\text{4}}\text{)}}_{\text{2}}{\text{-yCa}}_{\text{2}}{\text{SiO}}_{\text{4(ss)}}+{\text{2xTiO}}_{\text{2(in slag)}}\\ ={\text{xCaTiSiO}}_{\text{5(S)}}+{\text{xCaTiO}}_{\text{3(S)}}+{\text{Ca}}_{\text{3}}{\text{(}{\text{PO}}_{\text{4}}\text{)}}_{\text{2}}{\text{-(y-x)Ca}}_{\text{2}}{\text{SiO}}_{\text{4(ss)}}\end{array}$$(5)

3.5. Effect of Slag without Fluoride on P₂O₅ Solubility

To further verify the effect of fluorine on the slag P₂O₅ solubility, the No. 2 slag is especially treated with no fluorine, synthetic slag composition is shown in Table 4. Figure 11 is the comparison of P₂O₅ solubility between industrial modified slag and modified slag without fluoride.

Table 4. Chemical composition of industrial modified slag without fluoride (mass fraction, %).

Slag sample	CaO	SiO2	Fe2O3	P2O5	Al2O3	MgO	TiO2	CaF2	MnO	R	P2O5 solubility/%
Original slag 1	46.72	11.68	19.47	10	1.07	5.46	0.97	0.37	4.26	4.0	58.35
No. 2	41.35	20.68	17.23	10	0.95	4.83	0.86	0.33	3.77	2.0	72.43
No. 2 slag without fluoride	41.50	20.76	17.29	10	0.95	4.85	0.86	0	3.78	2.0	95.95

Fig. 11.

Effect of slag without fluoride on slag P₂O₅ solubility.

As shown in Fig. 11, the P₂O₅ solubility of original slag 1 is 58.35%, the P₂O₅ solubility of No. 2 slag is increased to 72.43% after SiO₂ modification. And after the No. 2 slag without fluoride, the slag P₂O₅ solubility is increased to 95.95%, which is mainly that fluorapatite that can’t be dissolved in 2% citric acid is not generated after the No. 2 slag without fluoride, the XRD results of industrial modified slag without fluoride are shown in Fig. 12. Therefore, fluorine content in slag has an important influence on the slag P₂O₅ solubility, and the fluorine content in slag must be controlled to obtain higher slag P₂O₅ solubility.

Fig. 12.

XRD results of industrial modified slag without fluoride.

4. Conclusion

The P₂O₅ solubility of P-bearing steelmaking slag was fully researched; the effect of different conditions on the P₂O₅ solubility of P-bearing slag was analyzed. The key findings are as follows:

(1) For the CaO–SiO₂–Fe₂O₃–P₂O₅ slag system, when slags have same basicity, slag P₂O₅ solubility is decreased with the increasing of P₂O₅ content. As for the slags with other same factors, slag P₂O₅ solubility is decreased with the increasing of Fe₂O₃ content and slag basicity.

(2) When adding MgO, MnO and Na₂O into CaO–SiO₂–P₂O₅–Fe₂O₃ slag system, the citric acid solubility of slag is generally more than 90%. And adding MgO, MnO and Na₂O into CaO–SiO₂–P₂O₅–Fe₂O₃ slag system can break the complex net structure formed by Si–O on certain degree, and also hinder the precipitation of β-Ca₃(PO₄)₂ crystal with low citric acid solubility during the melting-cooling process, so adding appropriate MgO, MnO or Na₂O content in CaO–SiO₂–P₂O₅–Fe₂O₃ slag can improve the slag P₂O₅ solubility. While for CaO–SiO₂–P₂O₅–Fe₂O₃–X (X represents MgO, MnO or Na₂O) slag system, different amounts of MgO, MnO or Na₂O has little effect on the P₂O₅ solubility, which is mainly due to MnO mainly enters into RO phase which has no effect on phosphorus existence form in slag, and Na ions in Na₂O can partly replace Ca ions in phosphorus-rich phase (nC₂S–C₃P solid solution), then Na₂Ca₄(PO₄)₂SiO₄ and Na₃PO₄ are formed, and they and phosphorus-rich phase (nC₂S–C₃P solid solution) have almost same good citric acid solubility.

(3) Adding CaF₂ into slag can easily form the fluorapatite that can’t be dissolved in 2% citric acid solution, so the slag P₂O₅ solubility was significantly decreased. Adding 0.5% of CaF₂ into slag, the slag P₂O₅ solubility is decreased to lower than 80%, when the added CaF₂ content is more than 3%, the slag P₂O₅ solubility is decreased to lower than 20%. Slag forming route with no fluoride or less fluoride (the fluoride content in slag is lower than 0.5%) should be used in the steel-making process to make the dephosphorizing converter slag produced slag phosphate fertilizer, and achieve resource utilization of dephosphorizing slag.

(4) Al₂O₃ and TiO₂ modification only increases the phosphorus content in phosphorus-rich phase, does not change the existing form of phosphorus in slag, thus the slag citric acid -solubility of melting slag is increased slightly after Al₂O₃ and TiO₂ modification. However, adding suitable modifier SiO₂ prevents the precipitation of β-Ca₃(PO₄)₂ with low P₂O₅ solubility, slag P₂O₅ solubility is increased significantly. So considering the phosphorus enrichment and resource utilization recovery during modification process, the suitable slag bacisity is 1.5–2.0.

References

• 1)   H. J.  Li,  H.  Suito and  M.  Tokuda: ISIJ Int., 35 (1995), 1079.
• 2)   M.  Ishikawa: ISIJ Int., 46 (2006), 530.
• 3)   T.  Takahashi,  H.  Tanabe,  Y.  Yao,  S.  Inoue,  K.  Watanabe,  S.  Kohira,  R.  Kawabata,  A.  Shiroyama,  H.  Matsuno and  T.  Murai: Chinese Patent, CN1509261A, (2004).
• 4)   H.  Motz and  J.  Geiseler: Waste Manage, 21 (2001), 285.
• 5)   H.  Suito and  R.  Inoue: ISIJ Int., 46 (2006), 180.
• 6)   H. S.  Gao: World Metal Bulletin, Chinese iron and Steel Industry Association, Beijing, (2012).
• 7)   X. R.  Wu,  P.  Wang,  L. S.  Li,  Z. J.  Wu and  R. H.  Chen: Ironmaking Steelmaking, 38 (2011), 185.
• 8)   P. K.  Son and  Y.  Kashiwaya: ISIJ Int., 48 (2008), 1165.
• 9)   H.  Kubo,  K.  Matsubae-Yokoyama and  T.  Nagasaka: ISIJ Int., 50 (2010), 59.
• 10)   B.  Deo,  J.  Halder,  B.  Snoeijer,  A.  Overbosch and  R.  Boom: Iron Steel Making, 32 (2005), 54.
• 11)   X.  Yang,  H.  Matsuura and  F.  Tsukihashi: ISIJ Int., 49 (2009), 1298.
• 12)   S.  Kitamura,  S.  Saito,  K.  Utagawa,  H.  Shibata and  D. G. C.  Robertson: ISIJ Int., 49 (2009), 1838.
• 13)   K.  Shimauchi,  S.  Kitamura and  H.  Shibata: ISIJ Int., 49 (2009), 505.
• 14)   R.  Boom,  S.  Riaz and  K. C.  Mills: Ironmaking Steelmaking, 32 (2005), 21.
• 15)   Y. H.  Wang,  B.  Xie,  J.  Diao,  X. J.  Li and  J. X.  Zhang: J. Univ. Sci. Technol. Beijing, 33 (2011), 323.
• 16)   J.  Diao,  B.  Xie,  Y. H.  Wang and  X.  Guo: ISIJ Int., 50 (2010), 768.
• 17)   K.  Ito,  M.  Yanagisawa and  N.  Sano: Tetsu-to-Hagané, 68 (1982), 342.
• 18)   K.  Matsubae-Yokoyama,  H.  Kubo and  T.  Nagasaka: ISIJ Int., 50 (2010), 65.
• 19)   J.  Diao,  B.  Xie,  Y. H.  Wang and  X.  Guo: ISIJ Int., 52 (2012), 955.
• 20)   K.  Yokoyama,  H.  Kubo,  K.  Mori,  H.  Okada,  S.  Takeuchi and  T.  Nagasaka: ISIJ Int., 47 (2007), 1541.
• 21)   L.  Lin,  Y. P.  Bao,  M.  Wang and  H. M.  Zhou: Ironmaking Steelmaking, 40 (2013), 521.
• 22)   X. H.  Huang: Metallurgical Principle, Metallurgical Industry Press, Beijing, (2002), 167.
• 23)   Z. H.  Xu and  S. L.  Li: Steel Slag Phosphatic Fertilizer Production, China Industry Press, Beijing, (1962), 36.