Steel slag has been proven to be an effective environment remediation media for acid neutralization, and a potential aid to mitigate acid mine drainage (AMD) in passive treatment process. But its acid neutralization capacity (ANC) is frequently inhibited by precipitate after a period of time, while the characteristic of the formation process are unclear yet. In this work, ANC of the SS sample was tested using simulated AMD (H2SO4, 0.1M) and real AMD. Steel slag and AMD has been characterized on pH, ANC, as well as Ca, Mg, Al, Mg-bearing ingredients. Microscopic properties characterization and neutralization experiment results shown that Ca-bearing constitutes leaching and sulfate formation were the two main categories reactions throughout the neutralization process. A prominent transition point of the two kind reactions was selected at 40 % of the neutralization process. Microscopic properties characterization indicated Ca3SiO5 (C3S) played a dominant role among Ca-bearing components in alkaline releasing stage for the present sample. Morphology, pore distribution, composition, surface area and other microscopic properties of the neutralized slag were significantly changed by the crystallized CaSO4 precipitates in sulfate formation stage, which hindered the alkaline releasing behaviors gradually. Neutralization experiments conducted by real AMD suggested that the steel slag ANC property was also influenced by the contained high concentration metal ion due to the precipitate reactions except for sulfate formation reactions.
Research Article
Microscopic properties and neutralization mechanism of steel slag in acid mine drainage (AMD) remediation process
https://doi.org/10.21203/rs.3.rs-1887274/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
acid mine drainage
steel slag
acid neutralization capacity
alkaline
Steel slags are a category of alkaline industrial waste produced in steel making process. The crude steel output from China was 1053.3 million tons in 2020, accounting for 56.7% of the whole world production (Yang et al. 2021). The amount of steel slag produced in 2020 was 158 million ton according to the ratio that there were 150 Kg steel slags produced per ton of crude steel (Mohammed and Yaacob 2016, Feng et al. 2004, Zhang et al. 2020).
As the largest crude steel production country, China’s steel slag utilization ratio was only about 30%, which is lower than the developed countries. The present utilzation ratios in various applications for China and developed countries are listed in Table 1 (Wu et al. 2021). It can be seen that slag utilization are mainly in road project construction, cement production and other construction materials for the countries (Saha et al. 2019, Yang et al. 2021). Although many efforts had been made to increase the steel slag utilization ratio as construction materials, performance of produced construction materials was frequently substandard (Zvimba et al. 2017), because free CaO (f-CaO), free MgO (f-MgO) and other easily hydrated components could deteriorate the cement stability associated with the volume stability (Zhang et al. 2020, Yu et al. 2015). When steel slag was used in unbound and bitumen bound layers, its acceptable contents of f-CaO and f-MgO should be lower than 7% and 4% (Zhang et al. 2020). Therefore, it is necessary to promote the SS utilization ratios on other alternative methods, such as soil remediation (Sheng, Huang et al. 2014), water treatment (Tabelin et al. 2020, Sithole et al. 2020, Zhan et al. 2019, Asere et al. 2019), AMD neutralization (Zvimba et al. 2017, Saha et al. 2019) and other promising pollution control applications.
| Inner recycling | Road | Cement | Construction | Temporary storage | Others | Unutilized | |
|---|---|---|---|---|---|---|---|
| Japan | 20.8 | 32.4 | 3.4 | 34.8 | - | 7 | 1.6 |
| USA | - | 49.7 | 3.3 | 16 | - | 15.4 | 15.6 |
| China | 7.5 | 2.6 | 9.3 | - | - | 4.7 | 70 |
| Europe | 11 | 43 | 5 | 6 | 19 | 3 | 13 |
Previous studies confirmed that steel slags were more excellent in alkaline production than lime stone (Molahid et al. 2019), lime, soda alkaline and dolomite (Sukati et al. 2018). Therefore, when the steel slag was used in the AMD treatment, acidity in waste water was neutralized by alkaline components in steel slags, including f-CaO, f-MgO, Ca2SiO4 (C2S), Ca3SiO5 (C3S), CaFe2O4 (CF2) (Kruse et al. 2012). Previous experiments found that alkalinity concentrations of limestone leach bed were around 5 mg/L as CaCO3 with a maximums concentration of 80 mg/L as CaCO3, while the average alkalinity concentration of steel slag was above 100 mg/L as CaCO3 (Goetz and Riefler 2014a, Kruse et al. 2012). Another superiority of steel slag was its persistence on releasing time at high alkaline concentrations because steel slag does not absorb CO2 from air and convert back to relatively insoluble calcite (Wang and Yan 2010, Goetz and Riefler 2014b, Kruse et al. 2012). It is an extremely important property for AMD remediation, since the high levels of alkalinity could last for many years. However, using steel slag to treat AMD was not a widespread application in practice, as reported it was only used in the area of West Virginia, U.S. and South Africa in the form of steel slag leach bed (SSLB) (Goetz and Riefler 2014, Manchisi et al. 2020). To the authors' knowledge, remediation of AMD by steel slag has not been attempted in China.
Although the steel slag does not react with CO2 in air, its alkalinity releasing behaviors was influenced by many factors when it was used in applications. Performance of operated steel slag leach bed (SSLB) showed that the ANC of steel slag were influenced by effluent quality (Goetz and Riefler 2014b), slag alkalinity loading (Name and Sheridan 2014), slag particle size distribution (Alizadeh and Naseri 2014), etc. Compared performance of the SSLB built for one year and historical bed operated for many years, it was found that the precipitates in the effluent piping of the SSLB and internal of slags reduced the ANC of the slag resulting in a lower effluent pH than the designed value (Tabelin et al. 2020, Goetz and Riefler 2014b).
To ensure AMD effluent meet designed value or emission standard, it is of significance to guarantee steel slag in SSLB have sufficient alkalinity to neutralize AMD acidity and understand the precipitation process in pipes. However in real AMD remediation process, it is extremely difficult to monitor the changing process for the alkalinity consumption process since the consumption time consuming. In this work, slag ANC and microscopic properties were systematically investigated, aiming to explain the precipitate formation process and to provide operational guide for AMD remediation using steel slag.
2.1 Materials
Basic oxygen furnace steel slag from Shougang Group Shuicheng Steel Co., Ltd. (Guizhou Province, China) was used in the study. Before used, the raw steel slag were dried overnight at 105°C, ball milled and screened through sieves of 20 mesh, 50 mesh, 100 mesh and 200 mesh.
Both simulated AMD samples and real AMD samples were used in the study. Simulated AMD was prepared by diluting concentrated sulfuric acid to 0.1 M. Real AMD was collected from an abandoned coal mine in Guizhou Province, China, its composition and pH were determined by inductively coupled plasma (ICP, PerkinElmer Avio 500) and pH meter (PHS-3E), shown in Table 2.
Table 2 Compositions of real AMD (mg/L)
2.2 Experimental procedure
2.2.1 Alkalinity production determination
The term alkalinity production refers to the alkaline content in the leaching solution of steel slags, which was determined according to the acid titration method (SL83-1994 Chinese). As the method described, 1.0 g steel slag sample of 200 mesh size was mixed with 100mL DI water under a rotating speed of 100r/min for preset time. Filtrate was separated using a 0.45 µm filter (Merck), then its pH and alkalinity were determined, alkalinity was calculated and expressed as calcium carbonate (CaCO3).
2.2.2 ANC determination
ANC determination process was conducted by typical titration method, in which AMD samples from automatic titrator was used to model the inflow into SSLB, while the flask containing slags was used as a neutralization reactor with pH monitored. Before titration procedure, 1.0 g and 100 mL DI water was mixed and stirred at 500 rpm for 4 hours. During titration process, solution pH was determined pre and after each titration procedure. It was found that when simulated AMD or real AMD added into solution, it would depleted immediately by the alkaline components in solution leached from steel slag, causing the solution pH had a re-rise value after the neutralization reactions. With the added AMD increasing, the re-rise value became smaller because the remained alkaline components in slag was depleted gradually. Therefore, the neutralization endpoint were specified when the raised pH value was less than 0.05 after a titration procedure, i.e. the alkaline components in the steel slag were completely neutralized. When reaching the endpoint, the mixtures were separated by negative pressure filtration method by 0.45 µm filter (Merck). The obtained filtrate was preserved in polypropylene bottles and determined by ICP, while the residues was characterized by X-ray diffraction, scanning electron microscopy with EDS analyzer, and nitrogen adsorption-desorption methods.
2.2.3 Leaching ratio
Leach-ability of slag components were measured by leaching ratios, which were calculated according to following formula.
$$r=\frac{{c}_{1}\bullet V}{{c}_{2}\bullet m}$$
1
Where, \(r\) is leaching ratio for specific components; \({c}_{1}\) is concentration in the leach liquor; \(V\) is volume of filtrate; \({c}_{2}\) is mass percentage in slag samples; \(m\) is mass of the slag samples.
2.2.4 Characterization techniques
Chemical compositions of the steel slag samples were quantitatively analyzed by X-ray fluorescence (XRF) using an XRF-1800 Analyzer (Shimadzu Corp., Kyoto, Japan), shown in Table 2. f-CaO in steel slag was determined according to industry standard YB/T 4328 − 2012 Method for the determination of content of free calcium oxide in steel slag in Chinese. Phase compositions were performed by Powder X-ray diffraction (XRD) using a Smartlab X-ray diffractometer (Rigaku Corp., Tokyo, Japan) with the operating parameters of Cu Kα radiation (λ = 1.5418 Å) and 40 kV/200 mA power generator. An angular range of 10 − 90° 2θ was measured with a step size of 0.02° and a 1 s counting time per step. Identifications of all crystalline phases were undertaken with JADE6 software (Materials Data, Inc., Livermore, CA, USA) and the PDF-2 2004 database (International Centre for Diffraction Data, Newton Square, PA, USA). Morphology of steel slag samples were observed using a scanning electron microscopy (SEM) (Zeiss, Oberkochen, Germany), coupled with Electron Dispersive X-ray Spectroscopy (EDS) to analyse the grain structure and mineralogy. All steel slag samples were sputtered coated with an approximately 5 nm thick layer of carbon before observation. Specific surface areas, average pore size, and pore volume of steel slag samples were obtained by the Brunauer–Emmett–Teller (BET, ASAP2460, Micromeritics, USA) through N2 adsorption at 77K, pore distribution was calculated by the Bearrett-Joyner-Halenda (BJH) method. Composition Ca, Mg, Si, Al, Mn, Fe and Cr in aqueous solutions were determined by inductively coupled plasma (ICP, Perkin Elmer Avio 500).
3.1 Characterization and alkalinity production
Main chemical components of steel slag were shown in Table 3. Although there were some coexisted minor constituents in samples, like ZnO, SrO, MoO3, CuO, they made negligible contribution to ANC. Therefore, only main chemical constituents containing Ca, Fe, Si, Mg, Al, Mn and Cr were considered in the present study.
| Composition | CaO | Fe2O3 | SiO2 | MgO | MnO | Al2O3 | P2O5 | V2O5 | Cr2O3 |
|---|---|---|---|---|---|---|---|---|---|
| Content | 41.82 | 25.05 | 11.89 | 6.61 | 5.48 | 2.18 | 1.59 | 0.73 | 0.69 |
Steel slag is a heterogeneous material containing many crystalline phases. Diffraction lines revealed that major phases in the present sample contained C2S, C3S, C2F, f-CaO, RO phase and 2CaO·Fe2O3 (C2F) as shown in Fig. 2, of which the determined f-CaO in raw steel slag was 2.3%. Microstructure and morphology of raw steel slag samples shown that the slag had a porous surface as shown in Fig. 2 (b). Alkalinity production test result in Fig. 2 (c) suggested that, for the tested L/S ratio (1:100), the stable alkalinity was in a range of 100–110 mg/L as CaCO3, and its corresponding pH was 10.5. The alkalinity releasing result was consistent with previous results by other researchers (Goetz and Riefler 2014a), however, the f-CaO content was not as high as the samples reported previous. It is known that the calcium containing components in BOF slags were mainly originated from the added flux materials among steel making process, like lime, limestone (Zvimba et al. 2017). The unreacted limestone left in steel slag was existed in f-CaO form, while the reacted lime stone were transferred to C2S, C3S, CF and other forms by molten reactions (Manchisi et al. 2020). In other words, f-CaO component in steel slag did not play a dominant role in the alkaline production process, but its porous microstructure provided a big surface area and inner channels which had a positive influence on ANC property (Yang et al. 2021).
3.2 Acid neutralization capacity
ANC test results by simulated AMD was shown in Fig. 3 (a), it can see that with the simulated AMD amount increasing, solution pH decreased from 10.5 to 2.15 at the endpoint. For 1 g sample, its consuming simulate AMD volume was 115mL at endpoint, and it was 40 mL when pH reached a neutral value of 7.03. There was a linear relationship between solution pH and added simulated AMD volume with a square R of 0.94. When pH decreased to 7.03, its calculated ANC values was 8.23 mmol H+/g slag which was agreed well with the experiment result conducted by electric arc furnace (EAF) slag. The EAF slag was neutralized by similar method using HNO3 (1M) as simulated AMD, when pH decreased to 7, the calculated ANC of EAF slag was about 5.5 mmol H+/g slag and 7.5 mmol H+/g slag for short and long experiments respectively (Jinying Yan et al. 2000).
Another significant factor on steel slag ANC property was particles size distribution, because smaller particles had larger surface area and are easier to release the alkaline components (Alizadeh and Naseri 2014). ANC results for slag sample of 20, 50 and 100 mesh size indicated that slag of bigger size had smaller ANC values, as shown in Fig. 3 (b). For example, it was 8.53 mL simulated AMD for sample of 20 mesh size when reaching the endpoint. It was much less than the sample of 200 mesh size. It was reported that particles less than 3 mm were the most effective in AMD treatment because of rapid dissolution due to their high surface area (Simmons et al. 2002). However, fine particle slags in SSLB were more likely to precipitate on the bottom of beds, and to increase running resistance (Alizadeh and Naseri 2014, Iakovleva et al. 2015), especially for passive treatment applications. Furthermore, fine slag particles are liable to release the contained heavy metals. It is a more severe problem with fine slag particles, especially for electric arc furnace steel slag (2021, Alizadeh and Naseri 2014). Therefore, it is of great significance to optimize the slag size distribution according to the ANC demand and the geographical environment of remediation site, so as to balance the advantages and disadvantages of the fine particle size.
3.3 Slag characteristics variation during alkalinity consumption process
To investigate the properties variation in neutralization process, four intermediate ANC tests were conducted by adding 23mL, 46mL, 68mL and 92mL simulated AMD respectively, i.e. 20%, 40%, 60% and 80% of the endpoint volume. For brevity, the four intermediate samples and the previous endpoint sample were denoted as S1, S2, S3, S4, and S5.
3.3.1 Leaching characteristics
Leaching ratios for various slag components were listed in Fig. 4, it can be that leaching ratios of all components generally increased with simulated AMD increasing, but with respective characteristics .
As the dominant sector in steel slag, Ca, Mg, Si, Al-bearing components had been studied extensively in regards to leaching behaviors. Reported leaching experiment had confirmed that the Mg- and Ca-bearing constituents in steel slag contribute hugely to the short-term acid neutralization process, while Si and Al-bearing constituents discharge alkaline materials over a longer period of time due to their lower dissolution rate (Masindi et al. 2021). Slag leaching result by HCl (0.1M) also confirmed that CaO is more active than other Ca-containing minerals, its leaching ratio was more than 80%; followed by was the silicates, its leaching ratio was more than 45%; then were the aluminates and ferrite, their leaching ratios were less than 30% (Hall et al. 2014).
As for the leaching results in the present study, leaching ratio and leaching concentration of Calcium-bearing components increased firstly and then decreased. Sample S2 had the maximum Ca leaching ratio of 28.73%, and leaching concentration was 9.20 mg/L.
Mg-bearing constitutes was another important alkaline resource, whose leaching ratio increased directly with the adding simulated AMD increasing with a maximum value of 56.75% at endpoint. But Mg-bearing ingredients in slags were relatively small, so its final concentration was only 2.03 mg/L.
Si-bearing components in steel slag were mainly existed in form of C3S, C2S, C2MS, therefore, when the Ca/Mg-bearing silicate components dissolved in solution, both the Si-bearing components and Ca-, Mg-bearing components had similar variation trends as illustrated in Fig. 4.
For Al-bearing constitutes, their leaching ratios were close to zero until simulated AMD volume reached 69 mL. PH trends in Fig. 2 suggested that it was in neutral range till adding 60 mL simulated AMD, which suggested that the solution was alkaline until the consuming AMD amount achieved 60 mL. Under that conditions, Al in solutions were likely to precipitate as hydroxide. Leaching concentration of Al was 0.31 mg/L shown in Fig. 4 (b).
For Fe- and Cr-bearing constitutes, their leaching ratio trends were similar to the Al-bearing ones, which may due to the pH variations mentioned above. Final concentration of Fe and Cr were only 2.14 mg/L and 1.06 mg/L at the endpoint.
For Mn containing constitutes, their concentrations in solution increased directly with the increase of the simulated AMD amount, but MnO in steel slag and 0.06 mg/L. They had minor influence on the the slag alkalinity.
3.3.2 Phase characteristics
Steel slag phase transformation in the neutralization process was showed Fig. 5. Compared with the original steel slag sample, samples neutralized by simulated AMD had stronger diffraction intensity, which may be caused by the amorphous mineral components removal when steel slag were washed by DI water and acid solution. Diffraction intensity of CaSO4·0.5 H2O became stronger with the consuming simulated AMD increasing, meanwhile components of the raw samples became weak. Secondly, steel slag was a complicated heterogeneous solid mixture formed in a molten state at high temperature (Sithole et al. 2020), making various mineral constituents complicated. Although the exact disappearance sequence for different mineral in real steel slags were difficult to obtained by XRD determination, some interesting clues were provided, which were highlighted by dotted line in Fig. 4. It is known that f-CaO was one of the most active alkaline component in steel slag, but was not prominent in the XRD patterns because of its few mass percentage. But for sample S2 ,there was a obvious f-CaO, which maybe due to the dissolution of the outer layer of the f-CaO constitute. Further dissolution reactions make f-CaO peak disappeared again. Another interesting line was the CaFeSi2O6 peak in sample S1 to S5 because its persistence in the neutralization process. It can be speculated that the CaFeSi2O6 was not likely to dissolve in simulated AMD. As for Ca2SiO4 and Ca3SiO5, the major Ca-bearing components in steel slag, their peaks had different trends by comparison the diffraction peak in range of 26.2° and 37.4°. The disappearance of Ca2SiO4 peaks was accompanied by the formation of CaSO4·0.5 H2O throughout the neutralization process, while Ca3SiO5 peaks disappeared at the beginning period. Therefor, the Ca3SiO5 phase was possibly easier to react with the simulated AMD than phase Ca2SiO4.
To investigate the phase change for single component of real steel slags in acid solutions, thermodynamic calculation was used to explore the priority of different minerals. Possible reactions between slag components and simulated AMD (0.1 M H2SO4) was listed in Table 4. The related Gibbs free energy of the reactions indicated that the reaction sequence of components with simulated AMD were CaO > MgO > FeO > C2S > C2F > C3S under room temperature (Liang Z., et al. 2020). But some synthetic pure Ca and Mg bearing minerals studied by leaching experiments and thermodynamic analysis thought that the leaching order for different phase is CaO > (Ca3Al2O6, γ-Ca2SiO4, Ca3MgSi2O8, and Ca2MgSi2O7) > (Ca12Al14O33, and Ca2Fe2O5).
| Component | Reactions |
|---|---|
| \(\text{C}\text{a}\text{O}\) | \(\text{C}\text{a}\text{O}+2{H}^{+}\to {Ca}^{2+}+{H}_{2}O\) |
| \(\text{M}\text{g}\text{O}\) | \(\text{M}\text{g}\text{O}+2{H}^{+}\to {Mg}^{2+}+{H}_{2}O\) |
| \(\text{C}\text{a}{\left(\text{O}\text{H}\right)}_{2}\) | \(\text{C}\text{a}{\left(\text{O}\text{H}\right)}_{2}+2{\text{H}}^{+}\to {\text{C}\text{a}}^{2+}+{2\text{H}}_{2}\text{O}\) |
| \(\text{M}\text{g}{\left(\text{O}\text{H}\right)}_{2}\) | \(\text{M}\text{g}{\left(\text{O}\text{H}\right)}_{2}+2{\text{H}}^{+}\to {\text{M}\text{g}}^{2+}+{2\text{H}}_{2}\text{O}\) |
| \({{\text{C}\text{a}}_{2}\text{S}\text{i}\text{O}}_{4}\) | \({{\text{C}\text{a}}_{2}\text{S}\text{i}\text{O}}_{4}\to 2CaO+{SiO}_{2}\) |
| \(2\text{C}\text{a}\text{O}+4{\text{H}}^{+}\to {2\text{C}\text{a}}^{2+}+2{\text{H}}_{2}\text{O}\) | |
| \({{\text{C}\text{a}}_{3}\text{S}\text{i}\text{O}}_{5}\) | \({{\text{C}\text{a}}_{3}\text{S}\text{i}\text{O}}_{5}\to 3CaO+{SiO}_{2}\) |
| \(3\text{C}\text{a}\text{O}+6{\text{H}}^{+}\to {3\text{C}\text{a}}^{2+}+3{\text{H}}_{2}\text{O}\) | |
| \({{\text{C}\text{a}\text{F}\text{e}}_{2}\text{O}}_{5}\) | \({{\text{C}\text{a}\text{F}\text{e}}_{2}\text{O}}_{5}\to CaO+{{Fe}_{2}O}_{3}\) |
| \(3\text{C}\text{a}\text{O}+6{\text{H}}^{+}\to {3\text{C}\text{a}}^{2+}+3{\text{H}}_{2}\text{O}\) | |
| \({\text{C}\text{a}\text{S}\text{O}}_{4}\) | \({\text{C}\text{a}}^{2+}+{\text{S}\text{O}}_{4}^{2-}\to {\text{C}\text{a}\text{S}\text{O}}_{4}\) |
For the typical components of BOF slags, the alkaline components in steel slag could be classified into three groups according to their solubility. The first group were f-CaO, MgO, C2F; the second group included weakly bound CaO, MgO and some loosed bound C2S; while the third group contained some tightly bound C2S, C3S, MgO and FeO) (Bodurtha and Brassard 2000). It means that the dissolved CaO, MgO and FeO were preferential to react with the simulated AMD over C2S, C3F2 and C3S. (Manchisi et al. 2020, Bodurtha and Brassard 2000, Xue et al. 2013).
3.3.3 Morphology characteristics
There was a significant change for slag particle appearance before and after acid neutralization process, morphological features of sample S1 to S5 were shown in Fig. 6. For samples S1 and S2, most slag particles still had a block shapes as raw slag samples. But for samples S3, long trips well crystallized particles were emerged. And as for sample S4 and S5, crystal dimension was much bigger than that in sample S3, in addition, there were some new formed porous precipitate on surface of the crystals. Therefore, with the added simulated AMD increasing, there were more mineral composition dissolved and precipitate. The dissolving reaction made the original steel slag particle size became smaller, while the residual was then wrapped by the new formed precipitate and grown bigger.
Chemical compositions for several specific positions selected on slag surface of various morphology were tested by EDS, shown in Table 5. Position 1 was selected on the surface of a block porous particle in sample S1, its major elements includes calcium, sulfur, oxide, the less content elements were Si, Al, and Fe was not detected. Its chemical position was not conform to sulfate calcium or Ca-bearing silicates in original steel slags. Its high sulfur content indicated the sulfate formation, while the less Si content may due to the dissolution of the silicates in steel slags. Combined the phase transition and the micro-structure determination results, it can speculated that the S1 sample was under the coupled reactions of dissolution and sulfate formation, but the dissolution reaction play a dominant role. Similar analysis were conducted for position 2 to position 6. Take position 2 for example, its Ca, S and O percentage were higher than that of position 1, especially the Si content was very small, thus it may be on sulfate surface. The high content of Ca and S of position 3 further confirmed the existence of sulfate crystal with the simulated AMD increasing. As for position 4, its major elements were Si and O, it maybe the leaching residual after Ca, Mg bearing components dissolution. Position 5 and position 6 were selected on the surface of well-crystallized sulfate in sample S4 and S5, their main elements were Ca, S and O, with minor content of Fe, and Si.
| Composition | Ca | S | O | Mg | Si | Al | Fe |
|---|---|---|---|---|---|---|---|
| 1 | 16.72 | 12.12 | 47.64 | 0.02 | 22.7 | 0.26 | 0.54 |
| 2 | 24.36 | 19.74 | 53.94 | 0.03 | 1.88 | 0.05 | - |
| 3 | 48.05 | 34.12 | 17.7 | - | 0.07 | - | - |
| 4 | 5.8 | 3.96 | 40.62 | 0.02 | 48.37 | 0.58 | 0.58 |
| 5 | 42.12 | 31.62 | 26.09 | 0.04 | - | 0.04 | - |
| 6 | 57.06 | 34.4 | 6.63 | - | 0.54 | - | 1.30 |
Surface area and pore volume determination results were describe in Fig. 7 and Table 5. The results indicating that calculated surface area and total volume increased with the consuming simulated AMD. It is easy to understand that with the dissolution reactions occurring, such as Ca- and Mg-bearing constitutes on the surface and inner of slag particles, producing new pores and dissolution channels formed, which could promote the surface area and pore diameters. Especially, the overall SEM image indicated the slag particle appearance change into strip shape from block shape, which suggested the sulfate crystal was selected on the original slag surface, while the reactions of the dissolved calcium and the added sulfate radical making the crystal grown up. Then the slag particles was encased, and was connected to adjacent particles by the grown sulfate crystals. Meanwhile, the inner undissolved ingredients of the steel slag was blocked by the new grown sulfate, leading the neutralization reaction stopped gradually. In summary, the 40% of the neutralization endpoint was seem to be a turning point for the dissolution reaction and precipitate reaction.
3.4 ANC test by real AMD
To further investigate the influence of real acid mine drainage components on ANC property of the slag sample, neutralization procedure was conducted using real AMD sample.
For acid neutralization capacity test, the titration procedures and endpoint regulation was consistent with the simulated AMD experiments. Its pH variation trends in neutralization process was described in Fig. 8. The consuming real AMD volume was 670 mL and 230 mL for 1g steel slag sample when neutralization process reached endpoint and neutral rang (pH = 7), and the corresponding ANC results were 2.34 mmol H+/g slag and 0.85 mmol H+/g slag. Compare with simulated AMD, steel slag ANC for real AMD was relatively low, which was due to the metal ions, like Fe2+, Mg2+, Al3+. The metals ion in real AMD would also consume alkaline components of steel slag (Sephton and Webb 2019, Tabelin et al. 2020). Component distribution was tested on steel slag surface was detected by EDS detector, the results indicated that the Ca-bearing areas were overlapped to the S area, thus it can be speculated the CaSO4 containing compound was formed and precipitated in the neutralization process. Moreover, the Fe-, Si- and Mg- enriched areas on the surface were observed at else areas, which may be caused by the hydroxides the of the metal ions precipitated (Eq. 1) or adsorbed (Eq. 2) from AMD solution as followings.
> Si - OH··· H- O- H [Me(OH2)3]2+⟷ > Si- O Me + H3O+ (1)
Me2++OH−⟷ Me(OH)++OH−⟷ Me(OH)2 (2)
Combining the obtained microscopic properties and ANC results in the neutralization process of the steel slag samples, overall reaction between AMD and slag samples could be summarized. In summary, slag leaching and sulfate precipitate were two categories reactions throughout the neutralization process with 40% of endpoint simulated AMD volume as a turning points. Although the two kinds reaction occurred simultaneous, with solution pH decreased from alkaline to acidic, each component presented diverse changes.
In the initial stage, slag leaching reaction played a dominant role since f-CaO, MgO, some loosed bound CaO and C2S were released into the solution and hydrated with the acidic component in AMD, which causing Ca concentration increase and sulfate calcium on slag surface. As for the slag samples in this work, the easily leaching Ca-bearing constitutes were f-CaO, C2S, C3S and CF, among which C3S may contributed more to alkaline neutralization capacity, for the f-CaO amount is few and the C2S is relative less active. With the simulated AMD volume continues increasing, the formed sulfate calcium crystal grown up making the original steel slag appearance change into long strip forms by wrapping or connecting steel slag particles. Also the dissolution of the the alkaline ingredients make the samples had a larger pores diameter and surface areas, but the channels of the steel slag particles were blocked by the produced sulfate which make the alkaline releasing behaviors gradually stopped. Similar external coating, armors of iron hydroxides were also found when AMD was neutralized by lime in the passive treatment (Skousen et al. 2019, Manchisi et al. 2020). For the real AMD titration sample, the precipitate were a mixture of sulfate and metal hydroxides from real AMD, and that is why the real AMD sample had a smaller ANC values.
AMD neutralization remediation by steel slag could make full use of the alkaline substances in steel slag. With the dissolution of slag proceeding, calcium and magnesium concentration in solution increased. The maximum leaching ratio of Ca was 28.73% when 40% of the endpoint amount simulated AMD used. It can be considered as the turning point of dissolution stage and precipitate stage. The new formed crystallized sulfate on the surface of steel slag prevent the inner constitutes in the steel slags further leach out. The ANC results indicated that there was a linear relationship between pH variation and AMD amount for both simulated and real AMD. When the pH decreased to neutral range, the calculated ANCs were 8 mmol H+/g slag and 0.85 mmol H+/g slag for simulated AMD and real AMD, respectively. In addition, the particle size distribution had a significant influence on the ANC property of slag. Therefore, the utilization ratio of alkaline substances in steel slag was not high, especially for large particles employed in the AMD treatment application. To further improve the alkaline substance usage rate, the precipitate process on the surface of slag should be inhibited in future investigation.
Acknowledgments
The authors are grateful for the financial support provided by the Project of Guizhou science and Technology Department (No. 20204Y007), Science and Technology Research Program of Chongqing Municipal Education Commission (KJQN201801126).
Author contribution L. Yang: Conceptualization, Investigation, Methodology, Formal analysis, Validation, Writing - Original draft; Y. Tang: Project administration, Resources, Supervision, Writing -review & editing; M. Yang: Resources, Formal analysis; D. Cao: Data verification, Visualization; C. Lu: Resources, Supervision, Writing -review & editing; Y. He: Supervision, Writing -review & editing;
Data availability All data generated or analysed during this study are included in this published article and its supplementary information files.
Ethics approval and consent to participate Not applicable.
Consent for publication Not applicable.
Competing interests The authors declare no competing interests.
- Alizadeh, Naseri (2014) An Investigation on the Influence of EAF Slag Particle Size on AMD Neutralization Behavior in Static and Dynamic Slag Leaching Systems. Mine Water and the Environment, 34: 204-212.https://doi:10.1007/s10230-014-0302-8
- Asere, Stevens, Du Laing (2019) Use of (modified) natural adsorbents for arsenic remediation: A review. Sci Total Environ, 676: 706-720.https://doi:10.1016/j.scitotenv.2019.04.237
- Bodurtha, Brassard (2000) Neutralization of Acid by Steel-Making Slags. Environmental Technology, 21: 1271-1281.https://doi:10.1080/09593332108618151
- Feng, Van Deventer, Aldrich (2004) Removal of pollutants from acid mine wastewater using metallurgical by-product slags. Separation and Purification Technology, 40: 61-67.https://doi:10.1016/j.seppur.2004.01.003
- Goetz, Riefler (2014a) Geochemistry of CO2 in Steel Slag Leach Beds. Mine Water and the Environment, 34: 42-49.https://doi:10.1007/s10230-014-0290-8
- Goetz, Riefler (2014b) Performance of steel slag leach beds in acid mine drainage treatment. Chemical Engineering Journal, 240: 579-588.https://doi:10.1016/j.cej.2013.10.080
- Hall, Large, Adderley, et al. Calcium leaching from waste steelmaking slag: Significance of leachate chemistry and effects on slag grain mineralogy. Minerals Engineering, 2014, 65: 156-162.https://doi:10.1016/j.mineng.2014.06.002
- Iakovleva, Mäkilä, Salonen, et al. (2015) Acid mine drainage (AMD vs unmodified and modified limestone. Ecological Engineering, 81: 30-40.https://doi:10.1016/j.ecoleng.2015.04.046
- Jinying Yan (2000) The long-term acid neutralizing capacity of steel slag. WASTE MANAGEMENT, 20: 217-223.https://doi:DOI: 10.1016/s0956-053x(99)00318-9
- Kruse, Mackey, Bowman, et al. (2012) Alkalinity production as an indicator of failure in steel slag leach beds treating acid mine drainage. Environmental Earth Sciences, 67: 1389-1395.https://doi:10.1007/s12665-012-1583-5
- Li, Zai Bo, et al. “Leaching Behaviors of Chemical Components from Steel Slag.” Advanced Materials Research, vol. 634–638, Trans Tech Publications, Ltd., Jan. 2013, pp. 245–248.https:// doi:10.4028/www.scientific.net/amr.634-638.245.
- Liang Zhu, Li Jia, et al. Selective leaching behaviors of calcium and magnesium from steel slag. China Metallurgy, 2020, 30(5), 51-57. https://doi: 10. 13228/j.boyuan,issn 1006-9356.20200033.
- Manchisi, Matinde, Rowson, et al. (2020) Ironmaking and Steelmaking Slags as Sustainable Adsorbents for Industrial Effluents and Wastewater Treatment: A Critical Review of Properties, Performance, Challenges and Opportunities. Sustainability, 12: 2118.https://doi:10.3390/su12052118
- Mohammed, Yaacob (2016) Remediation of AMD using industrial waste adsorbents. AIP Conference, 1784: 060043.https://doi:10.1063/1.4966881
- Molahid, Mohd Kusin, Madzin (2019) Role of multiple substrates (spent mushroom compost, ochre, steel slag, and limestone) in passive remediation of metal-containing acid mine drainage. Environ Technol, 40: 1323-1336.https://doi:10.1080/09593330.2017.1422546
- Mulopo, Mashego, Zvimba (2012) Recovery of calcium carbonate from steelmaking slag and utilization for acid mine drainage pre-treatment. Water Sci Technol, 65: 2236-2241.https://doi:10.2166/wst.2012.143
- Name, Sheridan (2014) Remediation of acid mine drainage using metallurgical slags. Minerals Engineering, 64: 15-22.https://doi:10.1016/j.mineng.2014.03.024
- Saha, Sarkar, Sinha (2019) Use of Basic Oxygen Furnace (BOF) Steel Slag for Acid Mine Drainage Treatment: A Laboratory Study. Mine Water and the Environment, 38: 517-527.https://doi:10.1007/s10230-019-00615-3
- Sephton, Webb (2019) The role of secondary minerals in remediation of acid mine drainage by Portland cement. J Hazard Mater, 367: 267-276.https://doi:10.1016/j.jhazmat.2018.12.035
- Sithole, Ntuli, Okonta (2020) Fixed bed column studies for decontamination of acidic mineral effluent using porous fly ash-basic oxygen furnace slag based geopolymers. Minerals Engineering, 154: 106397.https://doi:10.1016/j.mineng.2020.106397
- Skousen, Ziemkiewicz, Mcdonald (2019) Acid mine drainage formation, control and treatment: Approaches and strategies. The Extractive Industries and Society, 6: 241-249.https://doi:10.1016/j.exis.2018.09.008
- Sukati, De Jager, Annandale, et al. (2018) The Hazardous Status of High Density Sludge from Acid Mine Drainage Neutralization. Sustainability, 10: 4185.https://doi:10.3390/su10114185
- Tabelin, Corpuz, Igarashi, et al. (2020) Acid mine drainage formation and arsenic mobility under strongly acidic conditions: Importance of soluble phases, iron oxyhydroxides/oxides and nature of oxidation layer on pyrite. J Hazard Mater, 399: 122844. https://doi:10.1016/j.jhazmat.2020.122844
- Wang, Yan (2010) Hydration properties of basic oxygen furnace steel slag. Construction and Building Materials, 24: 1134-1140. https://doi:10.1016/j.conbuildmat.2009.12.028
- Wu, Peng (2021) Review on global development of treatment and utilization of steel slag. Environmental Enginerring (Chinese), 39: 161-165. https://doi:10.13205/j.hjgc.202101025
- Xue, Wu, Zhou (2013) Adsorption characterization of Cu(II) from aqueous solution onto basic oxygen furnace slag. Chemical Engineering Journal, 231: 355-364.https://doi:10.1016/j.cej.2013.07.045
- Yang, Lu, Quan, et al. (2021) Mechanism of Acid Mine Drainage Remediation with Steel Slag: A Review. ACS Omega.https://doi:10.1021/acsomega.1c03504
- Yu, Liang, Wang, et al. (2015) Phosphate removal from domestic wastewater using thermally modified steel slag. J Environ Sci (China), 31: 81-88.https://doi:10.1016/j.jes.2014.12.007
- Zhan, Xiao, Liang (2019) Removal of Pb(II) from Acid Mine Drainage with Bentonite-Steel Slag Composite Particles. Sustainability, 11: 4476. https://doi:10.3390/su11164476
- Zhang, Chen, Jiang, et al. (2020) The potential utilization of slag generated from iron- and steelmaking industries: a review. Environment Geochem Health, 42: 1321-1334. https://doi:10.1007/s10653-019-00419-y
- Zvimba, Siyakatshana, Mathye (2017) Passive neutralization of acid mine drainage using basic oxygen furnace slag as neutralization material: experimental and modelling. Water Sci Technol, 75: 1014-1024. https://doi:10.2166/wst.2016.579