Journal of Hazardous Materials 286 (2015) 369–378

Contents lists available at ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

Effects of thin-film accelerated carbonation on steel slag leaching R. Baciocchi a , G. Costa a , A. Polettini b,∗ , R. Pomi b a b

Laboratory of Environmental Engineering, University of Rome “Tor Vergata”, Via del Politecnico 1 – 00133 Rome, Italy Department of Civil and Environmental Engineering, University of Rome “La Sapienza”, Via Eudossiana 18 – 00184 Rome, Italy

h i g h l i g h t s • • • •

Thin-film accelerated carbonation of EAF and AOD steelmaking slag was investigated. The leaching behaviour of untreated and carbonated slag samples was studied. Carbonation affected the leaching of both major elements and metal contaminants. Changes in pH and controlling solids explained the observed metal leaching.

a r t i c l e

i n f o

Article history: Received 8 October 2014 Received in revised form 26 December 2014 Accepted 30 December 2014 Available online 31 December 2014 Keywords: Steel slag Accelerated carbonation Metal leaching CO2 sequestration

a b s t r a c t This paper discusses the effects of accelerated carbonation on the leaching behaviour of two types of stainless steel slags (electric arc furnace and argon oxygen decarburisation slag). The release of major elements and toxic metals both at the natural pH and at varying pH conditions was addressed. Geochemical modelling of the eluates was used to theoretically describe leaching and derive information about mineralogical changes induced by carbonation. Among the investigated elements, Ca and Si were most appreciably affected by carbonation. A very clear effect of carbonation on leaching was observed for silicate phases; geochemical modelling indicated that the Ca/Si ratio of Ca-controlling minerals shifted from ∼1 for the untreated slag to 0.5–0.67 for the carbonated samples, thus showing that the carbonation process left some residual Ca-depleted silicate phases while the extracted Ca precipitated in the form of carbonate minerals. For toxic metals the changes in leaching induced by carbonation appeared to be mainly related to the resulting pH changes, which were as high as ∼2 orders of magnitude upon carbonation. Depending on the specific shape of the respective solubility curves, the extent of leaching of toxic metals from the slag was differently affected by carbonation. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Several types of alkaline residues are produced during steel manufacturing. The quantities generated worldwide are quite significant and are estimated to amount to roughly 10–15% by weight of the produced steel [1]. From current steel production data [2], a total crude steel production of 1.58 × 106 Mt was estimated for the year 2013 worldwide, with a main contribution of Asia accounting for 1.06 × 106 Mt (mainly due to China: 74% of the overall Asian production). According to the World Steel Association [3], about 69.5% and 29.4% of the steel produced worldwide is obtained from the basic oxygen furnace (BOF) and the electric arc furnace (EAF) processes, respectively. In Italy, these percentages are different, with a prevalence (∼65.6%) of the EAF technology [4].

∗ Corresponding author. Tel: +39 06 44585037; fax: +39 06 44585037. E-mail address: [email protected] (A. Polettini). http://dx.doi.org/10.1016/j.jhazmat.2014.12.059 0304-3894/© 2014 Elsevier B.V. All rights reserved.

Steel manufacturing slags are generated either during the conversion of iron ores into crude steel in basic oxygen converters (BOF slag) when the integrated steel manufacturing cycle is adopted, or during iron scrap melting in electric arc furnaces (EAF slag from carbon steel [EAF-C] and stainless/high alloy steel production [EAFS]) [1,5,6]. Secondary metallurgical slags are also part of the solid residues of the steel industry, and include argon oxygen decarburisation (AOD) slag (produced during the refining of stainless steel) and ladle furnace (LF) slag (produced during secondary processing of crude steel) [1,6]. In most plants, the slag after cooling is ground and subjected to magnetic separation in order to recover metal components to recycle in the steel production furnace [1]. On average, the steel industry generates about 125.8 kg of basic oxygen furnace (BOF) slag and 168.6 kg of electric arc furnace (EAF) slag per t of crude steel produced [2]. In Europe, the latest available statistics [5] report a production of 21.8 Mt of steelmaking slag in 2010 (of which 10.5 Mt of BOF slag, 6.8 Mt of EAF-C slag, 1.7 Mt of EAF-S slag and 2.8 Mt of secondary slag).

370

R. Baciocchi et al. / Journal of Hazardous Materials 286 (2015) 369–378

All these types of residues are alkaline materials characterised by a prevalence of Ca-containing silicate phases and generally also display relevant amounts of Mg, Al, Fe and Cr, among others, in the form of oxides and/or silicates [6]. The presence of such minerals provides the background for recent research in which steel slag is used as a source of reactive elements for the purpose of CO2 sequestration using accelerated carbonation whereby the material reacts with gaseous CO2 and the alkaline elements are converted into thermodynamically stable carbonate minerals [7–20]. The slag generated during steel manufacturing is usually regarded as a non-hazardous waste owing to its limited release of trace contaminants, as mirrored by compliance with standard leaching criteria (see e.g. [1]). Nevertheless, depending on the leaching conditions some types of slag may in fact still release non negligible amounts of toxic metals, such as typically Ba, Cr, V and Mo [21–28]. As for the environmental behaviour of steel slag, surprisingly only limited literature studies have specifically focused on the mechanisms underlying the leaching process and governing the potential release of toxic elements. The lack of systematic information about the environmental behaviour of steel slag is also evident when accelerated carbonated slags are concerned, with only relatively few studies investigating the mobility of elements of environmental concern in carbonated steel slag [13–15,29–35]. In particular, the mechanisms through which accelerated carbonation affects the extent of leaching of major constituents and toxic elements from the solid matrix are still unclear and thus need to be further elucidated. In the present work, an attempt was made at filling the existing gaps in the knowledge of the leaching-controlling phenomena for accelerated carbonated steel slag. To this aim, an experimental investigation of pH-dependent leaching of untreated and carbonated steel slag, along with geochemical modelling of the resulting leaching solutions, was carried out.

Batch carbonation tests were performed as detailed in Reference [15] using the direct thin-film (L/S ratio = 0.4 l/kg) carbonation route at an operating temperature of 50 ◦ C and a CO2 pressure between 1 and 10 bar. After carbonation, the samples were oven-dried at 50 ◦ C and analysed by calcimetry testing. The mineralogy and leaching behaviour of the carbonated samples were also investigated as described above on samples carbonated under the following conditions: L/S = 0.4 l/kg, T = 50 ◦ C, t = 1 h, pCO2 = 3 bar (EAF slag) or 1 bar (AOD slag). The two samples selected to explore the leaching behaviour were chosen as those that, while keeping the carbonation conditions as mild as possible in view of a potential full-scale application of the process, still allowed to attain a significant carbonation yield. To derive information on the mechanisms governing the leaching of contaminants from the slag, geochemical modelling of the ANC eluates was conducted using Visual Minteq. The standard thermodynamic database included in the modelling code was extended including the stability constants of several phases (see [36] for details). The application of the geochemical speciation code followed a three-step procedure: (1) the program was run using the measured eluate concentrations and pHs as the input data and suppressing precipitation for all solid phases; (2) potential solubility-controlling minerals where chosen among those displaying saturation indices (SI) in the range − 1.5 ≤ SI ≤ + 1.5 and on the basis of likelihood of formation in steel slag materials; and (3) the predicted concentrations of each element in equilibrium with the selected mineral phases were derived according to the empirical equation Ceq = Cmeas (10−SI )1/n [37], where Ceq and Cmeas are the theoretical and measured element concentrations, respectively, and n is the stoichiometric coefficient of the element in the mineral under concern. 3. Results and discussion

2. Materials and methods Slag samples were collected from an Italian stainless steel manufacturing plant after metals removal downstream of the electric arc furnace (EAF slag), and at the outlet of the desulfurisation unit subsequent to the argon oxygen decarburisation section (AOD slag). The EAF slag was quite heterogeneous in particle size and had the following grain size distribution (AASHTO classification): 18% gravel, 31% coarse sand, 30% fine sand, 18% silt and ∼3% clay. The AOD slag was considerably more homogeneous and finer in grain size (90% of particles below 0.150 mm). Considering the critical influence exerted by particle size on carbonation [13], the EAF slag was milled to a particle size below 150 ␮m, while the 150 ␮moversize fraction was simply discarded from the bulk of the AOD slag. Slag characterisation involved the determination of elemental composition, calcite content, mineralogy, acid neutralization capacity (ANC) and leaching behaviour. The elemental composition was determined by alkaline digestion with Li2 B4 O7 at 1050 ◦ C, followed by dissolution in 10% HNO3 and measurement of element concentrations using atomic absorption spectrometry (AAS). The carbonate content was evaluated by calcimetry analysis using a Dietrich-Frühling calcimeter. The mineralogical composition was evaluated by powder XRD analysis with Cu K␣ radiation using a Philips Expert Pro diffractometer equipped with a copper tube operated at 40 kV and 40 mA. The leaching behaviour was investigated through the EN 12457-2 and the CEN/TS 14429 (ANC) leaching tests. Eluate concentrations were determined by AAS analysis, whereas anion concentrations were measured by ion chromatography. All chemical analyses were performed in triplicate; the batch compliance leaching test was run in duplicate, while the ANC test was not replicated.

3.1. Chemical and mineralogical characterisation of the slag The chemical composition of the EAF and AOD slag is reported in Table 1. In view of carbonation, the Ca and Mg content of the slag is particularly relevant: in both types of residues the Ca concentration (35% and 40% for the EAF and AOD samples, respectively), was significantly higher than the Mg content (2.4% and 1.8%, respectively). Appreciable contents of Cr, Fe, Al, Mo and V were also measured in the EAF slag, in particular. The initial calcite content accounted for 3.5% and 4.0% of the EAF and AOD slag mass, respectively. Fig. 1 shows the results of the XRD analysis of the two types of slag. The most relevant phases present in the EAF slag were found to include different silicate minerals such as dicalcium silicate (Ca2 SiO4 , in its ␥-type polymorph), akermanite (Ca2 MgSi2 O7 ), cuspidine (Ca4 Si2 O7 (F,OH)2 ) and gehlenite (Ca2 Al(AlSi)O7 ); other identified minerals included oxide phases (periclase [MgO], a 0.86:1Ca–Al oxide with the formula (CaO)12 (Al2 O3 )7 , a 2:1Cr–Mg oxide having the composition MgCr2 O4 , the Fe(II)–Fe(III) oxide magnetite [FeO·Fe2 O3 ] and quartz [SiO2 ]); fluorite (CaF2 ) and small amounts of calcite (CaCO3 ) were also detected. The XRD pattern of the AOD slag revealed a lower number of peaks, indicating a simpler mineralogical composition. The main phases identified included ␥-dicalcium silicate, fluorite and magnetite, with traces of calcite. The results obtained from the mineralogical analysis confirm the findings of previous investigations on stainless steel slag [21,38]. 3.2. Carbon sequestration capacity and degree of carbonation The carbonation yield attained by the two slag samples is reported in Table 2 in terms of CO2 uptake as a function of pressure

R. Baciocchi et al. / Journal of Hazardous Materials 286 (2015) 369–378

371

the mineralogical changes upon carbonation of the two types of slag, including the full XRD patterns, are reported in [14]. For the EAF sample, the reduction in concentration from the untreated to the carbonated material was particularly relevant for the two oxide phases periclase and Ca–Al oxide as well as for dicalcium silicate, still evident for silicates including akermanite and cuspidine, and only slight for quartz and gehlenite. Other authors [17,38–42] showed elsewhere that mineral phases including, among others, dicalcium silicate, periclase, akermanite and Ca–Al oxides display considerable degrees of conversion upon carbonation. For the AOD slag, the dissolution upon carbonation was found to be particularly evident in the case of dicalcium silicate and periclase, and either hardly or not appreciable for magnetite and fluorite, respectively. For both types of slag, a significant increase in the calcite content was observed after carbonation. 3.3. Acid neutralisation capacity and leaching behaviour Fig. 1. XRD patterns for the EAF and AOD samples (a) dicalcium silicate [Ca2 SiO4 ], (b) akermanite [Ca2 MgSi2 O7 ], (c) Ca–Al oxide [Ca12 Al14 O33 ], (d) magnetite [Fe3 O4 ], (e) periclase [MgO], (f) fluorite [CaF2 ], (g) Cr–Mg oxide [MgCr2 O4 ], (h) calcite [CaCO3 ], (i) gehlenite [Ca2 Al(AlSi)O7 ], (j) cuspidine [Ca4 Si2 O7 (F,OH)], (k) quartz [SiO2 ]). Table 2 CO2 uptake attained upon carbonation of the EAF and AOD slag. CO2 uptake (%) Time

0.5 h

1h

2h

4h

8h

24 h

Pressure EAF AOD EAF AOD EAF AOD EAF AOD EAF AOD EAF AOD 1 bar 3 bar 10 bar

9.3 20.6 9.0 21.1 15.5 23.1 15.8 24.2 15.2 24.2 18.2 25.9 13.7 18.1 14.9 20.1 16.1 21.1 17.1 23.5 14.9 25.9 18.2 28.8 12.6 17.2 14.4 21.4 14.3 24.2 15.2 27.0 15.5 30.7 17.6 30.7

and contact time. While the study of the influence of the operating parameters on the process yield and carbonation kinetics was described elsewhere [15], the notably higher CO2 sequestration capacity of the AOD slag in comparison to the EAF sample must be highlighted here. The final calcite content of the EAF and AOD slag after 24 h of carbonation was 38 and 56% by weight, respectively. Accordingly, while a maximum CO2 uptake of 18% was displayed by the EAF slag, the corresponding value for the AOD sample was 31%. The mineralogical analysis of the treated samples showed that accelerated carbonation was quite effective in dissolving several mineral constituents of the slag. The results from mineralogical (XRD) analysis are reported qualitatively in Table 3 in terms of the observed reduction upon carbonation of the content of the different phases detected in the untreated slag. Further details about

A comparison of the acid neutralisation behaviour of the untreated and carbonated slag samples is shown in Fig. 2. The natural pH of the material decreased upon carbonation from 12.60 to 10.75 for the EAF slag, and from 12.47 to 10.45 for the AOD sample, which is typical for carbonated alkaline materials of different origin (see e.g. [12,33,43–47]) and is also consistent with the higher degree of carbonation attained by the AOD slag. The untreated AOD slag also displayed a larger acid buffering capacity compared to the EAF sample, as indicated by the higher amount of acid required to attain a predetermined pH value. In addition to the native pH, the whole ANC curves were in both cases found to change upon carbonation; the formation of two pH plateaus was observed (one in the pH range ∼9.0–9.5 for the EAF slag and ∼8.5–9.0 for the AOD Table 3 Observed qualitative reduction of the two slags’ contents of different mineral phases upon carbonation as derived from XRD analysis [14]. Mineral

EAF slag

AOD slag

Dicalcium silicate [␥-Ca2 SiO4 ] Akermanite [Ca2 MgSi2 O7 ] Cuspidine [Ca4 Si2 O7 (F,OH)2 ] Periclase [MgO] Ca–Al oxide [(CaO)12 (Al2 O3 )7 ] Quartz [SiO2 ] Gehlenite [Ca2 Al(AlSi)O7 ] Fluorite [CaF2 ] MgCr2 O4 Magnetite [FeO·Fe2 O3 ]

+++ ++ ++ +++ +++ + + n.a. n.a. n.a.

+++ – – +++ – – – n.a. – +

+++ very significant; ++ significant; + detectable; n.a.: not appreciable; − not present in the untreated slag.

Table 1 Chemical composition of the EAF and AOD slag. EAF slag

AOD slag

Major elements

Conc.(%)

Minor elements

Conc.(mg/kg)

Major elements

Conc.(%)

Minor elements

Conc.(mg/kg)

Ca Mg Al Si Na K Fe Cr

35.2 2.47 2.05 13.15 0.34 0.039 3.23 3.73

As Cd Cu Mn Mo Ni Pb Sb V Zn

n.a. 0.3 171 n.a. 227 480 90.7 n.a. 973 260

Ca Mg Al Si

40.33 1.44 0.68 14.15

As Cd Cr Cu Fe K Mn Mo Na Ni Pb Sb V Zn

17.4 3.5 399 528 616 113 329 8, particularly notable for K; however, while K release occurred at similar levels for the two slag samples, Na leaching was one order of magnitude larger for the EAF slag, mirroring the higher total content in the original material. 3.4. Geochemical modelling of metal leaching The study of the mechanisms governing the release of major elements and metal contaminants from the slag is deemed to be particularly important to predict the environmental behaviour of the material in view of either utilization or final disposal. While the original mineralogy of the slag certainly plays a role in determining the degree of element dissolution from the solid matrix, the occurrence of multiple complex phenomena during leaching, including incongruent dissolution, precipitation of secondary minerals, hydration of solid phases and many others, implies that the leaching-controlling phases may not necessarily correspond with the mineralogical constituents of the original slag. Geochemical modelling is therefore believed to be a powerful tool in order to identify the mechanisms governing leaching and the possible solid phases controlling dissolution/precipitation equilibria. Geochemical modelling results are reported in Figs. 6 and 7 for the AOD and EAF slag, respectively. For the untreated slags, the mineral xonotlite, a Ca silicate hydroxide related in structure to wollastonite represented by the chemical formula Ca6 Si6 O17 (OH)2 , was found to be the best candidate for solubility control of Ca at pH values above ∼9.5; ettringite was also selected as a possible leaching-controlling phase in the same pH range. The solubility curve for xonotlite also showed a good match with Si leaching data in the same pH range. Since xonotlite is a C–S–H phase known to be formed under hydrothermal conditions at temperatures above 140 ◦ C and to be unstable at ambient temperatures (see e.g. [50]), solubility control by this mineral is unlikely under the investigated conditions. However, modelling results may be interpreted in the sense that a C–S–H phase with a Ca/Si molar ratio of ∼1 having a stability constant similar to that used for xonotlite (Ca6 Si6 O17 (OH)2 + 5H2 O + 12H+ → 6Ca2+ + 6H4 SiO4 ; log K = 93.37 [51]) may explain the release of Ca from the untreated slag samples. Both C–S–H(1.1), which has the closest Ca/Si ratio to

Fig. 2. ANC curves for untreated and carbonated samples.

R. Baciocchi et al. / Journal of Hazardous Materials 286 (2015) 369–378

373

Fig. 3. Results of the EN 12457 leaching test (* measured concentration below the indicated detection limit).

Fig. 4. Ca release during the ANC test as a function of acid addition.

xonotlite, and C–S–H(0.8) displayed SI values between −0.2 and 1.0 in the pH range 9.5–12.2 (AOD) and 9.8–12.7 (EAF), indicating a general slight oversaturation of leachates in these solids. The presence of C–S–H phases as leaching-controlling minerals for Ca from steel slag is supported by the findings of previous studies which observed C–S–H at Ca/Si ratios of 1.1 [30,34] and 1.7 [24] being formed in slag leachates as a result of the alteration of Ca silicates during leaching. The presence of ettringite as a candidate for solubility control of Ca was proposed by Huijgen and Comans [30] but excluded by De Windt et al. [24]. In our study, ettringite was not detected by XRD analysis, and although its subsequent forma-

tion during leaching cannot be excluded, the fact that the solubility curves for the above mentioned silicate minerals well matched the measured Ca concentrations in the eluates in a wider pH range compared to ettringite may support the conclusion by De Windt et al. [24] that this phase was not responsible for solubility control of major elements for both slag samples. Carbonation involved a marked decrease in Ca concentrations in the eluates at pH > 8, with a change in the solubility-controlling phases compared to the untreated slag, as already noted by Huijgen and Comans [30] and van Zomeren et al. [34]. For both slag samples, under alkaline conditions two new silicate minerals, namely the

374

R. Baciocchi et al. / Journal of Hazardous Materials 286 (2015) 369–378

Fig. 5. Comparison of ANC test results for untreated and carbonated EAF and AOD slag.

hydrated Ca silicate hydroxide phases okenite (CaSi2 O4 (OH)2 ·H2 O) and gyrolite (Ca2 Si3 O7 (OH)2 ·1.5H2 O), were found to explain the measured leachate concentrations as a function of pH; wollastonite (CaSiO3 ) was also identified as a potential solubility-controlling phase (although with a poorer match), while C–S–H(0.8) was capable of explaining Ca solubility at the natural pH of the material only. Okenite and gyrolite are rare silicate phases (the latter known to be formed under hydrothermal conditions similarly to xonotlite) which have not been documented so far for materials such as steelmaking slag. However, as noted for the untreated slag, it may be possible that silicate minerals having Ca/Si ratios of 0.5–0.67 control Ca release from the carbonated slag; this appears to be consistent with previous studies [7,10,32,34] indicating the formation of a Ca-depleted silicate rim around steel slag particles during carbonation. Reasonably, the modelling results may thus be interpreted in the sense that the Ca/Si ratio of minerals controlling Ca leaching decreased from the initial value of ∼1 in the untreated slag to the final value of 0.5–0.67 in the carbonated samples, due to the Ca content reduction of the residual silicate phases caused by the mentioned incongruent dissolution of the original minerals, with preferential release of Ca into the liquid phase. It should also

be emphasised that the eluates were always found to be oversaturated by 1–2 orders of magnitude with respect to Ca carbonate minerals (see the solubility curves for CaCO3 ·H2 O in Figs. 6 and 7), which were indeed clearly detected by calcimetry and XRD testing. According to Meima and Comans [43], oversaturation of leachates with respect to calcite is a common phenomenon observed for weathered waste materials which still needs to be elucidated. On the basis of our modelling results, this is explained considering that Ca leaching from carbonated slag is governed by low-Ca/Si silicate minerals having higher solubilities than the Ca carbonate forms. As observed by Huijgen and Comans [30], Ca depletion of residual silicate minerals as a result of carbonation may also explain the increased Si leaching from the carbonated slag at pH > 10 (as also observed by Chen et al. [31]); in carbonated slag leachates this was clearly found to be controlled by amorphous SiO2 instead of by silicate phases. While amorphous silica dictated Si release from the untreated slag only below pH values of ∼8, the same phase was capable of explaining Si leaching from the carbonated material within the whole pH range investigated (see Fig. 5). The appearance of amorphous silica as a release-controlling phase for Si in the untreated slag can be explained considering the dissolution

R. Baciocchi et al. / Journal of Hazardous Materials 286 (2015) 369–378

375

Fig. 6. Comparison between measured leaching data and theoretical solubility-controlling minerals for the untreated (white circles) and carbonated (grey circles) AOD slag (legend: Art: artinite; Bru: brucite; Diasp: diaspore; Ettr: ettringite; For: forsterite; Gyr: gyrolite; Oke: okenite; Will: willemite; Woll: wollastonite; Xon: xonotlite).

of solid phases from the slag matrix during the leaching process. As discussed above, evidence was gained of dissolution of minerals from the slag occurring incongruently. As a result of this, the dissolution of silicate minerals during leaching leaves a residual Si-enriched leached layer (see e.g. [48,52,53]) that governs the equilibrium between the liquid and solid phases. Solubility control by amorphous SiO2 in the whole pH range investigated for the carbonated slag is further motivated by the fact that a Ca-depleted

silicate rim is present around the solid particles of the slag since the start of the leaching process due to the mineral dissolution effects produced by carbonation (see above). As evident from the data reported in Fig. 5, carbonation led to a reduction in Al leaching at pH > 6. While for the untreated slag the solubility-controlling minerals identified included Al2 O3 at pH < 8, diaspore (AlOOH) at pH > 10 and gehlenite hydrate at intermediate pHs, for the carbonated slag only Al2 O3 at pH < 8 was found to

376

R. Baciocchi et al. / Journal of Hazardous Materials 286 (2015) 369–378

Fig. 7. Comparison between measured leaching data and theoretical solubility-controlling minerals for the untreated (black asterisks) and carbonated (grey asterisks) EAF slag (legend: Art: artinite; Boeh: boehmite; Bru: brucite; Diasp: diaspore; Ettr: ettringite; For: forsterite; Gyr: gyrolite; Oke: okenite; Smith: smithsonite; Will: willemite; Woll: wollastonite; Xon: xonotlite)

explain the observed Al leaching (see Figs. 6 and 7); at higher pH values, leachate concentrations were below the analytical detection limit, so that it was not possible to infer about any potential solubility-controlling mineral. Mg leaching appeared to be unaffected by carbonation at pHs < 9, while a reduction in leachability by up to one order of magnitude was observed under more alkaline conditions. For the untreated

materials, geochemical modelling suggested that solubility control was dictated by the silicate forsterite (Mg2 SiO4 ) at pHs above ∼8. Another potential leaching-controlling mineral included brucite (Mg(OH)2 ); while this phase was also indicated by other studies [24,30] and its solubility curve well matched the leaching data, forsterite appeared to describe the experimental results more closely and in a wider pH range compared to brucite. For the

R. Baciocchi et al. / Journal of Hazardous Materials 286 (2015) 369–378

carbonated slag, possible solubility control appeared to occur by the hydrated Mg carbonate artinite (Mg2 (OH)2 CO3 ·3H2 O) at pH values of 8.6–9.6 (AOD slag) and 8.2–9.5 (EAF slag). At higher pHs, forsterite was again identified as the potential controlling mineral. As a result, the increase in Mg release from the carbonated slag observed in the EN 12457 test (see Fig. 3) was merely related to the decrease in the natural pH of the material upon carbonation rather than to a change in the nature of solubility-controlling minerals in this pH range. As noted above, both K and Na (to a lesser extent) appeared to be affected by carbonation, with a reduction in the amount released at pHs above ∼8, as high as two times for Na and up to one order of magnitude for K. However, geochemical modelling was inadequate to predict potential leaching-controlling elements for K and Na, for both the untreated and the carbonated slags. This was likely related to the association of such elements to complex minerals that are not included neither in the standard nor in the expanded thermodynamic database used. Nevertheless, considering that the EAF slag contained 3.4 times more K and 4.7 times more Na than the AOD sample, a comparison of the experimental leaching curves for these elements may lead to qualitatively infer that, while K release from the two slag samples was controlled by the same mineral form(s), this did not happen for Na. The leaching data for K and Na showed, for both slag samples, a positive correlation with Ca release, possibly indicating that Ca silicates in the original slag contained small inclusions of K and Na that were quantitatively released into the liquid phase during leaching. Among the toxic elements, the three oxyanionic metals Cr, Mo and V (these two analysed only for the EAF slag due to their very low content in the AOD slag sample) were influenced by carbonation apparently with different mechanisms and effects. At the natural pH, Cr and V were more leachable from the carbonated slag, while the opposite occurred for Mo. On the basis of the shape of the leaching curves, it may be inferred that Cr leachability was affected by carbonation only poorly for the EAF slag and to a larger extent for the AOD sample; the latter appeared to display some moderately higher Cr release after carbonation in the whole pH range investigated. Chromium solubility could be modelled at selected pH values by either Cr(III) oxide or hydroxide phases (Cr2 O3 , Cr(OH)3 ) both before and after treatment, although the pH range in which a satisfactory fitting degree was observed differed for the two slag samples: this corresponded to pHs of 4.5–7.5 and 6.5–11.8 for the EAF and AOD slag, respectively. The presence of Cr in the trivalent form in steel slag eluates was also observed by Huijgen and Comans [30] and De Windt et al. [24]; the former work also showed that Cr leaching as a function of pH was not considerably affected by carbonation. Mo leaching from the EAF slag (results not shown graphically here) indicated a shift of the release curves towards lower concentrations upon carbonation, with a reduction in leaching by a factor of 2–4 depending on pH. Geochemical modelling results suggested possible solubility control by powellite (CaMoO4 ) for the carbonated material, although with some underestimation of the solution concentrations in the pH range 4.7–8.2. This feature may tentatively indicate that the observed reduction in Mo release due to carbonation was associated to an indirect effect exerted on Ca availability in the liquid phase. The cationic metal Zn displayed a reduced leachability upon carbonation at the natural pH of the slag. Modelling results suggest that Zn release from the untreated slag was controlled by Zn silicates (willemite, Zn2 SiO4 , at pH above ∼11 and ZnSiO3 at pHs of 8–10), with the same minerals being unable to describe leaching from the carbonated material; however, since the eluate concentration values as a function of pH were quite scattered due to the low release levels, it was not possible to derive any conclusive remark about

377

the mechanisms governing the leaching of Zn from the carbonated slag. 4. Conclusions The specific focus of this manuscript was to investigate the effects of a thin-film carbonation treatment carried out at relatively mild operating conditions on the release of major elements and toxic metals from two types of stainless steel slag. Among the major elements, Ca and Si were found to be those most appreciably affected by carbonation, the shape of the leaching curves being significantly altered after the treatment, owing to changes in the mineral phases responsible for solubility control of these elements. This was tentatively ascribed to a reduction in the Ca/Si ratio of the residual minerals in the carbonated slag caused by the dissolution of (mainly) dicalcium silicate during the carbonation process, as evidenced by XRD investigations. Contaminants including the cationic metal Zn and the oxyanionforming elements Cr, V and Mo were variously affected by carbonation. While the release of Zn and Mo upon the EN 12457 test decreased upon carbonation, Cr and V displayed the opposite trend. However, for Cr the leachability as a function of pH appeared not to be significantly affected by carbonation, whereas Mo showed a certain reduction in leaching after carbonation in the entire pH range investigated. Since in a number of cases modelling was not capable to describe the experimental leaching data, further investigation is currently underway to refine the modelling by including sorption mechanisms in the calculations. References [1] D.M. Proctor, K.A. Fehling, E.C. Shay, J.L. Wittenborn, J.J. Green, C. Avent, et al., Physical and chemical characteristics of blast furnace basic oxygen furnace and electric arc furnace steel industry slags, Environ. Sci. Technol. 34 (2000) 1576–1582. [2] World Steel Association, Statistics archive, http://www.worldsteel.org/statistics/statistics-archive.html, accessed April 2013, 2013. [3] World Steel Association, Steel Statistical Yearbook 2013, Brussels, 2013, pp. 16. [4] Federacciai, The Italian Steel Industry Key Statistics 2013, Milano, 2013, pp. 7. [5] European Slag Association (Euroslag), Position paper on the status of ferrous slag, 2012. [6] C. Shi, Steel slag – its production, processing, characteristics, and cementitious properties, J. Mater. Civ. Eng. 16 (2004) 230–236. [7] W.J.J. Huijgen, G.-J. Witkamp, R.N.J. Comans, Mineral CO2 sequestration by steel slag carbonation, Environ. Sci. Technol. 39 (2005) 9676–9682. [8] S. Teir, S. Eloneva, C.-J. Fogelholm, R. Zevenhoven, Dissolution of steelmaking slags in acetic acid for precipitated calcium carbonate production, Energy 32 (2007) 528–539. [9] D. Bonenfant, L. Kharoune, S. Sauvé, R. Hausler, P. Niquette, M. Mimeault, et al., CO2 sequestration potential of steel slags at ambient pressure and temperature, Ind. Eng. Chem. Res. 47 (2008) 7610–7616. [10] S.N. Lekakh, C.H. Rawlins, D.G.C. Robertson, V.L. Richards, K.D. Peaslee, Kinetics of aqueous leaching and carbonization of steelmaking slag, Metall. Mater. Trans. B. 39 (2008) 125–134. [11] S. Kodama, T. Nishimoto, N. Yamamoto, K. Yogo, K. Yamada, Development of a new pH-swing CO2 mineralization process with a recyclable reaction solution, Energy 33 (2008) 776–784. [12] R. Baciocchi, G. Costa, E. Di Bartolomeo, A. Polettini, R. Pomi, The effects of accelerated carbonation on CO2 uptake and metal release from incineration APC residues, Waste Manage. 29 (2009) 2994–3003. [13] R. Baciocchi, G. Costa, A. Polettini, R. Pomi, Influence of particle size on the carbonation of stainless steel slag for CO2 storage, Energy Procedia 1 (2009) 4859–4866. [14] R. Baciocchi, G. Costa, E. Di Bartolomeo, A. Polettini, R. Pomi, Carbonation of stainless steel slag as a process for CO2 storage and slag valorization, Waste Biomass Valorization 1 (2010) 467–477. [15] R. Baciocchi, G. Costa, E. Di Bartolomeo, A. Polettini, R. Pomi, Wet versus slurry carbonation of EAF steel slag, Greenhouse Gases 1 (2011)312–319. [16] R. Santos, D. Ling, M.B. Guo Blanpain, T. Van Gerven, Valorisation of thermal residues by intensified mineral carbonation, in: 6th Int. Symp. Waste Process. Recycl. Miner. Metall. Ind. 2–5 Oct. 2011, Montreal (Canada), 2011.

378

R. Baciocchi et al. / Journal of Hazardous Materials 286 (2015) 369–378

[17] M. Uibu, R. Kuusik, L. Andreas, K. Kirsimäe, The CO2 -binding by Ca–Mg–silicates in direct aqueous carbonation of oil shale ash and steel slag, Energy Procedia 4 (2011) 925–932. [18] E.-E. Chang, S.-Y. Pan, Y.-H. Chen, C.-S. Tan, P.-C. Chiang, Accelerated carbonation of steelmaking slags in a high-gravity rotating packed bed, J. Hazard. Mater. 227–228 (2012) 97–106. [19] R.M. Santos, D. Ling, A. Sarvaramini, M. Guo, J. Elsen, F. Larachi, et al., Stabilization of basic oxygen furnace slag by hot-stage carbonation treatment, Chem. Eng. J. 203 (2012) 239–250. [20] R.M. Santos, D. Franc¸ois, G. Mertens, J. Elsen, T. Van Gerven, Ultrasound-intensified mineral carbonation, Appl. Therm. Eng. 57 (2013) 154–163. [21] H. Shen, E. Forssberg, An overview of recovery of metals from slags, Waste Manage. 23 (2003) 933–949. [22] A.-M. Fällman, Leaching of chromium and barium from steel slag in laboratory and field tests – a solubility controlled process? Waste Manage. 20 (2000) 149–154. [23] M. Tossavainen, F. Engstrom, Q. Yang, N. Menad, M. Lidstrom Larsson, B. Bjorkman, Characteristics of steel slag under different cooling conditions, Waste Manage. 27 (2007) 1335–1344. [24] L. De Windt, P. Chaurand, J. Rose, Kinetics of steel slag leaching: batch tests and modeling, Waste Manage. 31 (2011) 225–235. [25] M. Loncnar, M. Zupanˇciˇc, P. Bukovec, A. Jakliˇc, The effect of water cooling on the leaching behaviour of EAF slag from stainless steel production, Mater. Tehnol. 43 (2009) 315–321. [26] J.F.P. Gomes, C.G. Pinto, Leaching of heavy metals from steelmaking slags, Rev. Metal. 42 (2006) 409–416. [27] P. Chaurand, J. Rose, V. Briois, L. Olivi, J.-L. Hazemann, O. Proux, et al., Environmental impacts of steel slag reused in road construction: a crystallographic and molecular (XANES) approach, J. Hazard. Mater. 139 (2007) 537–542. [28] D. Mombelli, C. Mapelli, S. Barella, A. Gruttadauria, G. Le Saout, E. Garcia-Diaz, The efficiency of quartz addition on electric arc furnace (EAF) carbon steel slag stability, J. Hazard. Mater. 279 (2014) 586–596. [29] D.C. Johnson, C.L. MacLeod, C.D. Hills, Acid neutralisation capacity of accelerated carbonated stainless steel slag, Environ. Technol. 24 (2003) 545–551. [30] W.J.J. Huijgen, R.N.J. Comans, Carbonation of steel slag for CO2 sequestration: leaching of products and reaction mechanisms, Environ. Sci. Technol. 40 (2006) 2790–2796. [31] Q. Chen, D.C. Johnson, L. Zhu, M. Yuan, C.D. Hills, Accelerated carbonation and leaching behavior of the slag from iron and steel making industry, J. Univ. Sci. Technol. Beijing 14 (2007) 297–301. [32] S.R. van der Laan, C.J.G. van Hoek, A. van Zomeren, R.N.J. Comans, J.B.A. Kobesen, P.G.J. Broersen, Chemical reduction of CO2 to carbon at ambient conditions during artificial weathering of converter steel slag, in: ACEME08, 2nd Int. Conf. Accel. Carbonation Environ. Mater. Eng., Rome 1–3 Oct. 2008, 2008, pp. 229–237. [33] S. Diener, L. Andreas, I. Herrmann, H. Ecke, A. Lagerkvist, Accelerated carbonation of steel slags in a landfill cover construction, Waste Manage. 30 (2010) 132–139. [34] A. van Zomeren, S.R. van der Laan, H.B.A. Kobesen, W.J.J. Huijgen, R.N.J. Comans, Changes in mineralogical and leaching properties of converter steel slag resulting from accelerated carbonation at low CO2 pressure, Waste Manage. 31 (2011) 2236–2244. [35] P. Suer, J.-E. Lindqvist, M. Arm, P. Frogner-Kockum, Reproducing ten years of road ageing-accelerated carbonation and leaching of EAF steel slag, Sci. Total Environ. 407 (2009) 5110–5118.

[36] R. Onori, A. Polettini, R. Pomi, Mechanical properties and leaching modeling of activated incinerator bottom ash in Portland cement blends, Waste Manage. 31 (2011) 298–310. [37] T. Astrup, J.J. Dijkstra, R.N.J. Comans, H.A. van der Sloot, T.H. Christensen, Geochemical modeling of leaching from MSWI air-pollution-control residues, Environ. Sci. Technol. 40 (2006) 3551–3557. [38] D.C. Johnson, C.L. MacLeod, P.J. Carey, C.D. Hills, Solidification of stainless steel slag by accelerated carbonation, Environ. Technol. 24 (2003) 671–678. [39] M. Bodor, R.M. Santos, L. Kriskova, J. Elsen, M. Vlad, T. Van Gerven, Susceptibility of mineral phases of steel slags towards carbonation: mineralogical, morphological and chemical assessment, Eur. J. Mineral. 25 (2013) 533–549. [40] R. Baciocchi, G. Costa, M. Di Gianfilippo, A. Polettini, R. Pomi, A. Stramazzo, Thin-film versus slurry-phase carbonation of steel slag: CO2 uptake and effects on mineralogy, J. Hazard. Mater. 283 (2015) 302–313. [41] M. Salman, Ö. Cizer, Y. Pontikes, R.M. Santos, R. Snellings, L. Vandewalle, et al., Effect of accelerated carbonation on AOD stainless steel slag for its valorisation as a CO2 -sequestering construction material, Chem. Eng. J. 246 (2014) 39–52. [42] R.M. Santos, J. Van Bouwel, E. Vandevelde, G. Mertens, J. Elsen, T. Van Gerven, Accelerated mineral carbonation of stainless steel slags for CO2 storage and waste valorization: effect of process parameters on geochemical properties, Int. J. Greenhouse Gas Control 17 (2013) 32–45. [43] J.A. Meima, R.N.J. Comans, Geochemical modeling of weathering reactions in municipal solid waste incinerator bottom ash, Environ. Sci. Technol. 31 (1997) 1269–1276. [44] J.A. Meima, R.N. Comans, The leaching of trace elements from municipal solid waste incinerator bottom ash at different stages of weathering, Appl. Geochem. 14 (1999) 159–171. [45] S. Arickx, T. Van Gerven, C. Vandecasteele, Accelerated carbonation for treatment of MSWI bottom ash, J. Hazard. Mater. 137 (2006) 235–243. [46] X. Li, M. Fernández Bertos, C.D. Hills, P.J. Carey, S. Simons, Accelerated carbonation of municipal solid waste incineration fly ashes, Waste Manage. 27 (2007) 1200–1206. [47] A. Antemir, C.D. Hills, P.J. Carey, M.-C. Magnié, A. Polettini, Investigation of 4-year-old stabilised/solidified and accelerated carbonated contaminated soil, J. Hazard. Mater. 181 (2010) 543–555. [48] E. Ruiz-Agudo, C.V. Putnis, C. Rodriguez-Navarro, A. Putnis, Mechanism of leached layer formation during chemical weathering of silicate minerals, Geology 40 (2012) 947–950. [49] W.J.J. Huijgen, G.-J. Witkamp, R.N.J. Comans, Mechanisms of aqueous wollastonite carbonation as a possible CO2 sequestration process, Chem. Eng. Sci. 61 (2006) 4242–4251. [50] S. Shaw, S. Clark, C.M. Henderson, Hydrothermal formation of the calcium silicate hydrates, tobermorite (Ca5 Si6 O16 (OH)2 ·4H2 O) and xonotlite (Ca6 Si6 O17 (OH)2 ): an in situ synchrotron study, Chem. Geol. 167 (2000) 129–140. [51] U.S. Department of Energy (DoE), Qualification of thermodynamic data for geochemical modeling of mineral-water interactions in dilute systems, document no. ANL-WIS-GS-000003, 2004. [52] S.L. Brantley, Kinetics of mineral dissolution, in: S. Brantley, J. Kubicki, A. White (Eds.), Kinet. Water-Rock Interact., Springer, New York, 2010, pp. 151–210. [53] E.J. Weissbart, J.D. Rimstidt, Wollastonite: incongruent dissolution and leached layer formation, Geochim. Cosmochim. Acta 64 (2000) 4007–4016.

Effects of thin-film accelerated carbonation on steel slag leaching.

This paper discusses the effects of accelerated carbonation on the leaching behaviour of two types of stainless steel slags (electric arc furnace and ...
3MB Sizes 6 Downloads 8 Views