Waste Management 43 (2015) 376–385
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Recycling of ladle slag in cement composites: Environmental impacts Vesna Zalar Serjun a, Ana Mladenovicˇ a, Breda Mirticˇ b, Anton Meden c, Janez Šcˇancˇar d, Radmila Milacˇicˇ d,⇑ a
Slovenian National Building and Civil Engineering Institute, Dimicˇeva 12, 1000 Ljubljana, Slovenia University of Ljubljana, Faculty of Natural Sciences and Engineering, Department of Geology, Aškercˇeva 12, 1000 Ljubljana, Slovenia c University of Ljubljana, Faculty of Chemistry and Chemical Technology, Vecˇna pot 113, 1000 Ljubljana, Slovenia d Department of Environmental Sciences, Jozˇef Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia b
a r t i c l e
i n f o
Article history: Received 21 February 2015 Accepted 4 May 2015 Available online 23 May 2015 Keywords: Cement composites Ladle slag Environmental impacts Hexavalent chromium Physico-mechanical characteristics
a b s t r a c t In the present work compact and ground cement composites in which 30% of cement by mass was replaced by ladle slag were investigated for their chemical and physico-mechanical properties. To evaluate long-term environmental impacts, leachability test based on diffusion, which combined both, diffusion and dissolution of contaminants, was performed in water and saline water. Total element concentrations and Cr(VI) were determined in leachates over a time period of 180 days. At the end of the experiment, the mineralogical composition and the physico-mechanical stability of cement composites was also assessed. The results revealed that Cr(III) and Cr(VI) were immobilized by the hydration products formed in the cement composites with the addition of ladle slag. Cr(VI) content originating from the cement was also appreciably reduced by Fe(II) from minerals present in the added ladle slag, which thus had significant positive environmental effects. Among metals, only Mo and Ba were leached in elevated concentrations, but solely in ground cement composites with the addition of ladle slag. Lower V concentrations were observed in leachates of ground than compact composite. It was demonstrated that the presence of ladle slag in cement composites can even contribute to improved mortar resistance. The investigated ladle slag can be successfully implemented in cement composites as supplementary cementitious material. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction The alloying of steel through various refining processes in the steelmaking industry has led to the generation of different types of secondary metallurgical slags, such as ladle slag. The European Union (EU) production of ladle slag has been estimated to be about 4 million tonnes per year (Murri et al., 2013). At present, some steel plants have declared this slag to be a waste material, whereas others consider it to be a by-product. Ladle slag has two main adverse environmental effects when landfilled: its disintegration into a fine grained material causes the spread of dust and the leaching of hexavalent chromium (Cr(VI)) and the release of some other toxic metals (Bignozzi et al., 2013; Iacobescu et al., 2011). In EU about 80% of ladle slag is currently landfilled (PREWARC, 2008). The toxicity, solubility, mobility of Cr, as well as its environmental fate, depends on its chemical forms (He and Traina, 2005; Šcˇancˇar and Milacˇicˇ, 2014). Cr(VI) is highly toxic, whereas trivalent Cr (Cr(III)) compounds are far less toxic and in low concentrations essential micronutrients (Unceta et al., 2010). Sodium and ⇑ Corresponding author. Tel.: +386 1 4773560; fax: +386 1 2519385. E-mail address: [email protected] (R. Milacˇicˇ). http://dx.doi.org/10.1016/j.wasman.2015.05.006 0956-053X/Ó 2015 Elsevier Ltd. All rights reserved.
potassium chromates are highly mobile within the entire pH range, while Cr(III) compounds are mobile mainly at acidic pH values. Cr in industrial by-products can be successfully immobilized when they are implemented in cementitious building composites (Laforest and Duchesne, 2005; Giergiczny and Król, 2008; Chen et al., 2009; Leisinger et al., 2012). At the same time such utilization significantly improves resource efficiency as an essential part of green building, which is strongly supported by Directive 2008/98/ES on waste (Official Journal of the EU, 2008) and Construction Products Regulation No. 305/2011 (Official Journal of the EU, 2011). During the hardening process, the incorporation of potentially hazardous elements into hydration products, such as calcium silicate hydrate (C–S–H) and calcium aluminate hydrates (C–A–H) phases, takes place (Shao et al., 2013; Qian et al., 2006). Especially C–A–H, with alumina ferric oxide tri-sulfate (AFt) and alumina ferric oxide mono-sulfate (AFm) phases have a high anion immobilization potential due to their structures and possibility of formation from a wide range of starting materials (Leisinger et al., 2012; Park et al., 2006; Renaudin et al., 2011). As an example, Friedel’s salt (AFm), which is a double-layered hydroxide and is formed by the hydration of cement in saline water, represents a good adsorbent for
V.Z. Serjun et al. / Waste Management 43 (2015) 376–385
the immobilization of Cr(VI) in cement (Dai et al., 2009). On the other hand, the implementation of ladle slag into cementitious materials could be beneficial, as certain part of ladle slags consists of pozzolanic/hydraulic phases and therefore, it can be used as a supplementary cementitious materials (SCM) (Setién et al., 2009; Papayianni and Anastasiou, 2012; Manso et al., 2011). Materials which contain different industrial wastes, should fulfil the requirements on hygiene, health and the environmental aspects, set by Construction Product Regulation (Official Journal of the EU, 2011). Therefore, it is necessary to assure the immobilization of contaminants, and to estimate the environmental hazard due to their possible release from the final cementitious products during their service life (Chaurand et al., 2007; Hartlen, 1996). The extent of leaching was estimated by applying different leaching tests (Kosson et al., 2002; Van der Sloot and Kosson, 2012). Several test methods have also been issued e.g. the European test method for leaching aggregates SIST EN 1744-3 (2002), CEN technical protocols (SIST-TS CEN/TS 15862, 2013; SIST-TS CEN/TS 14405: 2004), the SW-846 test methods of the U.S. guide for waste characterization (USEPA, 2012, 2013), the pH dependence test SIST prEN 14429 (2013), and the compliance tests for the leaching of granular waste materials and sludges (SIST EN 12457-4, 2004; SIST EN 12457-2, 2004). In order to estimate the long-term environmental impacts, leaching tests based on diffusion were proposed in the Dutch NEN 7345 (1995) standard and applied to concretes (Hohberg et al., 2000). Similar protocol was used also to assess the environmental impacts of asphalt mixes with the addition of electric arc furnace (EAF) dust (Vahcˇicˇ et al., 2008) or EAF steel slag (Milacˇicˇ et al., 2011) and cement composites with addition of EAF dust (Šturm et al., 2009). The objective of this study was to critically evaluate the potential use of ladle slag as a supplementary cementitious material in cement composites in which 30% of cement by mass was replaced by ladle slag. In order to investigate the long-term environmental impacts leachability test based on diffusion and dissolution of contaminants was applied.
2. Materials and methods 2.1. Apparatus The total concentrations of elements in the ladle slag and Portland cement, and the element concentrations in the leachates of the cement composites, were determined by inductively coupled plasma mass spectrometry (ICP-MS) on an Agilent (Tokyo, Japan) 7700 ICP-MS and by flame atomic absorption spectrometry (FAAS), using a Varian (Mulgrave, Victoria, Australia) SpectrAA 110 atomic absorption spectrometer. Ca, Mg and Fe were determined by FAAS in a nitrous-oxide–acetylene flame, while Na and K in an air–acetylene flame. In determination of alkaline and alkaline earth elements, CsCl was used as ionization buffer. Operating parameters for the determination of elements by ICP-MS are provided in Table 1 (Supplementary). To reduce polyatomic interferences, Helium Collision mode (He mode) or High Energy Collision mode (HECM) were applied. The content of Cr(VI) was determined on a HACH DR/2010 (Loveland, CO, USA) Portable Datalogging Spectrophotometer. A CEM Corporation (Matthews, NC, USA) CEM MARS 5 Microwave Acceleration Reaction System was used for digestion of ladle slag and Portland cement. A CILAS 920 laser (Cilas, Orléans, France) was used to characterize particle size distribution. The X-ray powder diffraction analyses were performed on a PANalytical X’Pert PRO MPD diffractometer (PANalytical B.V., Almelo, Netherlands). Pore size distribution analyses were performed by a mercury AutoPore IV 9500 porosimeter (Micromeritics, Norcross, GA) and by nitrogen sorption ASAP 2020
377
equipment (Micromeritics, Norcross, GA). The latter instrument was also used for measurements of specific surface area. Polished cross-sections of samples were investigated using a scanning electron microscope (SEM, Jeol 5500 LV, Japan), equipped with energy dispersive spectroscopy (EDS, Oxford instruments, Great Britain). 2.2. Reagents Merck (Darmstadt, Germany) suprapur acids and Milli-Q water (Direct-Q 5 Ultrapure water system, Millipore Water town, MA, USA) were used for the preparation of samples and standard solutions. Stock standard solutions of metals (1000 ± 2 mg L 1 in 5% HNO3) and aqueous Cr(VI) solution (1.000 ± 0.002 g L 1 CrO24 ) were also obtained from Merck. Sartorius (Goetingen, Germany). 0.45 lm cellulose nitrate membrane filters were used for filtration. The certified reference materials CRM 320R Trace Elements in River Sediment, Community Bureau of Reference (Geel, Belgium), SLRS-5, River water reference material purchased from the National Research Council (Ottawa, Ontario, Canada) and CRM 544, Cr(VI) in a lyophilised solution, (Community Bureau of Reference), were used to check the accuracy of the analytical procedures. 2.3. Determination of Cr(VI) and total element concentrations in the leachates Before analysis the leachates were filtered. Cr(VI) was determined by the 1,5 diphenylcarbazide spectrophotometric method according to ISO 11083 (1994), whereas the total concentrations of selected elements: Cr, Cu, Zn, Fe, Ni, Pb, Ca, Ba, Mn, Ti, Co, Cd, Mo, V, As, Sn, Sb, Se and Hg were determined by ICP-MS under optimal measurement conditions. 2.4. Ladle slag and cement Ladle slag generated at the steelworks Acciaierie Bertoli Safau (ABS), Italy was used. It consisted of a mixture of slag derived from a vacuum oxygen decarburization process and ladle furnace slag in the approximate weight ratio of 40:60. The Portland cement used was CEM I (52.5R) according to standard SIST EN 197-1 (2011). The content of Cr(VI) in cement was reduced by the addition of iron(II) sulfate, so that its concentration, when hydrated, was less than 2 mg Cr(VI) kg 1 (SIST EN 196-10, 2006). The total element concentrations in the ladle slag and cement were determined by ICP-MS and FAAS after microwave assisted digestion, using a mixture of nitric, hydrochloric and hydrofluoric acids, and the addition of H3BO3 for complexation of fluorides (Zuliani et al., 2013). 2.5. Characterization of ladle slag, Portland cement and cement composites Granulometric analysis of the ladle slag and Portland cement was accomplished by a laser analyser in inert medium (isopropanol). Specific surface area (BET) was measured by nitrogen adsorption. The SEM/EDS analyses were performed in low-vacuum mode with a chamber pressure of 12–13 Pa and an accelerating voltage of 20 kV. X-ray powder diffraction analyses of the ladle slag and the hydration products of the ground cement composites were performed by a diffractometer in Cu Ka1 configuration. Data were collected at room temperature in the 2h range from 8° to 70° in steps of 0.033° 2h. X’Pert HighScore Plus software (PANalytical B.V.) involving the PDF database as a source of reference data was
378
V.Z. Serjun et al. / Waste Management 43 (2015) 376–385
employed in order to assist in the identification of the crystalline phases of the investigated samples. Quantitative phase analyses were performed by the Rietveld method, using Topas Academic V4.1 software (Burker-AXS) and the ICSD database. In order to characterize the sample porosity, the pore size distribution of the compact cement composites was measured by mercury intrusion porosimetry (MIP) at a pressure capacity of 413 MPa and by nitrogen adsorption at 105 °C within an evacuation rate of 0.67 Pa/s. 2.6. Leachability of elements and Cr(VI) from the samples of ladle slag and Portland cement: shaking with water In order to estimate the leachability of elements from the samples of ladle slag and Portland cement, a compliance test for the leaching of granular waste materials and sludges (SIST EN 12457-2, 2004) was followed (using a liquid to solid ratio of 10:1). 90 g of the sample was shaken with 900 mL of water on a mechanical shaker (10 rpm) for 24 h. After that the samples were filtered through 0.45 lm filter, and the pH, conductivity and concentrations of selected elements and Cr(VI) were determined. 2.7. Leachability test based on diffusion In this set of experiments, cement composites and cement composites in which 30% of cement by mass had been replaced by ladle slag were used. The proportion of binder replaced by ladle slag was the same as reported by Rodriguez et al. (2009), Papayianni and Anastasiou (2010) and Manso et al. (2011), who investigated the physico-mechanical properties of such building composites. The aggregate in cement composites (mortars) was crushed dolomite (Stranice, Slovenia) with a gradation of 0–4 mm. The cement/aggregate/ water ratio, by dry mass in mortars was 1:3:0.5. Composites casts were 4 cm 4 cm 16 cm bars, with a volume of approximately 0.26 L. Preparation and conditioning of the test specimens was performed according to SIST EN 196-1 (2005) and SIST EN 1015-11 (2001). Compact and ground cement composites and cement composites with the addition of ladle slag were prepared. Specimens were ground to a particle size below 0.5 cm on the disc mill with grinding sets made from tungsten carbide. To assess the long-term environmental impacts, the leachability test based on diffusion was applied, which combined both, diffusion and dissolution of contaminants. Water and saline water (3.8% NaCl) were used as leaching solutions. The saline water simulated a marine environment, whereas the ground composites represented the decomposition of mortar over time. The experiments were carried out in duplicate in 3 L polyethylene beakers. The volume ratio of the tested composites to the added leaching solution was 1:5, as set by the NEN 7375 leachability test based on diffusion (NEN 7375, 2004). To each compact and/or ground composite 1.3 L of water or saline water was added, and the beakers were covered with plastic lids. Blank samples of water (pH 12) and saline water (pH 13) were also prepared and analyzed along with the samples. Prior to each sampling the solution was mixed with a glass rod. 10 mL of the sample was taken with a plastic syringe and filtered. After each sampling the same amount of leaching solution was added, and the beakers were re-covered. Samples were taken after 7, 14, 30, 60, 90, 120, 150 and 180 days. The times intervals were selected so, that it was possible to follow the extent of leaching of a particular contaminant, until the equilibrium between the solid and liquid phase was established and the concentration of the contaminant released into the leaching solution remained constant. In the leaching solutions, the pH, and the concentrations of selected elements and Cr(VI) were determined. In comparison to NEN 7375 (2004) test, there was no replenishing of eluate at specified times in test executed. In addition, the time interval was
much longer (180 days) than in NEN 7375 (2004) test (64 days). The leaching test applied enabled simulation of diffusion and dissolution of contaminants in the environment, when building material composed from cement and ladle slag is immersed into the stagnant environmental waters. At the end of the experiment the mechanical strength of the compact composites was also determined (SIST EN 1015-11, 2007; SIST EN 1015-11/A1, 2007). To remove the water from the pores of the cement composites, mortar samples were dried in a convective oven at 105 ± 1 °C until a constant mass (Chen and Wu, 2013). For the XRD analyses of the hydration products that were performed at the end of the experiment for the ground composites, the aggregate was carefully separated from binder. 2.8. Quality control of the analytical data The accuracy of determination of total element concentrations in ladle slag and Portland cement samples was checked by analysis of certified reference material CRM 320R. The quality control material SLRS-5 was used to check the accuracy of the element concentrations determined in the leachates, whereas certified reference material CRM 544 was analyzed to check the accuracy of Cr(VI) determination. The determined concentrations were in good correlation with the reported certified values (the agreement between the results was better than ±5%), confirming the accuracy of the applied analytical procedures (Table 2 of Supplementary material). 3. Results and discussion 3.1. Characterization of the ladle slag and Portland cement The particle size distribution and BET specific surface area of the ladle slag and Portland cement used in the tests is presented in Table 3 of the Supplementary material. The BET specific surface area was higher in the ladle slag than in the Portland cement. In comparison to Portland cement, ladle slag consisted of coarser grained material and had a wider particle size distribution. The most abundant phases in ladle slag identified by the XRD (Fig. 1) were calcium aluminates (tricalcium aluminate (PDF #38-1429), mayenite (PDF #9-0413)) and metallic mineral compounds (wuestite (PDF #6-0615), spinels (PDF #19-0629 and #34-0140), Fe alloy (PDF #6-0696), hematite (PDF #13-534)) followed by calcium silicates (larnite (PDF #33-0302), c dicalcium silicate (PDF #31-0297)), portlandite (PDF #4-0733), periclase (PDF #45-0946), gehlenite (PDF #35-0755), calcite (PDF #5-0586) and brown millerite (PDF #30-0226). 3.2. Total concentrations of selected elements in the ladle slag and Portland cement and their concentrations and concentrations of Cr(VI) in aqueous leachates The total concentrations of selected elements in ladle slag and in the Portland cement are presented in Table 1. The concentrations of elements in Portland cement presented in Table 1 are comparable to those found in the literature (Nourredine, 2011; Truc et al., 2000; Orellan et al., 2004) and are, in the case of Ca and Si, higher than those in the ladle slag. Since ladle slag is a by-product of the steel-making industry, it contains elevated concentrations of Al, Fe, Ni, Cr, Mn and Mo. To estimate the environmental hazard due to the possible release of elements from the ladle slag and the Portland cement, a compliance test for the leaching of granular waste materials and sludges was applied (SIST EN 12457-2, 2004), and the results compared to the maximum permitted values set by the Slovenian
379
V.Z. Serjun et al. / Waste Management 43 (2015) 376–385
Fig. 1. XRD pattern of the ladle slag.
Table 1 Total element concentrations in the ladle slag (LS), and in the Portland cement (PC) determined by ICP-MS and FAAS⁄. The results indicate the mean value obtained on two parallel samples. Element ⁄
Ca Si Al Mg⁄ Fe⁄ K⁄ Na⁄ Cu Zn Pb Ni Cr total Ba Mn Ti Co Cd Mo V As Sn Sb Se Hg
LS (mg kg
1
)
226,300 ± 4500 35,500 ± 700 68,400 ± 1370 19,700 ± 400 185,300 ± 3700 200 ± 4 610 ± 10 469 ± 9 295 ± 6 35.5 ± 0.7 985 ± 20 3058 ± 60 589 ± 12 3800 ± 76 1060 ± 21 23.0 ± 0.5 0.75 ± 0.02 564 ± 11 305 ± 6 25.0 ± 0.5 94 ± 2 18.5 ± 0.9
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Recycling of ladle slag in cement composites: Environmental impacts Vesna Zalar Serjun a, Ana Mladenovicˇ a, Breda Mirticˇ b, Anton Meden c, Janez Šcˇancˇar d, Radmila Milacˇicˇ d,⇑ a
Slovenian National Building and Civil Engineering Institute, Dimicˇeva 12, 1000 Ljubljana, Slovenia University of Ljubljana, Faculty of Natural Sciences and Engineering, Department of Geology, Aškercˇeva 12, 1000 Ljubljana, Slovenia c University of Ljubljana, Faculty of Chemistry and Chemical Technology, Vecˇna pot 113, 1000 Ljubljana, Slovenia d Department of Environmental Sciences, Jozˇef Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia b
a r t i c l e
i n f o
Article history: Received 21 February 2015 Accepted 4 May 2015 Available online 23 May 2015 Keywords: Cement composites Ladle slag Environmental impacts Hexavalent chromium Physico-mechanical characteristics
a b s t r a c t In the present work compact and ground cement composites in which 30% of cement by mass was replaced by ladle slag were investigated for their chemical and physico-mechanical properties. To evaluate long-term environmental impacts, leachability test based on diffusion, which combined both, diffusion and dissolution of contaminants, was performed in water and saline water. Total element concentrations and Cr(VI) were determined in leachates over a time period of 180 days. At the end of the experiment, the mineralogical composition and the physico-mechanical stability of cement composites was also assessed. The results revealed that Cr(III) and Cr(VI) were immobilized by the hydration products formed in the cement composites with the addition of ladle slag. Cr(VI) content originating from the cement was also appreciably reduced by Fe(II) from minerals present in the added ladle slag, which thus had significant positive environmental effects. Among metals, only Mo and Ba were leached in elevated concentrations, but solely in ground cement composites with the addition of ladle slag. Lower V concentrations were observed in leachates of ground than compact composite. It was demonstrated that the presence of ladle slag in cement composites can even contribute to improved mortar resistance. The investigated ladle slag can be successfully implemented in cement composites as supplementary cementitious material. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction The alloying of steel through various refining processes in the steelmaking industry has led to the generation of different types of secondary metallurgical slags, such as ladle slag. The European Union (EU) production of ladle slag has been estimated to be about 4 million tonnes per year (Murri et al., 2013). At present, some steel plants have declared this slag to be a waste material, whereas others consider it to be a by-product. Ladle slag has two main adverse environmental effects when landfilled: its disintegration into a fine grained material causes the spread of dust and the leaching of hexavalent chromium (Cr(VI)) and the release of some other toxic metals (Bignozzi et al., 2013; Iacobescu et al., 2011). In EU about 80% of ladle slag is currently landfilled (PREWARC, 2008). The toxicity, solubility, mobility of Cr, as well as its environmental fate, depends on its chemical forms (He and Traina, 2005; Šcˇancˇar and Milacˇicˇ, 2014). Cr(VI) is highly toxic, whereas trivalent Cr (Cr(III)) compounds are far less toxic and in low concentrations essential micronutrients (Unceta et al., 2010). Sodium and ⇑ Corresponding author. Tel.: +386 1 4773560; fax: +386 1 2519385. E-mail address: [email protected] (R. Milacˇicˇ). http://dx.doi.org/10.1016/j.wasman.2015.05.006 0956-053X/Ó 2015 Elsevier Ltd. All rights reserved.
potassium chromates are highly mobile within the entire pH range, while Cr(III) compounds are mobile mainly at acidic pH values. Cr in industrial by-products can be successfully immobilized when they are implemented in cementitious building composites (Laforest and Duchesne, 2005; Giergiczny and Król, 2008; Chen et al., 2009; Leisinger et al., 2012). At the same time such utilization significantly improves resource efficiency as an essential part of green building, which is strongly supported by Directive 2008/98/ES on waste (Official Journal of the EU, 2008) and Construction Products Regulation No. 305/2011 (Official Journal of the EU, 2011). During the hardening process, the incorporation of potentially hazardous elements into hydration products, such as calcium silicate hydrate (C–S–H) and calcium aluminate hydrates (C–A–H) phases, takes place (Shao et al., 2013; Qian et al., 2006). Especially C–A–H, with alumina ferric oxide tri-sulfate (AFt) and alumina ferric oxide mono-sulfate (AFm) phases have a high anion immobilization potential due to their structures and possibility of formation from a wide range of starting materials (Leisinger et al., 2012; Park et al., 2006; Renaudin et al., 2011). As an example, Friedel’s salt (AFm), which is a double-layered hydroxide and is formed by the hydration of cement in saline water, represents a good adsorbent for
V.Z. Serjun et al. / Waste Management 43 (2015) 376–385
the immobilization of Cr(VI) in cement (Dai et al., 2009). On the other hand, the implementation of ladle slag into cementitious materials could be beneficial, as certain part of ladle slags consists of pozzolanic/hydraulic phases and therefore, it can be used as a supplementary cementitious materials (SCM) (Setién et al., 2009; Papayianni and Anastasiou, 2012; Manso et al., 2011). Materials which contain different industrial wastes, should fulfil the requirements on hygiene, health and the environmental aspects, set by Construction Product Regulation (Official Journal of the EU, 2011). Therefore, it is necessary to assure the immobilization of contaminants, and to estimate the environmental hazard due to their possible release from the final cementitious products during their service life (Chaurand et al., 2007; Hartlen, 1996). The extent of leaching was estimated by applying different leaching tests (Kosson et al., 2002; Van der Sloot and Kosson, 2012). Several test methods have also been issued e.g. the European test method for leaching aggregates SIST EN 1744-3 (2002), CEN technical protocols (SIST-TS CEN/TS 15862, 2013; SIST-TS CEN/TS 14405: 2004), the SW-846 test methods of the U.S. guide for waste characterization (USEPA, 2012, 2013), the pH dependence test SIST prEN 14429 (2013), and the compliance tests for the leaching of granular waste materials and sludges (SIST EN 12457-4, 2004; SIST EN 12457-2, 2004). In order to estimate the long-term environmental impacts, leaching tests based on diffusion were proposed in the Dutch NEN 7345 (1995) standard and applied to concretes (Hohberg et al., 2000). Similar protocol was used also to assess the environmental impacts of asphalt mixes with the addition of electric arc furnace (EAF) dust (Vahcˇicˇ et al., 2008) or EAF steel slag (Milacˇicˇ et al., 2011) and cement composites with addition of EAF dust (Šturm et al., 2009). The objective of this study was to critically evaluate the potential use of ladle slag as a supplementary cementitious material in cement composites in which 30% of cement by mass was replaced by ladle slag. In order to investigate the long-term environmental impacts leachability test based on diffusion and dissolution of contaminants was applied.
2. Materials and methods 2.1. Apparatus The total concentrations of elements in the ladle slag and Portland cement, and the element concentrations in the leachates of the cement composites, were determined by inductively coupled plasma mass spectrometry (ICP-MS) on an Agilent (Tokyo, Japan) 7700 ICP-MS and by flame atomic absorption spectrometry (FAAS), using a Varian (Mulgrave, Victoria, Australia) SpectrAA 110 atomic absorption spectrometer. Ca, Mg and Fe were determined by FAAS in a nitrous-oxide–acetylene flame, while Na and K in an air–acetylene flame. In determination of alkaline and alkaline earth elements, CsCl was used as ionization buffer. Operating parameters for the determination of elements by ICP-MS are provided in Table 1 (Supplementary). To reduce polyatomic interferences, Helium Collision mode (He mode) or High Energy Collision mode (HECM) were applied. The content of Cr(VI) was determined on a HACH DR/2010 (Loveland, CO, USA) Portable Datalogging Spectrophotometer. A CEM Corporation (Matthews, NC, USA) CEM MARS 5 Microwave Acceleration Reaction System was used for digestion of ladle slag and Portland cement. A CILAS 920 laser (Cilas, Orléans, France) was used to characterize particle size distribution. The X-ray powder diffraction analyses were performed on a PANalytical X’Pert PRO MPD diffractometer (PANalytical B.V., Almelo, Netherlands). Pore size distribution analyses were performed by a mercury AutoPore IV 9500 porosimeter (Micromeritics, Norcross, GA) and by nitrogen sorption ASAP 2020
377
equipment (Micromeritics, Norcross, GA). The latter instrument was also used for measurements of specific surface area. Polished cross-sections of samples were investigated using a scanning electron microscope (SEM, Jeol 5500 LV, Japan), equipped with energy dispersive spectroscopy (EDS, Oxford instruments, Great Britain). 2.2. Reagents Merck (Darmstadt, Germany) suprapur acids and Milli-Q water (Direct-Q 5 Ultrapure water system, Millipore Water town, MA, USA) were used for the preparation of samples and standard solutions. Stock standard solutions of metals (1000 ± 2 mg L 1 in 5% HNO3) and aqueous Cr(VI) solution (1.000 ± 0.002 g L 1 CrO24 ) were also obtained from Merck. Sartorius (Goetingen, Germany). 0.45 lm cellulose nitrate membrane filters were used for filtration. The certified reference materials CRM 320R Trace Elements in River Sediment, Community Bureau of Reference (Geel, Belgium), SLRS-5, River water reference material purchased from the National Research Council (Ottawa, Ontario, Canada) and CRM 544, Cr(VI) in a lyophilised solution, (Community Bureau of Reference), were used to check the accuracy of the analytical procedures. 2.3. Determination of Cr(VI) and total element concentrations in the leachates Before analysis the leachates were filtered. Cr(VI) was determined by the 1,5 diphenylcarbazide spectrophotometric method according to ISO 11083 (1994), whereas the total concentrations of selected elements: Cr, Cu, Zn, Fe, Ni, Pb, Ca, Ba, Mn, Ti, Co, Cd, Mo, V, As, Sn, Sb, Se and Hg were determined by ICP-MS under optimal measurement conditions. 2.4. Ladle slag and cement Ladle slag generated at the steelworks Acciaierie Bertoli Safau (ABS), Italy was used. It consisted of a mixture of slag derived from a vacuum oxygen decarburization process and ladle furnace slag in the approximate weight ratio of 40:60. The Portland cement used was CEM I (52.5R) according to standard SIST EN 197-1 (2011). The content of Cr(VI) in cement was reduced by the addition of iron(II) sulfate, so that its concentration, when hydrated, was less than 2 mg Cr(VI) kg 1 (SIST EN 196-10, 2006). The total element concentrations in the ladle slag and cement were determined by ICP-MS and FAAS after microwave assisted digestion, using a mixture of nitric, hydrochloric and hydrofluoric acids, and the addition of H3BO3 for complexation of fluorides (Zuliani et al., 2013). 2.5. Characterization of ladle slag, Portland cement and cement composites Granulometric analysis of the ladle slag and Portland cement was accomplished by a laser analyser in inert medium (isopropanol). Specific surface area (BET) was measured by nitrogen adsorption. The SEM/EDS analyses were performed in low-vacuum mode with a chamber pressure of 12–13 Pa and an accelerating voltage of 20 kV. X-ray powder diffraction analyses of the ladle slag and the hydration products of the ground cement composites were performed by a diffractometer in Cu Ka1 configuration. Data were collected at room temperature in the 2h range from 8° to 70° in steps of 0.033° 2h. X’Pert HighScore Plus software (PANalytical B.V.) involving the PDF database as a source of reference data was
378
V.Z. Serjun et al. / Waste Management 43 (2015) 376–385
employed in order to assist in the identification of the crystalline phases of the investigated samples. Quantitative phase analyses were performed by the Rietveld method, using Topas Academic V4.1 software (Burker-AXS) and the ICSD database. In order to characterize the sample porosity, the pore size distribution of the compact cement composites was measured by mercury intrusion porosimetry (MIP) at a pressure capacity of 413 MPa and by nitrogen adsorption at 105 °C within an evacuation rate of 0.67 Pa/s. 2.6. Leachability of elements and Cr(VI) from the samples of ladle slag and Portland cement: shaking with water In order to estimate the leachability of elements from the samples of ladle slag and Portland cement, a compliance test for the leaching of granular waste materials and sludges (SIST EN 12457-2, 2004) was followed (using a liquid to solid ratio of 10:1). 90 g of the sample was shaken with 900 mL of water on a mechanical shaker (10 rpm) for 24 h. After that the samples were filtered through 0.45 lm filter, and the pH, conductivity and concentrations of selected elements and Cr(VI) were determined. 2.7. Leachability test based on diffusion In this set of experiments, cement composites and cement composites in which 30% of cement by mass had been replaced by ladle slag were used. The proportion of binder replaced by ladle slag was the same as reported by Rodriguez et al. (2009), Papayianni and Anastasiou (2010) and Manso et al. (2011), who investigated the physico-mechanical properties of such building composites. The aggregate in cement composites (mortars) was crushed dolomite (Stranice, Slovenia) with a gradation of 0–4 mm. The cement/aggregate/ water ratio, by dry mass in mortars was 1:3:0.5. Composites casts were 4 cm 4 cm 16 cm bars, with a volume of approximately 0.26 L. Preparation and conditioning of the test specimens was performed according to SIST EN 196-1 (2005) and SIST EN 1015-11 (2001). Compact and ground cement composites and cement composites with the addition of ladle slag were prepared. Specimens were ground to a particle size below 0.5 cm on the disc mill with grinding sets made from tungsten carbide. To assess the long-term environmental impacts, the leachability test based on diffusion was applied, which combined both, diffusion and dissolution of contaminants. Water and saline water (3.8% NaCl) were used as leaching solutions. The saline water simulated a marine environment, whereas the ground composites represented the decomposition of mortar over time. The experiments were carried out in duplicate in 3 L polyethylene beakers. The volume ratio of the tested composites to the added leaching solution was 1:5, as set by the NEN 7375 leachability test based on diffusion (NEN 7375, 2004). To each compact and/or ground composite 1.3 L of water or saline water was added, and the beakers were covered with plastic lids. Blank samples of water (pH 12) and saline water (pH 13) were also prepared and analyzed along with the samples. Prior to each sampling the solution was mixed with a glass rod. 10 mL of the sample was taken with a plastic syringe and filtered. After each sampling the same amount of leaching solution was added, and the beakers were re-covered. Samples were taken after 7, 14, 30, 60, 90, 120, 150 and 180 days. The times intervals were selected so, that it was possible to follow the extent of leaching of a particular contaminant, until the equilibrium between the solid and liquid phase was established and the concentration of the contaminant released into the leaching solution remained constant. In the leaching solutions, the pH, and the concentrations of selected elements and Cr(VI) were determined. In comparison to NEN 7375 (2004) test, there was no replenishing of eluate at specified times in test executed. In addition, the time interval was
much longer (180 days) than in NEN 7375 (2004) test (64 days). The leaching test applied enabled simulation of diffusion and dissolution of contaminants in the environment, when building material composed from cement and ladle slag is immersed into the stagnant environmental waters. At the end of the experiment the mechanical strength of the compact composites was also determined (SIST EN 1015-11, 2007; SIST EN 1015-11/A1, 2007). To remove the water from the pores of the cement composites, mortar samples were dried in a convective oven at 105 ± 1 °C until a constant mass (Chen and Wu, 2013). For the XRD analyses of the hydration products that were performed at the end of the experiment for the ground composites, the aggregate was carefully separated from binder. 2.8. Quality control of the analytical data The accuracy of determination of total element concentrations in ladle slag and Portland cement samples was checked by analysis of certified reference material CRM 320R. The quality control material SLRS-5 was used to check the accuracy of the element concentrations determined in the leachates, whereas certified reference material CRM 544 was analyzed to check the accuracy of Cr(VI) determination. The determined concentrations were in good correlation with the reported certified values (the agreement between the results was better than ±5%), confirming the accuracy of the applied analytical procedures (Table 2 of Supplementary material). 3. Results and discussion 3.1. Characterization of the ladle slag and Portland cement The particle size distribution and BET specific surface area of the ladle slag and Portland cement used in the tests is presented in Table 3 of the Supplementary material. The BET specific surface area was higher in the ladle slag than in the Portland cement. In comparison to Portland cement, ladle slag consisted of coarser grained material and had a wider particle size distribution. The most abundant phases in ladle slag identified by the XRD (Fig. 1) were calcium aluminates (tricalcium aluminate (PDF #38-1429), mayenite (PDF #9-0413)) and metallic mineral compounds (wuestite (PDF #6-0615), spinels (PDF #19-0629 and #34-0140), Fe alloy (PDF #6-0696), hematite (PDF #13-534)) followed by calcium silicates (larnite (PDF #33-0302), c dicalcium silicate (PDF #31-0297)), portlandite (PDF #4-0733), periclase (PDF #45-0946), gehlenite (PDF #35-0755), calcite (PDF #5-0586) and brown millerite (PDF #30-0226). 3.2. Total concentrations of selected elements in the ladle slag and Portland cement and their concentrations and concentrations of Cr(VI) in aqueous leachates The total concentrations of selected elements in ladle slag and in the Portland cement are presented in Table 1. The concentrations of elements in Portland cement presented in Table 1 are comparable to those found in the literature (Nourredine, 2011; Truc et al., 2000; Orellan et al., 2004) and are, in the case of Ca and Si, higher than those in the ladle slag. Since ladle slag is a by-product of the steel-making industry, it contains elevated concentrations of Al, Fe, Ni, Cr, Mn and Mo. To estimate the environmental hazard due to the possible release of elements from the ladle slag and the Portland cement, a compliance test for the leaching of granular waste materials and sludges was applied (SIST EN 12457-2, 2004), and the results compared to the maximum permitted values set by the Slovenian
379
V.Z. Serjun et al. / Waste Management 43 (2015) 376–385
Fig. 1. XRD pattern of the ladle slag.
Table 1 Total element concentrations in the ladle slag (LS), and in the Portland cement (PC) determined by ICP-MS and FAAS⁄. The results indicate the mean value obtained on two parallel samples. Element ⁄
Ca Si Al Mg⁄ Fe⁄ K⁄ Na⁄ Cu Zn Pb Ni Cr total Ba Mn Ti Co Cd Mo V As Sn Sb Se Hg
LS (mg kg
1
)
226,300 ± 4500 35,500 ± 700 68,400 ± 1370 19,700 ± 400 185,300 ± 3700 200 ± 4 610 ± 10 469 ± 9 295 ± 6 35.5 ± 0.7 985 ± 20 3058 ± 60 589 ± 12 3800 ± 76 1060 ± 21 23.0 ± 0.5 0.75 ± 0.02 564 ± 11 305 ± 6 25.0 ± 0.5 94 ± 2 18.5 ± 0.9
