Steel slags are side products of iron and steel industry and are suitable for a variety of applications. In the present paper, the assessment of the hazard associated with steel slags used as harbor piers backfilling material was performed. The analyses were conducted on 64 samples (soil and groundwater) collected in 17 sites located in one of the piers of Taranto city (South Italy), a highly industrialized area, where steel slags have been used in the past as backfill material. The obtained data were compared to the national threshold levels and International Indexes (Igeo) for assessing anthropogenic impacts. The first comparative analysis revealed that only vanadium in soil and aluminum in groundwater, associated with high pH values in both matrixes (soil and groundwater) resulted of high concern. Nevertheless, Igeo values and the coupling of comparative analysis with multivariate techniques Principal component analysis and Hierarchical Cluster Analyses revealed that also Sb, Pb, Cr and Zn concentration have an extremely high hazard degree. All analyses confirm the impact of steel slags on contaminant leaching processes due to the strong alkaline nature of their leachates and the presence of strong relationship between metal mobility and other physical-chemical parameters related to the peculiar environmental context. These findings suggest that the use of such materials, albeit economically recommended, should be adopted under severe monitoring of all parameters that could promote the mobility of metals and other contaminants.
Slags are industrial wastes generated during the steelmaking process. Different steelmaking processes generate different types of slags characterized by distinct chemical, mineralogical and morphological properties. The amount of steel slags (SS) produced worldwide annually reach remarkable figures in the order of hundreds of millions of tonnes. Therefore, possible re-uses of these industrial wastes are highly recommended for both lightening the need for their disposal and to reduce the use of natural resources.
In the past years, significant amounts of SS have been used as construction material for civil and fill applications. Most of the utilized SS were granulated blast furnace slags, which are in high demand in the cement sector. Other applications of SS as aggregate material are for embankments construction, highway shoulders, hot mix asphalt pavement, quays and harbor piers, etc. (Rondi et al. 2016). In addition, environmental uses of steel slags have been increased with recent studies on their beneficial effects. Removal of phosphorus, nitrogen, or trace elements from solution and controlling unwanted industrial emissions are documented by Barca et al. (2012). Other studies have reported the effectiveness of using SS for removing phosphorous from agricultural runoff (Bowden et al., 2009). Use of ferrous slags as acid-neutralizing agents is documented by Gahan et al. (2009), that applied them in the treatment of acid-mine drainage resulting from the coal and base-metal operations. Lopez Gomez et al. (1999) reported on the preparation of fertilizers from SS and their influence on soil and grass composition and consequent economic benefits, while Anderson et al. (1991) examined the advantages in using calcium silicate slags as leaf nutrients in sugarcane soils.
Nevertheless, besides the documented opportunities of SS re-uses, recent studies have recognized the negative effects exerted by steel slags on groundwater and surface water as well as in soils (Roadcap et al. 2006; Baciocchi et al. 2015,; Gomez Nubla 2018 a,b). The main drawbacks of their use derive by their high potentiality in increasing the release of hazardous constituents from slags and/or from materials containing such industrial wastes. High swelling potential and high alkalinity revealed in slag leachates are well documented. Indeed, steel slags are mainly composed of calcium, silicon and iron oxides, but they contain also trace amounts of toxic elements, such as chromium, lead, cooper, nickel, zinc and vanadium, which can be released, entering into soil and water bodies under the leaching action of rainwater, that limits slag valorization (Wang et al., 2012). Moreover, chemical-physical parameters such as pH, electrical conductivity, redox potential and dissolved oxygen are demonstrated to impact the speciation of metal ions in the effluent and their mobility (Wang et al., 2012). When disposed in peculiar areas such as harbor piers under a salt-wedge regime, the presence of other dissolved ions (chlorine, borate, sulphate, phosphate, etc.) as well as peculiar hydrodynamic condition occurring in the seawater, can enhance the mobility and transport of contaminants (Alfarrah and Walraevens, 2018, Mali et al. 2018).
Defining technically sound civil or environmental application require deep knowledge of the chemical, mineralogical, and morphological properties of SS and the processes that generate them. Knowledge on the properties of SS is also important for predicting and preventing any potential environmental impact occurring by their long-term disposal, especially within some critical areas, such as port quays or backfilling material close to marine coasts characterized by salt-wedge regime. The recent enforcement of more stringent environmental regulations has also introduced the necessity of reliable evaluation of the negative impact deriving by use of SS wastes. Specific leaching tests on waste materials are compulsory before their reuse in environmentally related applications (REG. (EC) 2017/997, REG (EC)1357/2014, 2008/98/EC, Italian D.M.5/2/98, DLgs.152/2006)
Clear identification of the effects exerted by them on different matrixes (water-soil-biota) and consequent potential deleterious impact on living organisms is currently an important challenge.
The present research had a twofold aim: i) evaluating leaching processes of trace elements in the groundwater aquifer from the leachates of slag-filled soils; ii) discussing potential environmental issues related to weathering of slags in highly compromised environmental setting. For this purpose, the Port of Taranto, that falls within a wide catchment and was declared “at high risk of environmental crisis” by the national government (Italian Law n. 349/86), was selected as case study. The area represents one of the most complex industrial sites in Europe, located very close to an urban center, affected by different anthropogenic pressures, such as the biggest Steel Factory in Europe, the largest Italian Navy shipyard, an oil refinery (ENI), and shipbuilding activities. Slags waste of the nearby Steel Factory have been used in the past as backfilling material for embankments and piers of the port area and the analyses conducted in one of the quays subjected to Requalification Project of the port of Taranto are used in the present work for evaluation of the arising effects in the surrounding soil and ground water.
Materials and methods
Data collection - sampling strategy
The monitoring program for the study area was carried out by ARPA Puglia (Rel.Tech. 2016) in the framework of Site-Specific Risk Assessment performed according to Legislative Decree (D.Lgs.) No. 152/2006, for the Project of "Redevelopment of Polysectorial Pier - Modification of the Terminal Bulk of Port of Taranto" (Decree n.38/15 of 20.04.2015). The monitoring area has an extension of approximately 62,000 square meters falling in the Taranto Port. A total of 17 sites were sampled through core drilling methodology, up to a maximum depth of 3.0 m. Six of the 17 sites were equipped with piezometers (sampling wells) up to a maximum depth of 3.5 m in order to collect water samples from the shallow groundwater aquifer (Figure S1).
Thiessen Polygon Method was applied for selecting sampling sites for surface, subsurface soil and groundwater. For soil samples collected up to 3 m, three buckets were considered: surface and subsurface layers, collected up to 10 and 40 cm respectively, and bottom layer, F, reaching 3 m depth. For each site, three replicates were considered. A total of 64 soil samples and 6 groundwater samples were thus analyzed. Adequate clean plastic jars with Teflon coated lids were used for storage and transport of the samples to the laboratory. Once in laboratory, the collected soil samples were freeze-dried, gently ground in an agate mortar trying not to alter the grain size features, then passed through a 0.5 mm mesh sieves to remove debris and pebbles, and finally stored at −20 °C prior to analysis. Figure 1a shows the whole catchment subjected to the Monitoring Program and Figure 1b zooms on the study area considered for the present work. In this area, that belongs to one of the Taranto harbor piers built with heterogenous and steel slag materials, a drainage discharge channel (Channel n. 2, Figure 1a) of the Steel Factory is also located.
Given the small depth of maximum investigation (up to 3.0 m), the samples collected in the survey area intercepted only the surface lithostratigraphic units belonging to the anthropogenic backfilled material and sandy heterogeneous deposits, mainly mud with polygenic clasts (gravel and pebbles) varying in color from brown to greenish with different degrees of consistency (Figure S2). The thickness of the backfilling material vary from 2.8 m to 4.0 m. The latter has a chaotic structure and is made of mixtures of steel slags wastes and residues. The natural soil appears at the bottom end of this anthropogenic layer and belongs to the formation of the basic Sub-Apennine Clays. In the top part, in contact with anthropogenic layer, natural soils result altered, greenish in color and sometimes possess an evident sandy fraction. Considering the grain size distribution, despite a percentage of the silt and clay sum in the range of 4% ÷17% for unsaturated soil, the most predominant granulometric fraction is sand.
Analyses were carried out according to the standard protocols established by National Agency (ISPRA). Trace metal concentrations were measured by inductively coupled plasma mass spectrometry (ICP/MS X Series Thermo Fisher Scientific) after sample mineralization by total acid digestion (HCl, HNO3 and HF). The < 63 μm fraction, dried at 105 °C, was used for the determination of metals in order to reduce the grain size effect. The detection limits (LODs) were calculated from 3 replicates of procedural blanks. The LODs estimated were in μg/kg units and were equal to 1 ppb for all metals. Certified Reference Materials were used to control the analysis quality: the agreement between the analytical results for the certified and measured values was satisfactory, with recoveries ranging from 80% to 100% for all metals.
Assessment of the hazard distribution
In order to facilitate the interpretation of the obtained results, multivariate analyses with both supervised and unsupervised methods, were performed. PCA (Principal Component Analysis) and HCA (Hierarchical Cluster Analyses) are two important techniques commonly used in environmental studies (et al. 2017b). PCA was performed on normalized variables (mean and centered) in order to extract significant principal components (PCs) and further reduce the contribution of variables with minor significance. Coupling approach of PCA with HCA, using SIMCA software, was applied for identifying common pollution sources, patterns of sites with similar pollution character and recognizing the most discriminant contaminants within site clusters. The “benefits” of graphical representation allowed to present the main information by compressing the original data set, without great loss of information, providing thus, in a simple way, a comprehensive environmental status of the study area.
Results and discussion
Soil chemical contamination and statistical analyses
The basic statistical parameters and the measured concentrations of pollutants are shown in Table 1. Comparing the concentration of contaminants to the National Standard Values revealed that only vanadium concentration exceeded the national limit for Industrial Areas (D. Lgs.152/2006). Nevertheless, the very high pH values are indicative for abnormally strongly alkaline soils. In addition, all trace metal concentrations exceed in the same way both the Average Shale Values (ASVs), considered as Background Levels for metal content in natural soils, and the Regional Geochemical Background (RGB) values reported in a previous study (Mali et al., 2015). The respective hazard classes calculated on the base of Igeo values (Table 2) indicate that the pollution degree, besides vanadium, range from “middle” to “strongly polluted”, also for Sb, Pb, Cr, Cu and Zn (Table 1), reaching in two cases hazard classes of extremely high concern (Igeo > 6, Table 1). In addition, comparing the concentration of contaminants to Effect Range Medium (ERM) values (Table S1) established from Long et al, (1995) as metal concentration thresholds above which toxic effects on biota could be expected, all analyzed metals can exert negative effects, especially considering the peculiar environment setting [a harbour area subjected to hydrodynamic pressures (Malcangio et al. 2017; Mali et al. 2018)].
In order to identify common contaminant sources and site clusters with similar pollution trend, Principal Component Analyses (PCA) was performed. A model with four Principal Components (PCs) is considered, covering 69% of the total co-variance. Figure 2 shows the projection of Score plot (Fig. 3a) and Loading plot (Fig. 3b) of the first two components (PC1//PC2). Figure 2a shows that samples are grouped in two main clusters: the highly deep samples (F) in blue color are displayed in the part of PC1/PC2 space characterized by positive value of PC1 and negative value of PC2. On the contrary, the remaining samples (surficial samples (red and yellow), and sub-surficial samples (green color) are displayed on the opposite space. Analysis of the Loading Plots (Figure 2b) reveals two clusters of contaminants, differently controlled by pH values. The variability of Crtot, V, Be and Pb resulted proportionally correlated with pH values in all deep layers (F). Increase of pH corresponds to increase of Cr, V, Be and Pb concentrations. While the concentration of the other metals (Sn, Sb, As, Ni, Co and Cu) which reach high values in the surface layers, seems to be inversely correlated with pH values and proportionally with Total Hydrocarbons (THC). The latter index (THC) was selected in this study as representative of organic pollutants. The values of pH increases with the depth reaching very alkaline values (12.5) in highly deep layer F, while it remains in the range from pH 9 to 10 in the surficial and intermedium layers (such values are still over the normal soil pH ranges) (Figure S3). Only in one case (S14, intermedium layer) the pH is acid (5.8). It is supposed that the steel slags used for the embankment of the study area are constituted by alkaline oxides and leaching phenomena from the slags-filled material to the surrounding environment are expected to occur.
Groundwater chemical contamination and statistical analyses
Perusal of the groundwater data (Table S2) indicates that high values of pH in soils are reflected in the high pH values detected in shallow groundwater aquifers. Indeed, water pH values in the six sites investigated reach very alkaline levels (up to 12.5) with an average of pH = 11 which is abnormal for groundwaters. In addition, high Al concentration values, (exceeding in some cases five times the National Laws levels (200 mg l−1 - Tab, 2 Annex, 5 Part IV D. Lgs, 152/06) were detected, as in the case of the sampling well countersigned with S6, where the dissolved Al concentration reached 1720 mg l−1.
It should be underlined that data of Al concentration in the whole investigated groundwater area (Figure 1a) were consulted. High levels of dissolved Al were found in the whole catchment. It seems that the shallow groundwater aquifer, under the slag-filled area, is seriously impacted by alkaline environment that enhances the mobility of all elements that give amphoteric oxides, including Al that is not considered as a soil pollutant and thus not subjected to monitoring analysis in soils. It is reasonable to suppose that the increase of alkalinity of the groundwater and soils in the investigated area is due to the presence of alkaline steel slags that, under long-term weathering processes, undergo the dissolution of the primary mineral phases and consequent give rise to a release of the associated contaminants. This finding complies to literature (Gomez et al. 2018a,b). Moreover, other physical-chemical parameters such as electrical conductivity (σ), redox potential (Eh) and dissolved oxygen are known to impact the speciation of metal ions in the effluent and their transfer and transformation tendency (Wang et al., 2012). In fact, the groundwater samples were characterised also by high levels of Electrical Conductivity (up to 19000 μS/cm), principally related to concentration of Cl– (706 ÷ 25453 mg l−1), SO42– (12.5 ÷ 708 mg l−1), HCO3– (25 ÷ 1507 mg l−1), being these anions presumably brought by seawater intrusion (Leshmi and Mophin, 2017, Alfarrah and Walraevens, 2018). These findings are confirmed by Hierarchical Cluster Analysis (HCA) performed on groundwater dataset (Figure 3). Three clusters of variables are distinguishable in HCA. As expected, the Al values resulted related to pH and conductivity (σ), that influence also the variability of other contaminants, such as As, Cd, Sb and B. Although the concentration of these elements did not exceed the National Limits, it is reasonable to suppose that at long-term period in alkaline condition, contaminant release phenomenona from soils to groundwater could occur, causing in future concern for the quality status of the aquifer. The presence of organic pollutants such as toluene, benzene and tetrachloroethane, associated in the cluster with Cr, Zn, Ni and Se, seems to suggest that contamination sources other than slags may be held responsible for the quality of the groundwater. Broader studies are thus needed to assess the right weight to slags in the hazard pollution of the investigated site.
The usage of steel slag aggregates in industrial applications might constitute an useful alternative for their sustainable reuse and possibly divert large amount of these waste materials from landfills. Nevertheless, this study demonstrates that use of steel slags, especially in areas that are highly compromised by natural and anthropogenic factors, could exert serious effect especially in the shallow groundwater aquifer due to the alkaline leachate of slag-filled material under long-term weathering processes. Site-specific hazard assessment carried out in this study evidenced that slag-filled soils are characterized by high pH values that, in turn, causes alkaline conditions in the underlying groundwater with consequent dissolution of aluminum, with potential harm to aquatic life and human health. Multivariate analyses permitted to identify also other co-contaminants (Sb, Pb, Cd) that can enhance the ecological risk of the aquifer, especially in the presence of uncharacteristic values of pH, conductivity and redox potential. Therefore, the use of steel slags, albeit economically recommended, should be adopted under wise monitoring of long-term period. The chemical compositions of steel slags are very complex and depends on different factors, such as source of raw materials, furnace processing conditions and refining techniques. Therefore, the physical-chemical characteristics of steel slags can vary greatly, and the study of their effects should be performed on a case by case basis to meet the requirement for risk analysis assessment.
The authors thanks ISPRA and the SOGESID S.p.A. for providing all supporting data.