Stabilization of soils containing sulfates by using alternative hydraulic binders

Highlights

• Treatment of soils with OPC resulted in a high release of chromium into solution.

• Large volume expansions (>5%) were measured in soils treated with OPC.

• Soil treated with GGBS resulted in a high sulfate immobilization (>80%).

• Treatment of soils with GGBS reduced the swelling potential of the soil.

Abstract

Excavation operations during construction produce several tons of soil, which frequently contain high concentrations of sulfates (>0.5 wt%). In accordance with the requirements of the French decree for waste classification for disposal, a soil containing sulfates is classified as “inert and non-hazardous waste” if the leachable sulfate concentration is lower than a mass fraction of 0.1%. To prevent sulfate leaching from excavated soils, solutions for immobilizing sulfates are needed and the associated stabilization mechanisms must be understood. On the other hand, the reuse of soils containing sulfates for civil engineering purposes can lead to significant risks after their treatment with ordinary cementitious binders because of chemical reactions involving sulfates. These reactions can promote the formation of massive ettringite crystals resulting in expansion, cracking and eventually catastrophic damage of materials or structures. For these reasons, the stabilization of soils containing sulfates by adding alternative hydraulic binders is studied in this paper. Several binders were used to treat a sulfate-spiked soil. It was observed that treatment with cementitious binders having high C₃A content led to volume expansions greater than 5%, while treatments with binders containing a high fraction of ground granulated blast furnace slag (GGBS) showed volume expansions of less than 5% and about 89% of sulfates were immobilized in the solid matrices. These preliminary results suggest that GGBS binders are effective for the treatment of soils containing sulfates. Moreover, numerical calculations using PHREEQC were compared with experimental results to improve the understanding of sulfate immobilization mechanisms.

Introduction

In France, excavated soils are considered as waste and they are classified in different categories depending on their concentration of leachable characteristic constituents. In accordance with the French decree on waste classification for disposal (Legifrance, 2014), a soil containing sulfates is classified as “inert and non-hazardous waste” if the leachable sulfate concentration is lower than a mass fraction of 0.1%. To decrease sulfate leaching from excavated soils, solutions for immobilizing sulfates are needed and the associated stabilization mechanisms must be understood. On the other hand, the reuse of soils containing sulfates for civil engineering purposes can be considered.

In civil engineering, the mechanical properties of soil are improved by adding lime or cement. However, it has been reported that lime treatment of a soil containing sulfates represents a potential risk to the stability and durability of the material application because of expansive reactions induced in the treated soil. Several studies have highlighted that sulfate-rich soils treated with either ordinary Portland cement or lime lead to expansion of the material (Cabane, 2004; Celik and Nalbantoglu, 2013; Colas, 2012; Harris et al., 2006; Hunter, 1988; Kolani et al., 2012; Puppala et al., 2003; Talluri et al., 2013; Wang et al., 2003).

Moreover, it has been shown that soils containing clays with a mass fraction of at least 10% and 1% of sulfates produced significant swelling after treatment with lime (Dermatas, 1995; Hunter, 1988). In a high pH system (pH > 12), an expansive mineral known as hydrated calcium sulfoaluminate or ettringite ($3\mathrm{CaO}\cdot {\mathrm{Al}}_2{\mathrm{O}}_3\cdot 3{\mathrm{CaSO}}_4\cdot 32{\mathrm{H}}_2\mathrm{O}$) can precipitate when sulfate, alumina and calcium ions, and water are available to react. In treated soils, alumina ions could come from either the cement used in the treatment or from the dissolution of clays in the soil because of the increase of pH after the lime-treatment (Celik and Nalbantoglu, 2013; Dermatas, 1995; Hunter, 1988; Puppala et al., 2004).

It has been also reported that the swelling potential of sulfate-rich soils is decreased when they are treated with binders other than lime (Celik and Nalbantoglu, 2013; Harris et al., 2006; Kota et al., 1996; Mahedi et al., 2018; Puppala et al., 2004, 2003; Talluri et al., 2013; Wang et al., 2003). Wang et al. (2003) investigated the stabilization of sulfate-rich soils by partially replacing ordinary Portland cement (OPC) by alternative materials such as ground granulated blast furnace slag (GGBS), class C fly ash (CFA) and amorphous silica (AS). The authors showed that the replacement of OPC by GGBS decreased the amount of expansion in the soil and no expansion was detected when CFA and AS were used. Similarly, Puppala et al. (2003) studied the effectiveness of the treatment of sulfate-rich soils (2000 mg/kg to 5000 mg/kg of dry mass) with sulfate-resisting cement of types I/II and V. By adding 5% of such cements, they observed a decrease in the free swelling from 30% to less than 5%.

In the literature, preliminary work in this field focused primarily on the evaluation of the swelling phenomena in sulfate-contaminated soils. However, not much is known about the decrease in leachable sulfate concentration after treatment and the sulfate stabilization mechanisms. This work aims to (i) compare the capacity of several alternative binders to immobilize sulfates in a sulfate-spiked soil in natural pH conditions (pH 7) and (ii) understand the sulfate immobilization mechanisms by using a single step batch method. The stabilization technique was used to decrease the sulfate leaching from treated soils and the swelling phenomenon was studied for all the treatments. Stabilization groups together the chemical processes permitting the decrease in the potential risk of a hazardous waste due to the decrease in the toxicity and the water solubility of pollutants (Chen et al., 2009; Peysson, 2005).

In this study, four different binders were chosen to treat a sulfate-spiked soil: one Ordinary Portland Cement (OPC), one alternative clinker mainly composed of ye'elimite C₄A₃$\bar{\mathrm{S}}$ and belite C₂S, one GGBS CEM III/C cement and one experimental binder composed of 90% GGBS and 10% OPC. Leachable sulfate concentrations were determined by using ion chromatography by analyzing eluates extracted from leaching tests carried out in accordance with the European standard NF EN 12457-2, which is required by French law (Legifrance, 2014). Leachable heavy metals concentrations were also determined by using Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES). Swelling of untreated and binder-treated soils was evaluated by performing volume expansion tests on compacted specimens. Additionally, mechanical properties were evaluated by determining the indirect tensile strength. Finally, sulfate stabilization mechanisms were studied using Scanning Electron Microscopy with Energy Dispersive Spectroscopy (SEM/EDS) and X-ray diffraction (XRD) by analyzing mineral phases that could contain sulfates. In order to better understand the underlying sulfate stabilization mechanisms, experimental results were compared with numerical calculations obtained from a geochemical model using the PHREEQC code (Parkhurst and Appelo, 1999).