Valorization of deep soil mixing residue in cement-based materials

Abstract

This study examined the potential of recycling deep soil mixing residue (DSMR), a cement-soil waste, as a partial Portland cement substitute. The chemical and mineral compositions of DSMR before and after calcination were determined. For the calcined DSMR blended cement pastes, the compressive strength and phase assemblage (characterized using TGA and XRD) were evaluated. Results revealed pozzolanic and hydraulic properties of the calcined DSMR, due to the formation of reactive calcium-rich glassy phases and hydraulic C₂S. The calcined DSMR can contribute to the strength development of cement because of the pozzolanic reactions and the formation of additional hydrates. With up to 20% calcined DSMR, negligible side-effect on the strength of the blended cement paste was noticed. The compressive strength development of the blended cement pastes was linearly correlated to the reactive phase content in the calcined DSMR. This work demonstrated the potential of using DSMR to develop greener composite cement.

Introduction

Deep soil mixing (DSM), a ground improvement technique, has been widely utilized to treat problematic or soft soil ground over the last decades (De Silva et al., 2001; Güllü et al., 2017; Hasheminezhad and Bahadori, 2019; Jin-Jian et al., 2013; Madhyannapu and Puppala, 2014). DSM is carried out usinga machine equipped with mixing blades, as shown in Fig. 1(a), to break up in-situ soils and mechanically mix them with a cementitious binder, resulting in the formation of a cement-soil column beneath the ground surface. Cement is the most commonly used binder in DSM, often with a binder to soil ratio of up to 0.3 (Arulrajah et al., 2018; Ni and Cheng, 2014). The chemical reaction between the cement and the soil significantly enhances the characteristics of the weak soils, especially in terms of the load-bearing capacity, shear strength and modulus, to meet engineering requirements.

A major environmental challenge of the DSM technique is the generation of a large quantity of solid wastes. During the construction process, the cementitious binder is mixed with the in-situ soil to create a soil-cement mixture beneath the ground. During this procedure, a large quantity of excessive soil-cement mixtures tends to return to the ground surface and become the deep soil mixing residue (DSMR), as shown in Fig. 1(b). It is estimated that the amount of DSMR accounts for around 30% by the volume of the produced DSM column (Denies et al., 2012). The total quantity of DSMR generated in Flanders, Belgium, for instance, is more than 460,000 tons per year (Joseph et al., 2016). Currently, the generated DSMR is usually stacked next to the construction sites (see Fig. 1(c)) and later dumped in suburban landfills. The current dumping of DSMR results in a high economic cost and causes significant environmental issues, such as occupying valuable land resources and polluting surrounding soils. Therefore, exploring more efficient methods for the disposal of DSMR is of great importance to a circular construction economy.

The partial replacement of Portland cement by certain volumes of supplementary cementitious materials (SCMs), exhibiting hydraulic and/or pozzolanic or filler properties, has been commonly utilized to reduce the demand for cement clinker and thereby to decrease CO₂ emissions from cement production. Moreover, SCMs are frequently employed in concrete to improve its long-term mechanical performance and durability. However, in recent years, conventional SCMs, such as ground granulated blast furnace slag (GGBS) and coal fly ash, are subjected to a significant shortage in many countries and regions (Poowancum and Horpibulsuk, 2015). Moreover, the supply of those materials has kept decreasing in recent years and is expected to be reduced considerably in 2025 due to the rapid decrease of blast furnaces and coal-fired power plants (Alberici et al., 2017). As a result, it is necessary to explore novel and sustainable SCMs.

Calcined DSMR could be a promising candidate as an alternative SCM. This is because DSMR mainly consists of clay minerals, quartz, and hydrated cement. Several research works have documented the utilization of hydrated cement wastes (Bogas et al., 2020; Serpell and Lopez, 2013; Serpell and Zunino, 2017; Yu et al., 2013) and excavated soils (or clays) (Dang et al., 2013; Du and Pang, 2018; Hollanders et al., 2016; Yanguatin et al., 2019; Zhao and Khoshnazar, 2020; Zhou et al., 2017) as SCMs, where calcination was used to enhance their reactivity. For hydrated cement, thermal treatment at a high temperature (600–800 °C) for several hours confers its rehydration capacity, due to the formation of calcium oxide (CaO) and recovered clinkers (i.e., C₂S) (Bogas et al., 2020; Serpell and Zunino, 2017; Yu et al., 2013). For clay minerals, the dehydroxylation by thermal treatment at 500–900 °C transforms the crystal minerals into an amorphous phase, making the treated materials exhibit pozzolanic reactivity (Zhao and Khoshnazar, 2020). The reactivity of calcined clays is mainly governed by the types of clay minerals (Zhao and Khoshnazar, 2020). Kaolinitic clays, which are often regarded as high-quality clays, exhibit high pozzolanic reactivity after thermal treatment, while some other clay minerals (e.g., montmorillonites and illite) show low reactivity or even remain inert after being heated; hence, they have been classified as low-quality clays for the use as SCMs (Hollanders et al., 2016).

A major concern of using calcined DSMR as a possible SCM is that its mineral composition tends to be dominated by the low-quality clays in it, due to the limited availability of kaolin-rich clays in the local natural soils, which thus limits its reactivity (Hollanders et al., 2016). Nevertheless, the co-calcination of low-quality clays and cement hydrates, which are the main minerals in the DSMR, might be able to form some highly reactive phases, favoring the reactivity of the calcined DSMR. This is because recent research demonstrated that the reactivity of low-quality clays could be significantly increased when they are co-calcined with Ca-rich minerals (e.g., calcite or dolomite) (Bullerjahn et al., 2020; Cultrone et al., 2001). This is due to the reaction between the clays and the free CaO released from the decomposed calcite or dolomite during calcination, leading to the formation of Ca-rich amorphous phases with highly pozzolanic activity (Bullerjahn et al., 2020; Cultrone et al., 2001). Similar reactive phases are expected to be generated in calcined DSMR with low-quality clays, as the hydrated cement present in DSMR could work as a Ca-rich source, considering that free CaO is one of the main products of the thermally dehydrated cement (Carriço et al., 2020).

The research significance and novelty of the present study are that, for the first time, the potential of using DSMR after calcination as an SCM is investigated. For this purpose, three DSMR samples from different construction sites located in Flanders, Belgium, were collected. They were then ground into fine powders and thermally activated (calcined). The chemical and mineral compositions of the DSMR powders before and after the calcination were characterized using X-ray fluorescence (XRF), thermogravimetric analysis (TGA), X-ray diffraction (XRD), scanning electron microscopy-energy dispersive X-ray spectroscopy (SEM-EDS), and Fourier Transform Infrared Spectroscopy (FTIR). The hydration of cement pastes blended with the DSMR powders at different curing ages (1, 7, 14, and 28 days) was quantitatively monitored with XRD and TGA. Finally, the compressive strength development of the DSMR powder blended pastes was evaluated, and the results were compared with that of cement pastes blended with GGBS, which is a conventional highly reactive SCM (Yousefi Oderji et al., 2019).