Influence of aqueous carbonate species on hydration and carbonation of reactive MgO cement

Highlights

• Boundary nucleation and growth model for MgO hydration under ambient conditions.

• No detectable carbonation of reactive MgO from brine under ambient conditions.

• CO₂ curing of reactive MgO from brine leads to hydrated carbonates, strength gain.

• Sodium bicarbonate and sodium sesquicarbonate in MgO enhances compressive strength.

• Amorphous phase in MgO with NaHCO₃ and Na₃H(CO₃)₂∙2H₂O plays key strengthening role.

Abstract

Reactive MgO cement derived from brine could be a low-carbon alternative to Portland cement, as it avoids the need to burn carbon-containing raw materials and it can absorb CO₂ gas to form high-strength mineral carbonates. Additionally, the incorporation of a high concentration of aqueous carbonate species can further contribute to its negative CO₂ footprint. This paper aims to study the influence of aqueous carbonate species on the hydration and carbonation behavior of reactive MgO cement derived from brine. Although the hydrated MgO exhibited much higher compressive strength in the presence of bicarbonate solution, it still did not exhibit compressive strength sufficient for structural applications (> 10 MPa) under ambient conditions. TGA and XRD results revealed that the carbonate and/or bicarbonate solution altered the hydration degree and morphology of formed Mg(OH)₂ but did not form new hydration products. In contrast, when MgO paste was exposed to concentrated CO₂ gas (20 % CO₂) it produced magnesium carbonates and reached dramatically higher compressive strength, nearly 60 MPa by 28 days. TGA and XRD results suggest that an amorphous phase forms in sodium bicarbonate and sodium sesquicarbonate systems, which plays a key role in the enhancement of compressive strength.

Introduction

Increased carbon dioxide (CO₂) emissions is the most significant environmental challenge of this century [1]. Manufacturing of cement, whose dominant form used nowadays is Portland cement, is responsible for about 5∼7 percent of global greenhouse gas emissions [[2], [3], [4], [5], [6]]. The conventional feedstock for Portland cement manufacturing is limestone (CaCO₃), which is excavated and crushed. Then it is sintered with other materials (i.e. clays) at temperatures reaching ∼1450°C in a cement kiln to produce clinker. This process directly releases CO₂ during calcination of CaCO₃ (i.e. CaCO₃ → CaO + CO₂), which makes up ∼ 50–60% of the total emissions of Portland cement production [3]. Much of the remaining emissions are associated with the use of fossil fuels, such as coal, to generate the heat required to achieve the high temperatures needed for clinker manufacturing [3]. From a sustainability perspective, the cement industry is calling for alternative binders that can reduce associated CO₂ emissions without compromising performance.

One alternative to Portland cement is magnesium oxide (MgO)-based cement [7]. If calcined, reactive MgO powder is mixed with plain water, it will react to form brucite (Mg(OH)₂), which has a porous structure and is typically quite low in strength [8,9]. However, there has been renewed interest by both industry and academia in MgO-H₂O systems due to their ability to absorb CO₂, forming carbonated products that can be as strong as hydrated Portland cement. Similar approaches to carbonation curing of calcium silicate systems and calcium hydroxide are also under development and show great promise [[10], [11], [12]]. The U.S. patent granted under the title “Reactive magnesium oxide cements” in 2008 describes the strength gain of reactive MgO cement under carbonation [13]. Since then, some studies have investigated various aspects of the carbonation of reactive MgO, including carbonation condition [[14], [15], [16], [17]], microstructure [18], additives that enhance hydration and carbonation [16,[19], [20], [21], [22]], durability [23,24], and life cycle assessment [25]. MgO-H₂O systems are a promising candidate to replace conventional Portland cement in various applications, including porous blocks [9,26], concrete [8,27], and ground improvement [28,29].

The raw material, MgO, can be obtained by two main methods: (i.) a dry processing route in which magnesite (MgCO₃) is calcined to MgO while releasing CO₂ (i.e. MgCO₃ → MgO + CO₂), and (ii.) a wet processing route in which precipitated Mg(OH)₂ powder is recovered from Mg-rich seawater or brine sources and subsequently calcined to MgO, releasing H₂O in the process (i.e. Mg(OH)₂ → MgO + H₂O). The calcination temperature of either route is lower than Portland cement (700∼1000 °C vs. 1450 °C), which could be beneficial from a sustainability standpoint. Between the two processing routes, the drawback of the current wet processing route is its high energy consumption, which is three times higher than the dry process route [30]. However, MgCO₃ deposits are mainly concentrated in a few countries, such as China, North Korean, and Russia [31]. The wet process route provides an option for regions with limited availability of natural MgCO₃ deposits, and contributes to ∼14 % of global MgO production [32]. H. Dong studied the wet processing route by utilizing reject brine, obtained from desalination plants in Singapore, which once had no economic value and could have been harmful to the marine ecosystem [[33], [34], [35]]. Also, the MgO obtained through the wet process route has higher purity and reactivity [33,36,37], which can lead to higher potential for carbonation to improve properties and carbon storage. However, as most studies have utilized raw material derived from the dry process route, work on MgO from the wet process route is currently still lacking.

Under accelerated carbonation conditions, where the specimen is exposed to an environment with highly concentrated CO₂ gas, gaseous CO₂ dissolves into water and releases carbonate and bicarbonate ions:

CO₂ (g) ⇌ CO₂ (aq)

CO₂ (aq) + H₂O ⇌ H⁺+ HCO₃⁻

HCO₃⁻ ⇌ H⁺+ CO₃^2-

MgO-H₂O systems intake CO₂, leading to carbonate products in the form of hydrated magnesium carbonates (HMCs), including nesquehonite (MgCO₃·3H₂O), hydromagnesite (4MgCO₃·Mg(OH)₂·4H₂O), dypingite (4MgCO₃·Mg(OH)₂·5H₂O), and artinite (MgCO₃·Mg (OH)₂·3H₂O). These carbonate products develop a strong and durable microstructure that can be comparable to that of Portland cement hydration products [38]. In previous studies [21,39], it was found that there still exists a large amount of uncarbonated brucite (up to 55 %) after curing in a concentrated CO₂ condition. To address this limitation, [22,40] utilized sodium bicarbonate (NaHCO₃) to effectively enhance the carbonation degree of reactive MgO cement. However, the specific role of NaHCO₃ in the carbonation process is still not fully understood. It was suggested that HCO₃⁻ transfers to CO₃^2- and associates with Mg²⁺ to form HMCs in the presence of OH⁻ [22]. It was also assumed that NaHCO₃ can react with Mg(OH)₂ directly to form HMCs [22,40]. However, NaHCO₃ was not found to significantly affect the hydration products in MgO-hydromagnesite blends [41]. Thus, this study aims to clarify the influence of HCO₃⁻ and CO₃^2- on the hydration and carbonation of reactive MgO cement by incorporating a relatively high concentration of aqueous carbonate species. Additionally, reactive MgO with the addition of aqueous carbonate species is rich in carbonate even before exposure to a concentrated CO₂ environment, which could be beneficial in further reducing carbon footprint if the aqueous carbonate species could be produced by using exhaust CO₂ through a carbon negative process, such as a modified Solvay ammonia soda process mentioned in [42,43].

In this study, we focused on MgO derived from the wet process route. We investigated the hydration and carbonation behavior of MgO cement systems with the addition of sodium carbonate (Na₂CO₃), sodium bicarbonate (NaHCO₃) and sodium sesquicarbonate dihydrate (Na₃H(CO₃)₂·2H₂O). The influence of carbonate and bicarbonate ions on hydration and carbonation was studied through compressive strength testing, isothermal calorimetry, thermal gravimetric analysis (TGA), X-ray powder diffraction (XRD), and scanning electron microscopy (SEM).