Next Article in Journal
Design and Implementation of Sampling Wells in Phosphate Mine Waste Rock Piles: Towards an Enhanced Composition Understanding and Sustainable Reclamation
Previous Article in Journal
Geochemistry, Sr-Nd Isotope Compositions, and U-Pb Chronology of Apatite from Kimberlite in Wafangdian, North China Craton: Constraints on the Late Magmatic Processes
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Minerals
Volume 14
Issue 3
10.3390/min14030285
minerals-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Mineralogical Evolution of High-pH/Low-pH Cement Pastes in Contact with Seawater

by
Yutaro Kobayashi
1,2,* and
Tsutomu Sato
3
1
Taiheiyo Consultant Co., Ltd., Tokyo 101-0054, Japan
2
Division of Sustainable Resources Engineering, Graduate School of Engineering, Hokkaido University, Sapporo 060-8628, Japan
3
Division of Sustainable Resources Engineering, Faculty of Engineering, Hokkaido University, Sapporo 060-8628, Japan
*
Author to whom correspondence should be addressed.
Minerals 2024, 14(3), 285; https://doi.org/10.3390/min14030285
Submission received: 13 February 2024 / Revised: 4 March 2024 / Accepted: 6 March 2024 / Published: 8 March 2024
(This article belongs to the Section Environmental Mineralogy and Biogeochemistry)
Browse Figures
Versions Notes

Abstract

:
In facilities for the geological disposal of radioactive waste in coastal areas, the long-term alteration of cementitious materials in engineered barriers is expected to occur due to the ingress of groundwater derived from seawater. Although the reaction between cement and seawater has been studied, the alteration behavior caused by the reaction between seawater and low-pH cement, which is expected to be used in a disposal facility, has not yet been clarified. In this study, the effects of cement type on cement–seawater interactions were investigated, and the chemical stability and mineral evolution of cement pastes caused by reactions with seawater were determined. The dissolution of cement hydrates occurred upon increased contact with seawater, and the formation of secondary minerals, including carbonate and Mg-containing minerals, was observed. The progress of dissolution depended on the mineral composition of the initially formed cement hydrates, and low-pH cement containing pozzolanic materials showed less resistance to seawater. Differences in pH and Si concentration that are due to the type of cement used had a strong influence on the evolution of minerals (especially Mg-containing minerals), implying that the formed mineral species possibly affect the migration characteristics of radionuclide.

1. Introduction

Low-level radioactive waste containing long-half-life transuranic nuclides (TRU waste) is considered for the same geological disposal as high-level radioactive waste to isolate it from the human living environment [1]. For the geological disposal of TRU waste, an engineered barrier system is proposed for the long-term inhibition of radionuclide migration. Moreover, cementitious materials are expected to be used as solidifying materials for waste form manufacturing and as structural support for disposal tunnels [1]. Although Ordinary Portland cement (OPC) is the most common cementitious material, its pore solution has a pH > 12.5, potentially forming a hyper-alkaline plume due to groundwater percolation. The contact of a hyper-alkaline plume with buffer materials, such as compacted bentonite, that surround cementitious materials may lead to the alkaline alteration of these materials, which may cause the dissolution of minerals [2,3,4,5,6] and/or secondary mineral formation to occur [4,5,6,7,8,9]. These phenomena may change the permeability of barrier materials and affect the long-term performance of the repository.
Low-pH cements with pore water pH below 11 have been considered to minimize the alkali alteration of the compacted bentonite for several decades [10,11,12]. Low-alkaline cement (LAC), also termed high-volume fly ash silica fume cement (HFSC) was developed by the Japan Atomic Energy Agency (JAEA) as a mixture of siliceous materials (coal fly ash, silica fume) and OPC [13]. The reaction of cement hydrates, such as calcium hydroxide and calcium silicate hydrate (C-S-H), with siliceous materials in LAC increases the content and decreases the Ca/Si molar ratio of C-S-H, resulting in a decrease in porewater pH in the LAC due to the high consumption of calcium hydroxide and Na and K uptake in C-S-H.
Coastal areas are considered to be the preferred locations for geological disposal sites in Japan from the perspective of safe waste transportation [14]. It is assumed that disposal facilities will be exposed to saline groundwater of seawater origin, therefore the long-term alteration of cementitious materials in a saline environment needs to be assessed. To evaluate the durability of concrete structures in marine environments, laboratory, and field exposure tests on the interactions between cementitious materials and seawater have been conducted [15,16] and summarized in several review papers [17,18,19,20]. De Weerdt et al. [20] combined the results of previous studies with thermodynamic modeling and focused on Mg, CO3, SO4, Cl, and Na as components in the contact solution that affect the long-term degradation of cementitious materials under saturated conditions, summarizing their phase transitions. They reported that the main degradation factor under saturated conditions was the leaching of cement hydrates regardless of solution composition, with the degree of leaching and the type and amount of secondary phases depending on the solution composition. Additionally, the resistance to chemical attacks from seawater was reported to change when pozzolanic materials were mixed with OPC due to the suppression of calcium hydroxide formation, increased C-S-H formation, and changes in secondary minerals [18]. Although the mechanisms of interaction between cementitious materials and seawater are becoming progressively better understood, current knowledge on the mineral phase transition of cement hydrates during reactions with seawater is still lacking. In particular, cementitious materials, such as LAC, which was developed for applications in geological disposal facilities, are rarely used in marine concrete, and there are limited examples of studies on their interactions with seawater.
In this context, seawater immersion experiments were conducted on cement pastes with OPC and LAC to investigate the alteration behavior of the constituent minerals during reactions with seawater.

2. Materials and Methods

2.1. Materials

The cement used to prepare two types of cement pastes (OPC paste and LAC paste) was ordinary Portland cement (Japan Cement Association, Tokyo, Japan), which was made using only cement clinker and gypsum. Chemical additives were not used for either paste. LAC was prepared by mixing OPC, silica fume (SF, Microsilica® 940U, Elkem Japan, Tokyo, Japan), and fly ash (FA, JIS type-II, Hekinan Thermal Power Station, JERA, Aichi, Japan) in 40:20:40 wt.%. The chemical composition of each material is shown in Table 1 and the mineral composition of FA is shown in Table 2. The crystalline phases in the FA are chemically inert in an alkaline environment, and the glass in FA contributes to the pozzolanic reaction [21,22,23]. Before use, each powder material was put into a plastic bag and mixed by hand shaking until visually homogeneous.
Artificial seawater was prepared using ultra-pure water (<18.2 MΩ·cm, Milli-Q® Integral 3 system, Merck Millipore, Tokyo, Japan) and commercial artificial seawater (Aquamarine, Yashima Pure Chemicals, Osaka, Japan) whose chemical concentration is shown in Table 3. The pH of artificial seawater was adjusted to pH = 8.2 by adding a NaOH solution.

2.2. Sample Preparation

Both cement pastes were prepared at 20 °C with a water-to-binder ratio (W/B) of 0.6. To prevent material segregation, the fresh pastes were kneaded repeatedly until bleeding water could not be visibly observed after mixing. These pastes were cured in sealed polypropylene (PP) containers in an oven at 50 °C for 28 days. The cured samples were coarsely ground and vacuum-dried for 5 days, and then the powder samples ground to less than 250 µm were used for the immersion experiment.

2.3. Immersion Experiment

In the immersion experiment, OPC and LAC paste powders were dispersed in artificial seawater at liquid-to-solid ratios (L/S) of 10, 50, 100, 500, and 1000. The PP bottles containing the samples were sealed under an inert Ar gas atmosphere to prevent carbonation of the samples by CO2 gas in the atmosphere. The sealed PP bottles were placed in an oven at 60 °C for 28 days and hand-shaken once a day. The solid and aqueous phases were separated by suction filtration using a PTFE membrane filter with a pore size of 0.45 µm. The solid sample was vacuum-dried for 7 days.

2.4. Solid-Phase Analyses

The chemical compositions of FA and SF were measured using X-ray fluorescence (XRF: ZSX Primus, Rigaku, Tokyo, Japan) according to the JIS standards [24,25,26]. In the sample preparation, FA was finely ground using a disc mill, and the silica fume was dried at 105 °C before analysis.
To characterize the mineral composition in the samples before and after the experiment, the samples were analyzed using powder X-ray diffraction (XRD: X’Pert Pro MPD, PANalytical, Tokyo, Japan) with target Cu Kα, a tube voltage of 45 kV, a tube current of 40 mA, a scanning range of 5–60° 2θ, and a semiconductor detector (X’celerator, PANalytical, Tokyo, Japan). The X’Pert HighScore Plus version 4.9 software (PANalytical, Tokyo, Japan) and PDF-2 Release 2016 database (International Centre for Diffraction Data ICDD, Newton Square, CT, USA) were used for phase identification.
Raman microspectroscopy (XploRA PLUS, Horiba, Kyoto, Japan) was performed for compacted powder samples under the following measurement conditions: laser wavelength, 532 nm; grating, 2400; confocal hole diameter, 300 µm; and slit size, 100 µm. Samples were compacted using a hand press with a 2 mm die set (JASCO, Tokyo, Japan). The spectral acquisition time was 60 s and data for 3 spectra were recorded for each measurement point. The system was calibrated using a silicon standard. For the analysis, powder samples were compacted using a hand press and analyzed at 100 points (10 vertical × 10 horizontal) in a 100 µm square area, and the average spectrum of the measurement area was calculated from the spectral data of the 100 points.
Solid-state magic-angle spinning nuclear magnetic resonance (NMR: JNM-ECA600, JEOL RESONANCE, Tokyo, Japan) was measured under static field strength, 14.1 T. A 3.2 mm zirconia sample tube and a MAS probe were used to acquire the single-pulse MAS spectra at spinning speeds of 20 kHz. For 29Si NMR measurements, we used π/2 pulses (2.8 µs) with recycle delays of 40 s and an accumulation of 2000. Peak positions in 29Si NMR spectra were referred to tetramethyl silane (Si(CH3)4) as the external standard. For 27Al NMR measurements, we used π/10 pulses (0.47 µs) with recycle delays of 0.5 s, and an accumulation of 10,000. Peak positions in 27Al NMR spectra were referred to potassium aluminum sulfate hydrate (AlK(SO4)2·12H2O) as the external standard.

2.5. Aqueous-Phase Analyses

For pH measurements, samples were left in a laboratory at 20 °C for at least 6 h and then measured using the glass electrode method. After calibration with three different pH standard solutions (pH = 7, 9, and 12), a pH meter (F-72, Horiba, Kyoto, Japan) and glass electrode (9632-10D, Horiba, Kyoto, Japan) were used for measurements.
The chemical composition of the liquid phase was analyzed using inductively coupled plasma atomic emission spectroscopy (ICP-AES: SPECTRO BLUE EOP, SII, Tokyo, Japan) for Ca, Si, Al, Mg, and Fe ion concentrations, and an ion chromatograph system (IC: Dionex Integrion HPIC system, Thermo Fisher Scientific, Waltham, MA, USA) for SO42− ion concentration.

2.6. Thermodynamic Calculation

The Geochemist’s Workbench® 2023 Professional version 17 (GWB, Aqueous Solutions, Champaign, IL, USA) software package was used for the thermodynamic calculations to obtain the phase diagram for Ca-Si-H2O and Mg-Si-H2O systems, and the solution concentration and mineral composition were calculated under the same conditions as in the present experiment. In these calculations, the thermodynamic database “ThermoddemV1.10_15Dec2020”, provided by the French geological survey (BRGM, Orleans, France), was used after the addition of equilibrium constants for magnesium silicate hydrate (M-S-H), Mg-Al/Fe, layered double hydroxide (LDH), and siliceous hydrogarnet, as reported in a previous study (Table 4). The equilibrium constant K at 60 °C was calculated using the thermodynamic properties reported in the reference literature referring to the method (3-term approximation) described in Lothenbach et al. [27]. The Act2 module in the GWB software was used to obtain the phase diagram for Ca-Si-H2O and Mg-Si-H2O systems, and the GSS module was used to calculate the activity of aqueous species such as Ca, Si, and Mg. The React module was used to determine the solution pH at 60 °C and the solution concentration and mineral composition under same conditions as in the present experiment.

3. Results

3.1. Solid Characterization

The XRD patterns of the OPC pastes immersed in artificial seawater at different L/S ratios showed reflections of cement hydrates such as portlandite (Ca(OH)2), ettringite (Ca6Al2(SO4)3(OH)12·26H2O), and calcium silicate hydrate (C-S-H; xCaO·ySiO2·zH2O) in the initial sample (Figure 1a). These hydrates were dissolved in a reaction with artificial seawater; portlandite disappeared at L/S ≥ 50, and ettringite and C-S-H disappeared at L/S ≥ 500. Secondary minerals formed in the reaction between the OPC paste and artificial seawater were categorized into three groups: the (1) hydroxide group, (2) sulfate group, and (3) carbonate group.
Portlandite was present in the initial sample as a hydroxide mineral, which was decalcified by reacting it with artificial seawater with increasing L/S ratios; brucite (Mg(OH)2) was formed at L/S ≥ 50 by Mg derived from artificial seawater. Reflections of layered double hydroxide (LDH) were identified at L/S ≥ 500. LDH has a lamellar structure consisting of brucite-like hydroxide sheets in a solid solution of divalent and trivalent metal cations and is known to intercalate anions between the layers [32]. In previous studies on cement chemistry, hydrotalcite, consisting of Mg2+ and Al3+, was the Mg-containing phase in cement pastes [6]. The basal spacing of LDH changed depending on the content ratio of divalent and trivalent cations, and the interlayer anion type was reported [32,33]. The reflections of LDH identified in this experiment were broad, suggesting that the low crystallinity and composition of divalent/trivalent cations and interlayer anions in LDH may not be simple. In addition, the color of the samples with L/S > 500 was reddish, implying the possibility of the formation of Mg-Fe(III) LDH and/or amorphous ferric hydroxide as a mineral containing ferric ion.
Although carbonates were not present in the initial sample, Ca was leached from cement hydrate with increasing L/S ratios and reacted with carbonate ions in artificial seawater to form calcite (CaCO3) at L/S ≥ 50 and aragonite (CaCO3) at L/S ≥ 500. Calcite is the most stable calcium carbonate mineral at ambient temperature and pressure, while aragonite is in the metastable phase. However, under high Mg concentrations, the crystal growth of calcite was inhibited and aragonite formation was more common [34]. Similar phenomena were reported in environments with high Mg/Ca molar ratios (e.g., marine concrete structures [35], high alkaline springs [36]). Ettringite was present as a sulfate mineral in the initial sample, while a decrease in pH with increasing L/S ratios led to the decomposition of ettringite into gypsum (CaSO4·2H2O) at L/S = 100.
XRD patterns of the LAC pastes immersed in artificial seawater at different L/S ratios showed that the initial sample contained cement hydrates, such as ettringite and C-S-H, as well as quartz and mullite that originated from FA (Figure 1b). Quartz and mullite were present at all L/S ratios and no changes in their reflections were observed. These results showed that the crystalline phases in FA were chemically inert during the immersion experiment. The halo of glass in FA (2θ = 15−30°) could not be identified in the LAC paste and it was difficult to observe the changes in the glass phase before and after immersion in seawater. Although it is possible that the glass phase present in the LAC paste before the experiment was partially dissolved by contact with seawater and affected the formation of silicate hydrate which is described below, the details are unknown.
Cement hydrates were dissolved by smaller amounts of seawater compared to the OPC pastes; ettringite dissolved at L/S ≥ 10 and C-S-H dissolved at L/S ≥ 100. In the LAC paste, the initial pH was lower due to the mixing of pozzolanic materials, and no hydroxide minerals were formed, as identified in the OPC paste, leading to the early dissolution of the cement hydrates. Secondary minerals formed in the LAC pastes were categorized into three groups: the (1) silicate group, (2) sulfate group, and (3) carbonate group.
For silicate minerals, magnesium silicate hydrate (M-S-H; xMgO·ySiO2·zH2O) was formed at L/S ≥ 50 by the reaction between Si derived from the dissolution of cement hydrates and Mg in artificial seawater. For sulfate minerals, similar to the OPC paste, ettringite decomposition and gypsum formation were identified with increasing L/S ratios, but the L/S ratio where phase transformation occurred was smaller than in the OPC paste. For carbonate minerals, as in the OPC pastes calcite was formed but aragonite, which was identified in the OPC pastes with L/S ≥ 500, was not formed. Instead, in the LAC pastes with L/S ≥ 500, the reflections of calcite were shifted (2θ = 29.45° to 29.76°), indicating the incorporation of Mg into calcite (magnesian calcite). In this experiment, the Mg/Ca ratio in the solution was greater in the LAC pastes than in the OPC pastes for the same L/S ratio, so aragonite formation was generally expected in LAC, but not identified by XRD. In natural environments, the presence of layered silicate minerals, such as clay minerals, could potentially affect the formation of Mg-containing calcium carbonate phases [37,38]. Molnár et al. [39] reported that aragonite was initially formed under conditions characterized by large Mg/Ca ratios but transformed from aragonite to magnesian calcite when smectite was present. In the LAC pastes with L/S ≥ 500, M-S-H, which is considered a precursor of phyllosilicates [40,41], may play a similar role to smectite, resulting in the formation of magnesian calcite.
In the 29Si MAS NMR spectra of the OPC and LAC pastes (Figure 2), peaks of Q0 sites attributed to C2S and C3S were identified from around −65 to −75 ppm in the initial samples, but these peaks almost disappeared in the post-immersion samples due to hydration progress. The relative ratio of the peak intensity of the Q2 site (around −85 ppm) to the Q1 site (around −80 ppm) is higher in the LAC paste than in the OPC paste, indicating that the Ca/Si ratio of C-S-H formed in the LAC paste is smaller than that in the OPC paste. In silicate minerals, the peak position of the Qn (0 ≤ n ≤ 4) site shifts positively when the neighboring Si tetrahedra are replaced by Al tetrahedra [42]. In both samples, the contribution of the Q2(1Al) site was observed between the peaks of the Q1 and Q2 sites, suggesting that the Al was incorporated into C-S-H by substitution with Si in the C-S-H. Regardless of cement type, the peak intensity at the Q1 site decreased and that at the Q3 site (−100 to −90 ppm) increased as the L/S ratio increased. These changes were consistent with studies that reported the formation of M-S-H via the reaction of C-S-H with Mg [43], suggesting the dissolution of C-S-H in the cement paste and the formation of M-S-H. No M-S-H peaks were identified in the XRD pattern of the OPC pastes; therefore, it was assumed that the proportion of M-S-H in the sample was smaller than for other minerals. Peaks of the Q2 and Q3 site were identified in both samples at L/S = 1000, but the peaks were continuous in the LAC paste, whereas each peak was independent in the OPC paste. It was reported that in Al-incorporated M-S-H, the contribution of the Q2(Al) site appears between the Q1 and Q2 peaks, and the Q3(Al) site between the Q2 and Q3 peaks, respectively [44]. A significant amount of Al was incorporated into the M-S-H in the LAC paste due to the substitution with Si in M-S-H.
In the 27Al MAS NMR spectra of the OPC pastes (Figure 3a), the peaks attributed to 6-coordinated Al (0–20 ppm) were dominant in all samples compared to the peaks of 4-coordinated Al (60–80 ppm). The intensity of 6-coordinated Al increased while that of 4-coordinated Al decreased with increasing L/S ratio. The results indicate that although 4-coordinated Al in the initial sample seemed to be incorporated into C-S-H, the majority of Al in OPC samples was present as calcium aluminate hydrate or LDH rather than incorporated into silicate hydrates. LDH formed in the OPC pastes was considered to be a possible solid solution for Fe(III) and Al from XRD and 27Al MAS NMR results. In contrast, the intensity difference between the peak of 4- and 6-coordinated Al in the LAC pastes was smaller than for the OPC pastes, suggesting a higher Al incorporation into C-S-H in the initial sample. Layered phyllosilicate minerals including M-S-H are composed of tetrahedral sheets (T), which consist of tetrahedra such as Si-O and Al-O, and octahedral sheets (O), which consist of octahedra such as Mg-O and Al-O. The Al incorporated in M-S-H was reported to exist in both tetrahedral and octahedral sheets, with 4- and 6-coordinated Al [44]. The spectral shapes of the LAC paste at L/S = 100 and 1000 (Figure 3b) were similar to the spectra of Al-incorporated M-S-H in a previous study [44], in agreement with the 29Si MAS NMR results, indicating that Al incorporation into M-S-H occurred. Spectra of the LAC pastes also showed some contribution of 4- and 6-coordinated Al in unreacted FA (around 50 and 0 ppm, respectively) in all samples.
Raman spectra of the OPC and LAC pastes (Figure 4) showed spectral changes consistent with the mineral phase transitions identified using XRD analysis: portlandite dissolution (355 cm−1), brucite formation (440 cm−1), gypsum formation (490, 630 cm−1), and calcium carbonate formation (calcite: 150, 700 cm−1; aragonite: 150, 200, 700 cm−1). Peak shifts of approximately 650–700 cm−1 were observed with increasing L/S ratios in both the OPC and LAC pastes. The shift occurred at L/S >100 for the OPC pastes and L/S > 10 for the LAC pastes and was assumed to be caused by the dissolution of C-S-H and the formation of M-S-H as a secondary product. M-S-H structures were reported to have either a serpentine-like 1:1 type structure with tetrahedral and octahedral sheets combined 1:1 (T-O layer), or a talc-like 2:1 type structure with octahedral sheets sandwiched between tetrahedral sheets (T-O-T layer) [28,45,46,47]. The peak positions of M-S-H identified in the immersed samples were consistent with the serpentine group (around 680 cm−1) for the OPC pastes and talc (around 670 cm−1) for the LAC pastes, indicating that the formed structure of M-S-H was different depending on the type of cement. The Mg content in the system at larger L/S ratios depends more on the artificial seawater and less on the different cement types. Therefore, the formed structure of M-S-H was inferred to vary due to differences in pH and the supply of Si and/or Al caused by different cement types.

3.2. Aqueous Chemistry

The pH transitions of the OPC and LAC pastes in artificial seawater with increasing L/S ratio are shown in Figure 5a. In OPC samples, the pH decreased with increasing L/S ratios and remained almost constant at pH ≈ 9.5 for an L/S ratio > 500. Meanwhile, in LAC samples, the pH was already pH < 11 at an L/S ratio = 10 and decreased with increasing L/S ratios, reaching a pH ≈ 8.0 at an L/S ratio > 100, which is the same level as artificial seawater. These trends of decreasing pH were expected to be influenced by the dilution effect of the pore water in the cement pastes due to the addition of seawater, as well as by the dissolution of cement hydrates.
The chemical compositions of the OPC and LAC pastes in artificial seawater with the increasing L/S ratio are shown in Figure 5b–f. In a small L/S ratio, Ca concentrations were higher than artificial seawater due to the dissolution of cement hydrates, while the Mg and SO42− concentrations were lower than in artificial seawater due to the formation of secondary minerals, both of which converged to the concentration in artificial seawater with an increasing L/S ratio. Si concentrations were significantly higher in the LAC samples than in the OPC samples at all L/S ratios. Al concentrations were below the limit of quantification in most samples for both OPC and LAC, indicating only a minimal dissolution from the solid phase.
The chemical compositions of the immersion solution were plotted in phase diagrams for the Ca-Si-H2O and Mg-Si-H2O system obtained using thermodynamic calculation (Figure 6). For both the OPC and LAC pastes, solutions were plotted along the C-S-H solubility curve at small L/S ratios and at the boundary, with a lower Ca/Si ratio at C-S-H as the L/S ratio increased. This suggests that C-S-H in the cement paste was in equilibrium with the immersion solution in these samples, and that the Ca/Si ratio of C-S-H decreased with the increasing L/S ratio. On the other hand, solutions with large L/S ratios were plotted using the C-S-H solubility curve, indicating that C-S-H in the cement paste dissolved and was lost. These solutions were plotted around the M-S-H solubility curve, suggesting that M-S-H was formed by the reaction of the cement paste with Mg from seawater under a large L/S ratio. These changes in silicate hydrates were observed via solid-phase analysis and were supported by the solution composition and thermodynamic calculations.

4. Discussion

4.1. Influence of Cement Types on Cement–Seawater Interactions

4.1.1. Mineral Transitions during Cement–Seawater Interactions

In the disposal system constructed with Portland cement, considering the dissolution characteristics of cement hydrates and the leaching of the Na and K in the cement, cement hydrates were altered by contact with groundwater as follows [48]:
  • The leaching of Na and K contained in Portland cement;
  • The dissolution of portlandite (Ca(OH)2);
  • The dissolution of calcium aluminate hydrates and decalcification of C-S-H;
  • The dissolution of C-S-H.
De Weerdt et al. [20] calculated mineral transitions in contact with solutions with different compositions by thermodynamic calculations. They showed that the dissolution sequence of cement hydrates is generally similar with the above, although there are some differences in the timing of dissolution and the type of aluminate hydrates formed. In this study, the sequence of cement hydrates dissolution was generally similar for both the OPC paste and the LAC paste, and no influence of cement types was observed. However, the LAC pastes were shown to be less resistant to seawater than the OPC pastes. Experimental results showed that portlandite was not formed in the LAC pastes, and its Ca/Si ratio of C-S-H was smaller than that of the OPC pastes. The shorter dissolution sequence of cement hydrates in the LAC pastes compared with the OPC pastes may have resulted in the dissolution of cement hydrates such as C-S-H and ettringite in contact with a smaller amount of artificial seawater. In the case of concrete mixed with SF, C-S-H with a low Ca/Si ratio formed by the reaction of portlandite and SF is easily converted to M-S-H by the attack of Mg in seawater, and the compressive strength of the concrete is lower than in non-SF-mixed specimens [49,50]. It has been reported that cementitious materials mixed with pozzolanic materials can reduce chloride ingress due to the densification of the microstructure and formation of chloride-binding hydrates [51,52], but their resistance to seawater is lower than that of OPC, suggesting that cement hydrates may dissolve early and lose the properties expected of cementitious materials when LAC is used. This implies a negative impact on the durability of the disposal facility, and a further detailed evaluation in more realistic size and environmental conditions is needed in the future.
The mineral transitions of the OPC and LAC pastes in contact with artificial seawater are shown in Figure 7. The mineral transitions depend on the type of cement, especially for Mg-containing minerals and carbonate minerals. The species of Mg-containing minerals were dependent on the aqueous pH. The OPC pastes with a relatively high pH enhanced the formation of hydroxide minerals (brucite, LDH) and limited the Mg concentration to low levels. On the other hand, hydroxide minerals could not be formed in the LAC paste because the pH was lower than that of the OPC paste under all conditions. In addition, the formation of M-S-H was observed in all samples, but its structure, composition ratio, and Al incorporation behavior were different. The LAC pastes used Si-rich SF and FA as starting materials, and their bulk Si content was more than twice that of the OPC pastes, suggesting a difference in the available Si content during the M-S-H formation process. In fact, the Si concentration of the LAC pastes after the experiment was more than 10 times higher than that of the OPC pastes. It was considered that a 2:1 type structure M-S-H with a relatively small Mg/Si ratio was formed in the Si-rich LAC pastes, and a 1:1 type structure M-S-H with a relatively large Mg/Si ratio was formed in the OPC pastes with a low Si content. For the carbonate minerals, different calcium carbonate minerals were formed under conditions with large L/S ratios, where the Mg/Ca ratio in solution was higher, suggesting the influence of coexisting M-S-H. M-S-H was formed in both samples, but as noted previously, their composition ratios and structures were different, and the structural differences in M-S-H may have affected the polymorphism of the carbonate minerals.
Thus, the differences in pH and Si content caused by the type of cement used have a significant effect on mineral transition due to the reaction between cement hydrates and seawater. LDH, identified only in the OPC pastes, is reported to have an anion exchange capacity that can interact with anionic radionuclides, and the formation of LDH in the OPC pastes may delay the migration of some anionic radionuclides. M-S-H is also considered to be a thermodynamically metastable phase with a low crystallinity and likely to transform into a crystalline phase [53], and the transformation was also observed in a natural hyperalkaline condition [54,55,56,57]. The differences in the structure and Al incorporation of M-S-H identified in the OPC and LAC pastes can not only affect the type of mineral formed as the stable phase, but also the chemical and physical properties of M-S-H, such as cation exchange capacity, swelling properties, and the polymorphism of the coexisting calcium carbonate. Although these changes are also expected to affect radionuclide retention properties and migration characteristics, our knowledge of the stability and sorption properties of M-S-H is still limited, and further data collection is required.

4.1.2. Applicability of Thermodynamic Calculation for the Cement–Seawater Interaction

Thermodynamic calculations were carried out under the same conditions as for the seawater immersion experiment and compared with the experimental results. In the calculations, the cement was assumed to be fully hydrated. Crystalline minerals such as quartz, mullite, and magnetite in fly ash were assumed to be inert as they are not considered to significantly contribute to the reaction. Only the reactive degree of silica fume and glass in the fly ash was considered in the LAC, which was estimated to be 90% and 25% for each material, respectively, based on a fitting of experimental results.
The pH and the aqueous concentration of major elements (Ca, Si, Mg) of the OPC and LAC pastes in artificial seawater with increasing L/S ratios are shown in Figure 8, comparing experimental and calculated results, and the mineral composition calculated using thermodynamic calculations is shown in Figure 9. The calculated Ca and Mg concentrations were in good agreement with the experimental data, and the trends in pH change were generally similar. The calculated mineral phase transition was similar to the experimental results, which showed that the dissolution of portlandite, decalcification of C-S-H, and formation of secondary minerals such as gypsum, calcite, and Mg-containing minerals occurred with increasing L/S ratios.
On the other hand, Si concentrations were estimated to be generally lower than experimental data, and Al and Fe concentrations, that were below the lower limit of quantification of most samples experimentally, differed significantly between the experimental data and calculated results. These differences were assumed to be mainly due to insufficient thermodynamic data for the following two phases; (1) Al-incorporated silicate hydrate such as C-A-S-H (calcium aluminosilicate hydrate) and M-A-S-H (magnesium aluminosilicate hydrate) and (2) Mg-Al/Fe LDH.
While the composition of the immersion solution could be explained by the equilibrium with C-S-H and M-S-H (as shown in Figure 6), Al was incorporated into C-S-H and M-S-H in the solid phase, as confirmed by NMR analysis. The thermodynamic data for C-A-S-H were also included in the thermodynamic data used in this study, and its formation was predicted in the calculations for the Si- and Al-rich LAC pastes. In contrast, C-A-S-H formation was not predicted by the calculations for the OPC pastes, except for the initial sample. The models of C-A-S-H have been studied using various approaches [57,58,59], but their thermodynamic data are still debated. M-S-H has been studied intensively in recent years [28,53], but there are only limited thermodynamic data available, especially for the Al-incorporated M-S-H, i.e., M-A-S-H, which has not yet been researched. The experiments in this study confirmed that M-S-H is the major product of cement–seawater interactions, regardless of cement types, and thermodynamic calculations using existing data were able to evaluate the formation of M-S-H. However, thermodynamic calculations for the LAC pastes in which a significant incorporation of Al into M-S-H was confirmed by NMR analysis, could not consider Al incorporation into M-S-H due to the lack of thermodynamic data for M-A-S-H, resulting in an excess of Al and the predicted LDH formation, which was not confirmed experimentally. C-A-S-H and M-A-S-H are the major phases of cement–seawater interactions, and thus the improvement of these thermodynamic data is expected to significantly reduce the difference between the experimental and calculated results.
Mg-Al LDH is a common LDH phase, and many thermodynamic data have been reported for conditions with variable Mg/Al ratios and interlayer anions [60,61]. These thermodynamic data mainly include OH− and CO32− as interlayer anions, with relatively limited data including anions such as SO42− and Cl−, and even less data that were able to account for temperature dependence. Although these anions are reported to have less selectivity than OH− and CO32− [62], they are present in large amounts in seawater and may not be negligible in cement–seawater interactions. By totaling thermodynamic data for the SO42−-type Mg-Al LDH, its formation could be predicted using the calculations. Cl− ions are present in concentrations approximately one order of magnitude higher than SO42− ions; therefore, additional thermodynamic data are required for the further study of Cl−-type Mg-Al LDHs. Al in Mg-Al LDH can be replaced by Fe3+, but thermodynamic data for Mg-Fe LDH are limited [61] and only CO32−-type Mg-Fe LDH (pyroaurite) is available at present [29]. Under conditions of larger amounts of contact solution, Mg-Fe LDH is expected to be a candidate host mineral for Fe; therefore, the prediction of Fe concentrations can be greatly improved by the further development of thermodynamic data for Mg-Fe LDH.

4.2. Influence of Aqueous Phase Chemistry on Cement Hydrate Alteration

Cementitious materials are used in long-life civil engineering structures, such as bridges, tunnels, and dams, which must have a long lifespan but are expected to degrade over time due to exposure to solutions with various chemical compositions, depending on their location. Therefore, many researchers have studied the degradation mechanisms of cementitious materials in contact with solutions with various compositions using methods such as field research, laboratory experiments, and thermodynamic modeling. The salt concentration of river water, groundwater, and seawater in natural environments generally differs by orders of magnitude, although the composition of the water depends on the types of rocks and minerals that come into contact with it, the degree of dilution by freshwater, and other factors. De Weerdt et al. [20] combined the results of previous studies with thermodynamic modeling and reported that sulfate and carbonate in solution enhanced the leaching of Ca-rich cement hydrates by forming Ca-containing phases (calcite, ettringite, gypsum, etc.). In this study, artificial seawater with a high sulfate and carbonate content was used, leading to the formation of sulfate and carbonate minerals. As a result, leaching of cement hydrates and a significant decrease in solution pH were observed. Seawater-derived groundwater, which is expected to contact a repository located in a coastal area, is thought to contain high levels of sulfate and carbonate, so it may be more aggressive to cementitious materials than relatively low-salinity solutions, such as those of groundwater derived from precipitation. Although the subsurface environment in which the geological disposal facility is constructed is not a leaching dominated environment, the results suggest that the chemical stability of concrete structures, which varies with groundwater composition, requires careful attention.
Mg, which is abundant in seawater, was reported to precipitate as magnesium-containing phases (brucite, LDH, M-S-H) near the contact interface with cementitious materials [16,20,35], and the formation of M-S-H was also reported at the cement–claystone interface due to contact with a Mg-rich claystone interstitial solution under an in situ experiment in an underground environment [6,63]. These hydrates were also observed to be formed by the reaction with seawater in this study. The formation of such secondary products was reported to locally reduce porosity and limit the ingress of other elements and the leaching of cement hydrates [64,65]. The formation also governs the pore water chemistry and evolution of mineral phases in the cement pastes. The concentration of each component in groundwater derived from seawater is expected to be higher than that in groundwater derived from precipitation, therefore, the number of secondary products is also expected to be higher, corresponding to the solution concentration, suggesting that the effects of clogging may be more significant and inhibit further alterations caused by carbonates and sulfates in the solution. On the other hand, some studies have reported that the ingress of chloride ions and the leaching of cement hydrates were not limited in cases of apparently dense layers formed by the precipitation of secondary products [66]. If the ingress of chloride ions is not limited, the risk of steel corrosion in the disposal facility will be higher because seawater contains a much higher number of chloride ions, and the pore solution pH of cementitious materials is significantly lower due to the reaction with seawater.
The results of the seawater immersion experiments with powder samples and the findings of previous studies indicated that the leaching of cement hydrates is accelerated in seawater, which has a higher salt concentration than river water or groundwater, and confirmed mineral transitions, including the formation of secondary products from the reaction between the leached and seawater components. Although the stability of the material was investigated in this study, which was a suspension system, it was implied that changes in mass transfer characteristics due to the precipitation of secondary products may also affect the chemical stability of the material. The evaluation of the effect of secondary mineral precipitation on mass transfer is limited by experiments using powder samples as in this study. Further understanding of these phenomena is expected through experiments using bulk samples. Although it is difficult to compare the phenomena in actual structures with the results of this study because construction experience with LAC is currently limited, a comparative study also needs to be conducted with the results of industrial analog studies under conditions similar to disposal facilities.

5. Conclusions

In order to identify differences between cement types on the reaction between cement paste and seawater, an immersion experiment of cement paste powder in seawater under different L/S ratios was conducted using OPC and LAC, which is a mixture of OPC with pozzolanic materials such as fly ash and silica fume. The dissolution of cement hydrates occurred with increasing L/S ratios in both the OPC and LAC pastes, and the formation of carbonate minerals and Mg-bearing minerals were identified as secondary minerals, but the degree of dissolution progress and the type of secondary minerals formed were dependent on the cement type. The addition of pozzolanic material in the LAC paste reduces the pore solution pH by inhibiting the formation of calcium hydroxide and decreasing the Ca/Si ratio of C-S-H. This property shortened the dissolution process of cement hydrates, resulting in the dissolution of cement hydrates in reaction to a smaller amount of seawater than the OPC paste. The low-alkaline LAC paste was lower in resistance to seawater as a material than the OPC paste, implying that it is likely to lose the properties expected of cementitious materials at an early stage.
In the case of mineral formation caused by reactions with seawater, there were significant differences in Mg-containing minerals and carbonate minerals depending on the type of cement. The types of Mg-containing minerals formed suggested that differences in the pH and Si concentrations of the immersion solution due to the type of cement had a strong influence on mineral transition. The Mg-bearing minerals identified in this study included LDH, with anion exchange capacity and M-S-H, which is expected to have cation exchange capacity. The composition, structure, and formation of these minerals were observed to vary with cement type, suggesting that the radionuclide migration properties may differ depending on the type of cement used. M-S-H may have affected the mineral type of calcium carbonate formed, suggesting that Mg-containing minerals may be key minerals in the mineral transition from cement hydrates and seawater reactions. The selection of cement type for disposal facility construction requires consideration not only of the long-term mineralogical evolution as shown in this study, but also of various other aspects such as workability, mechanical effects, and construction experience. This paper discussed the long-term mineralogical evolution of cement pastes from the perspective of retardation of radionuclide migration.
Thermodynamic calculations, using previously published thermodynamic databases and thermodynamic data, were generally able to predict the mineral evolution and associated concentration changes in major elements in the contact solution during cement–seawater interactions identified in this study. However, there were differences between the calculated and experimental results for the concentration changes for minor elements in the solution due to the shortage of thermodynamic data for Al- and Fe-containing phases, such as M-A-S-H and Mg-Al/Fe LDH. The further development of thermodynamic data for these phases is required to reduce the differences between actual phenomena and the results predicted using thermodynamic calculations.
It was shown that the leaching of cement hydrates was more likely to occur in high-salinity seawater than in freshwater, and that various minerals were formed by reactions with components in seawater, such as Mg, SO42−, and CO32−. It was found that contact with seawater-based solutions led to more significant reactions than contact with river water or groundwater. However, the approach of contact with solutions in actual structures is different from the approach used in this experiment: a suspension system. It is expected that effects on mass transfer will be evaluated in future experiments using bulk specimens and will be compared with the results of industrial analog studies under conditions similar to disposal facilities.

Author Contributions

Conceptualization, Y.K. and T.S.; methodology, Y.K.; investigation, Y.K.; resources, Y.K.; data curation, Y.K.; writing—original draft preparation, Y.K. and T.S.; writing—review and editing, Y.K. and T.S.; visualization, Y.K.; supervision, T.S.; project administration, T.S.; funding acquisition, Y.K. and T.S. All authors have read and agreed to the published version of the manuscript.

Funding

This work was financially supported by Grants-in-Aid for Scientific Research A (No. 19H00878) from the Japan Society for the Promotion of Science (JSPS) to T.S.

Data Availability Statement

The experimental data used to support the findings of this study are included in the manuscript.

Conflicts of Interest

Yutaro Kobayashi is an employee of Taiheiyo Consultant and received financial support from Taiheiyo Consultant. The funders had no role in the design on the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

References

  1. NUMO. The NUMO Pre-Siting SDM-Based Safety Case. NUMO-TR-21-01. 2021. Available online: https://www.numo.or.jp/technology/technical_report/pdf/NUMO-TR21-01_rev220222.pdf (accessed on 30 January 2024).
  2. Nakayama, S.; Sakamoto, Y.; Yamaguchi, T.; Akai, M.; Tanaka, T.; Sato, T.; Iida, Y. Dissolution of montmorillonite in compacted bentonite by highly alkaline aqueous solutions and diffusivity of hydroxide ions. Appl. Clay Sci. 2004, 27, 53–65. [Google Scholar] [CrossRef]
  3. Rozalén, M.L.; Huertas, F.J.; Brady, P.V.; Cama, J.; García-Palma, S.; Linares, J. Experimental study of the effect of pH on the kinetics of montmorillonite dissolution at 25 °C. Geochim. Cosmochim. Acta 2008, 72, 4224–4253. [Google Scholar] [CrossRef]
  4. Fernández, R.; Mäder, U.K.; Rodríguez, M.; Vigil de la Villa, R.; Cuevas, J. Alteration of compacted bentonite by diffusion of highly alkaline solutions. Eur. J. Mineral. 2009, 21, 725–735. [Google Scholar] [CrossRef]
  5. Moyce, E.B.A.; Roschelle, C.; Morris, K.; Milodowski, A.E.; Chen, X.; Thornton, S.; Small, J.S.; Shaw, S. Rock alteration in alkaline cement waters over 15 years and its relevance to the geological disposal of nuclear waste. Appl. Geochem. 2014, 50, 91–105. [Google Scholar] [CrossRef]
  6. Mäder, U.; Jenni, A.; Lerouge, C.; Gaboreau, S.; Miyoshi, S.; Kimura, Y.; Cloet, V.; Fukaya, M.; Claret, F.; Otake, T.; et al. 5-year chemico-physical evolution of concrete–claystone interfaces, Mont Terri rock laboratory (Switzerland). Swiss J. Geosci. 2017, 110, 307–327. [Google Scholar] [CrossRef]
  7. Taubald, H.; Bauer, A.; Schäfer, T.; Geckeis, H.; Satir, M.; Kim, J.I. Experimental investigation of the effect of high-pH solutions on the Opalinus Shale and the Hammerschmiede Smectite. Clay Miner. 2000, 35, 515–524. [Google Scholar] [CrossRef]
  8. González-Santamaría, D.E.; Fernández, R.; Ruiz, A.I.; Ortega, A.; Cuevas, J. High-pH/low pH ordinary Portland cement mortars impacts on compacted bentonite surfaces: Application to clay barriers performance. Appl. Clay Sci. 2020, 193, 105672. [Google Scholar] [CrossRef]
  9. Savage, D.; Walker, C.; Arthur, R.; Rochelle, C.; Oda, C.; Takase, H. Alteration of bentonite by hyperalkaline fluids: A review of the role of secondary minerals. Phys. Chem. Earth 2007, 32, 287–297. [Google Scholar] [CrossRef]
  10. García Calvo, J.L.; Hidalgo, A.; Alonso, C.; Fernández Luco, L. Development of low-pH cementitious materials for HLRW repositories: Resistance against ground waters aggression. Cem. Concr. Res. 2010, 40, 1290–1297. [Google Scholar] [CrossRef]
  11. Lothenbach, B.; Le Saout, G.; Ben Haha, M.; Figi, R.; Wieland, E. Hydration of a low-alkali CEM III/B–SiO2 cement (LAC). Cem. Concr. Res. 2012, 42, 410–423. [Google Scholar] [CrossRef]
  12. Lothenbach, B.; Rentsch, D.; Wieland, E. Hydration of a silica fume blended low-alkali shotcrete cement. Phys. Chem. Earth 2014, 70, 3–16. [Google Scholar] [CrossRef]
  13. Mihara, M.; Iriya, K.; Torii, K. Development of low–alkaline cement using pozzolans for geological disposal of long–lived radioactive waste. Doboku Gakkai Ronbunshuu F 2008, 64, 92–103. [Google Scholar] [CrossRef]
  14. METI. Nationwide Map of “Scientific Features” Relevant for Geological Disposal. Available online: https://www.enecho.meti.go.jp/en/category/electricity_and_gas/nuclear/rwm/pdf/map_en.pdf (accessed on 30 January 2024).
  15. De Weerdt, K.; Justnes, H. The effect of sea water on the phase assemblage of hydrated cement paste. Cem. Concr. Compos. 2015, 55, 215–222. [Google Scholar] [CrossRef]
  16. Jakobsen, U.H.; De Weerdt, K.; Geiker, M.R. Elemental zonation in marine concrete. Cem. Concr. Res. 2016, 85, 12–27. [Google Scholar] [CrossRef]
  17. Yi, Y.; Zhu, D.; Guo, S.; Zhang, Z.; Shi, C. A review on the deterioration and approaches to enhance the durability of concrete in the marine environment. Cem. Concr. Compos. 2020, 113, 103695. [Google Scholar] [CrossRef]
  18. Ting, M.Z.Y.; Yi, Y. Durability of cementitious materials in seawater environment: A review on chemical interactions, hardened-state properties and environmental factors. Constr. Build. Mater. 2023, 367, 130224. [Google Scholar] [CrossRef]
  19. Qu, F.; Li, W.; Dong, W.; Tam, V.W.Y.; Yu, T. Durability deterioration of concrete under marine environment from material to structure: A critical review. J. Build. Eng. 2021, 35, 102074. [Google Scholar] [CrossRef]
  20. De Weerdt, K.; Bernard, E.; Kunther, W.; Pedersen, M.T.; Lothenbach, B. Phase changes in cementitious materials exposed to saline solutions. Cem. Concr. Res. 2023, 165, 107071. [Google Scholar] [CrossRef]
  21. Fernández-Jiménez, A.; Palomo, A. Characterisation of fly ashes. Potential reactivity as alkaline cements. Fuel 2003, 82, 2259–2265. [Google Scholar] [CrossRef]
  22. Oey, T.; Timmons, J.; Stutzman, P.; Bullard, J.W.; Balonis, M.; Bauchy, M.; Sant, G. An improved basis for characterizing the suitability of fly ash as a cement replacement agent. J. Am. Ceram. Soc. 2017, 100, 4785–4800. [Google Scholar] [CrossRef]
  23. Kuenzel, C.; Ranjbar, N. Dissolution mechanism of fly ash to quantify the reactive aluminosilicates in geopolymerisation. Res. Conse. Recycl. 2019, 150, 104421. [Google Scholar] [CrossRef]
  24. JIS R 5204; Chemical Analysis Method of Cement by X-ray Fluorescence. Japanese Standards Association (JSA): Tokyo, Japan, 2019.
  25. JIS A 6201; Fly Ash for Use in Concrete. Japanese Standards Association (JSA): Tokyo, Japan, 2015.
  26. JIS A 6207; Silica Fume for Use in Concrete. Japanese Standards Association (JSA): Tokyo, Japan, 2016.
  27. Lothenbach, B.; Kulik, D.A.; Matschei, T.; Balonis, M.; Baquerizo, L.; Dilnesa, B.; Miron, G.D.; Myers, R.J. Cemdata18: A chemical thermodynamic database for hydrated Portland cements and alkali-activated materials. Cem. Concr. Res. 2019, 115, 472–506. [Google Scholar] [CrossRef]
  28. Nied, D.; Enemark-Rasmussen, K.; L’Hopital, E.; Skibsted, E.; Lothenbach, B. Properties of magnesium silicate hydrates (M-S-H). Cem. Concr. Res. 2016, 79, 323–332. [Google Scholar] [CrossRef]
  29. Rozov, K.B.; Berner, U.; Kulik, D.A.; Diamond, L.W. Solubility and Thermodynamic Properties of Carbonate-Bearing Hydrotalcite—Pyroaurite Solid Solutions with A 3:1 Mg/(Al+Fe) Mole Ratio. Clays Clay Miner. 2011, 59, 215–232. [Google Scholar] [CrossRef]
  30. Zuo, Y. Thermodynamic modeling of the phase evolution in alkali-activated slag cements with sulfate salt exposure. J. Am. Ceram. Soc. 2022, 105, 7658–7675. [Google Scholar] [CrossRef]
  31. Dilnesa, B.Z.; Lothenbach, B.; Renaudin, G.; Wichser, A.; Kulik, D. Synthesis and characterization of hydrogarnet Ca3(AlxFe1−x)2(SiO4)y(OH)4(3−y). Cem. Concr. Res. 2014, 59, 96–111. [Google Scholar] [CrossRef]
  32. Cavani, F.; Trifirò, F.; Vaccari, A. Hydrotalcite-type anionic clays: Preparation, properties and applications. Catal. Today 1991, 11, 173–301. [Google Scholar] [CrossRef]
  33. Morimoto, K.; Anraku, S.; Hoshino, J.; Yoneda, T.; Sato, T. Surface complexation reactions of inorganic anions on hydrotalcite-like compounds. J. Colloid Interface Sci. 2012, 384, 99–104. [Google Scholar] [CrossRef]
  34. Berner, R.A. The role of magnesium in the crystal growth of calcite and aragonite from sea water. Geochim. Cosmochim. Acta 1975, 39, 489–504. [Google Scholar] [CrossRef]
  35. Buenfeld, N.R.; Newman, J.B. The development and stability of surface layers on concrete exposed to sea-water. Cem. Concr. Res. 1986, 16, 721–732. [Google Scholar] [CrossRef]
  36. Chavagnac, V.; Ceuleneer, G.; Monnin, C.; Lansac, B.; Hoareau, G.; Boulart, C. Mineralogical assemblages forming at hyperalkaline warm springs hosted on ultramafic rocks: A case study of Oman and Ligurian ophiolites. Geochem. Geophys. Geosyst. 2013, 14, 2474–2495. [Google Scholar] [CrossRef]
  37. Díaz-Hernández, J.L.; Sánchez-Navas, A.; Reyes, E. Isotopic evidence for dolomite formation in soils. Chem. Geol. 2013, 347, 20–33. [Google Scholar] [CrossRef]
  38. Nyirő-Kósa, I.; Rostási, Á.; Bereczk-Tompa, É.; Cora, I.; Koblar, M.; Kovács, A.; Pósfai, M. Nucleation and growth of Mg-bearing calcite in a shallow, calcareous lake. Earth Planet. Sci. Lett. 2018, 496, 20–28. [Google Scholar] [CrossRef]
  39. Molnár, Z.; Pekker, P.; Dódony, I.; Pósfai, M. Clay minerals affect calcium (magnesium) carbonate precipitation and aging. Earth Planet. Sci. Lett. 2021, 567, 116971. [Google Scholar] [CrossRef]
  40. Tosca, N.J.; Masterson, A.L. Chemical controls on incipient Mg-silicate crystallization at 25 °C: Implications for early and late diagenesis. Clay Miner. 2014, 49, 165–194. [Google Scholar] [CrossRef]
  41. Roosz, C.; Grangeon, S.; Blanc, P.; Montouillout, V.; Lothenbach, B.; Henocq, P.; Giffaut, E.; Vieillard, P.; Gaboreau, S. Crystal structure of magnesium silicate hydrates (M-S-H): The relation with 2:1 Mg–Si phyllosilicates. Cem. Concr. Res. 2015, 73, 228–237. [Google Scholar] [CrossRef]
  42. Engelhardt, G.; Michel, D. High-Resoluion Solid-State NMR of Silicates and Zeolites; John Wiley & Sons: New York, NY, USA, 1987. [Google Scholar]
  43. Bernard, E.; Lothenbach, B.; Le Goff, F.; Pochard, I.; Dauzères, A. Effect of magnesium on calcium silicate hydrate (C-S-H). Cem. Concr. Res. 2017, 97, 61–72. [Google Scholar] [CrossRef]
  44. Bernard, E.; Lothenbach, B.; Cau-Dit-Coumes, C.; Pochard, I.; Rentsch, D. Aluminum incorporation into magnesium silicate hydrate (M-S-H). Cem. Concr. Res. 2020, 128, 105931. [Google Scholar] [CrossRef]
  45. Vespa, M.; Lothenbach, B.; Dähn, R.; Huthwelker, T.; Wieland, E. Characterisation of magnesium silicate hydrate phases (M-S-H): A combined approach using synchrotron-based absorption-spectroscopy and ab initio calculations. Cem. Concr. Res. 2018, 109, 175–183. [Google Scholar] [CrossRef]
  46. Nishiki, Y.; Cama, J.; Otake, T.; Kikuchi, R.; Shimbashi, M.; Sato, T. Formation of magnesium silicate hydrate (M-S-H) at pH 10 and 50 °C in open-flow systems. Appl. Geochem. 2023, 148, 105544. [Google Scholar] [CrossRef]
  47. Inoue, K.; Nishiki, Y.; Fukushi, K.; Suma, R.; Sato, T.; Sakuma, H.; Tamura, K.; Yokoyama, S.; Shimbashi, M.; Mizukami, T.; et al. Systematic comparison of Mg K-edge XANES spectra of magnesium-bearing clay minerals and magnesium silicate hydrates: A promising tool for identifying magnesium silicate hydrate in natural samples. Appl. Clay Sci. 2023, 245, 107152. [Google Scholar] [CrossRef]
  48. Berner, U.R. Evolution of pore water chemistry during degradation of cement in a radioactive waste repository environment. Waste Manag. 1992, 12, 201–219. [Google Scholar] [CrossRef]
  49. Ganjian, E.; Pouya, H.S. Effect of magnesium and sulfate ions on durability of silica fume blended mixes exposed to the seawater tidal zone. Cem. Concr. Res. 2005, 35, 1332–1343. [Google Scholar] [CrossRef]
  50. Ganjian, E.; Pouya, H.S. The effect of Persian Gulf tidal zone exposure on durability of mixes containing silica fume and blast furnace slag. Constr. Build. Mater. 2009, 23, 644–652. [Google Scholar] [CrossRef]
  51. Lothenbach, B.; Scrivener, K.; Hooton, R.D. Supplementary cementitious materials. Cem. Concr. Res. 2011, 41, 1244–1256. [Google Scholar] [CrossRef]
  52. Thomas, M.D.A.; Hooton, R.D.; Scott, A.; Zibara, H. The effect of supplementary cementitious materials on chloride binding in hardened cement paste. Cem. Concr. Res. 2012, 42, 1–7. [Google Scholar] [CrossRef]
  53. Roosz, C.; Vieillard, P.; Blanc, P.; Gaboreau, S.; Gailhanou, H.; Braithwaite, D.; Montouillout, V.; Denoyel, R.; Henocq, P.; Madé, B. Thermodynamic properties of C-S-H, C-A-S-H and M-S-H phases: Results from direct measurements and predictive modelling. Appl. Geochem. 2018, 92, 140–156. [Google Scholar] [CrossRef]
  54. Shimbashi, M.; Sato, T.; Yamakawa, M.; Fujii, N.; Otake, T. Formation of Fe- and Mg-Rich Smectite under Hyperalkaline Conditions at Narra in Palawan, the Philippines. Minerals 2018, 8, 155. [Google Scholar] [CrossRef]
  55. Nishiki, Y.; Sato, T.; Katoh, T.; Otake, T.; Kikuchi, R. Percipitation of magnesium silicatehydrates in natural alkaline surface environments. Clay Sci. 2020, 24, 1–13. [Google Scholar] [CrossRef]
  56. Shimbashi, M.; Yokoyama, S.; Kikuchi, R.; Otake, T.; Sato, T. Characteristics and Formation Pathways of Iron- and Magnesium-Silicate-Hydrates and Smectites Under Natural Alkaline Conditions. Clays Clay Miner. 2022, 70, 492–513. [Google Scholar] [CrossRef]
  57. Kulik, D.A.; Miron, G.D.; Lothenbach, B. A structurally-consistent CASH+ sublattice solid solution model for fully hydrated C-S-H phases: Thermodynamic basis, methods, and Ca-Si-H2O core sub-model. Cem. Concr. Res. 2022, 151, 106585. [Google Scholar] [CrossRef]
  58. Myers, R.J.; Bernal, S.A.; Provis, J.L. A thermodynamic model for C-(N-)A-S-H gel: CNASH_ss. Derivation and validation. Cem. Concr. Res. 2014, 66, 27–47. [Google Scholar] [CrossRef]
  59. Haas, J.; Nonat, A. From C–S–H to C–A–S–H: Experimental study and thermodynamic modelling. Cem. Concr. Res. 2015, 68, 124–138. [Google Scholar] [CrossRef]
  60. Myers, R.J.; Lothenbach, B.; Bernal, S.A.; Provis, J.L. Thermodynamic modelling of alkali-activated slag cements. Appl. Geochem. 2015, 61, 233–247. [Google Scholar] [CrossRef]
  61. Bernard, E.; Zucha, W.J.; Lothenbach, B.; Mäder, U. Stability of hydrotalcite (Mg-Al layered double hydroxide) in presence of different anions. Cem. Concr. Res. 2022, 152, 106674. [Google Scholar] [CrossRef]
  62. Miyata, S. Anion-Exchange Properties of Hydrotalcite-Like Compounds. Clays Clay Miner. 1983, 31, 305–311. [Google Scholar] [CrossRef]
  63. Dauzeres, A.; Achiedo, G.; Nied, D.; Bernard, E.; Alahrache, S.; Lothenbach, B. Magnesium perturbation in low-pH concretes placed in clayey environment—Solid characterizations and modeling. Cem. Concr. Res. 2016, 79, 137–150. [Google Scholar] [CrossRef]
  64. Santhanam, M.; Cohen, M.; Olek, J. Differentiating seawater and groundwater sulfate attack in Portland cement mortars. Cem. Concr. Res. 2006, 36, 2132–2137. [Google Scholar] [CrossRef]
  65. El-Khoury, M.; Roziere, E.; Grondin, F.; Cortas, R.; Chehade, F.H. Experimental evaluation of the effect of cement type and seawater salinity on concrete offshore structures. Constr. Build. Mater. 2022, 322, 126471. [Google Scholar] [CrossRef]
  66. Schwotzer, M.; Konno, K.; Kaltenbach, J.; Gerdes, A. The rim zone of cement-based materials—Barrier or fast lane for chemical degradation? Cem. Concr. Compos. 2016, 74, 236–243. [Google Scholar] [CrossRef]
Minerals 14 00285 g001
Figure 1. XRD patterns of cement pastes before and after seawater immersion: (a) OPC pastes; (b) low alkaline cement (LAC) pastes and FA. P: portlandite; CSH: calcium silicate hydrate (C-S-H); E: ettringite; B: brucite; LDH: layered double hydroxide; C: calcite; A: aragonite; G: gypsum; MSH: M-S-H; Q: quartz; Mu: mullite, Mt: magnetite.
Figure 1. XRD patterns of cement pastes before and after seawater immersion: (a) OPC pastes; (b) low alkaline cement (LAC) pastes and FA. P: portlandite; CSH: calcium silicate hydrate (C-S-H); E: ettringite; B: brucite; LDH: layered double hydroxide; C: calcite; A: aragonite; G: gypsum; MSH: M-S-H; Q: quartz; Mu: mullite, Mt: magnetite.
Minerals 14 00285 g001
Minerals 14 00285 g002
Figure 2. 29Si MAS NMR spectra of cement pastes before (Initial) and after (L/S = 100, 1000) seawater immersion: (a) OPC pastes; (b) LAC pastes.
Figure 2. 29Si MAS NMR spectra of cement pastes before (Initial) and after (L/S = 100, 1000) seawater immersion: (a) OPC pastes; (b) LAC pastes.
Minerals 14 00285 g002
Minerals 14 00285 g003
Figure 3. 27Al MAS NMR spectra of cement pastes before (Initial) and after (L/S = 100, 1000) seawater immersion: (a) OPC pastes; (b) LAC pastes.
Figure 3. 27Al MAS NMR spectra of cement pastes before (Initial) and after (L/S = 100, 1000) seawater immersion: (a) OPC pastes; (b) LAC pastes.
Minerals 14 00285 g003
Minerals 14 00285 g004
Figure 4. Raman spectra of cement pastes before and after seawater immersion: (a) OPC pastes; (b) LAC pastes. P: portlandite; CSH: C-S-H; B: brucite; LDH: layered double hydroxide; C: calcite; A: aragonite; G: gypsum; MSH: M-S-H; Q: quartz; SF: silica fume.
Figure 4. Raman spectra of cement pastes before and after seawater immersion: (a) OPC pastes; (b) LAC pastes. P: portlandite; CSH: C-S-H; B: brucite; LDH: layered double hydroxide; C: calcite; A: aragonite; G: gypsum; MSH: M-S-H; Q: quartz; SF: silica fume.
Minerals 14 00285 g004
Minerals 14 00285 g005
Figure 5. Dissolved ion concentrations and pH of the solutions from OPC and LAC experiments at different L/S ratios: (a) pH; (b) Ca; (c) Si; (d) Al; (e) Mg; (f) SO42−. The results below the lower limit of quantification (LOQ) are not plotted.
Figure 5. Dissolved ion concentrations and pH of the solutions from OPC and LAC experiments at different L/S ratios: (a) pH; (b) Ca; (c) Si; (d) Al; (e) Mg; (f) SO42−. The results below the lower limit of quantification (LOQ) are not plotted.
Minerals 14 00285 g005
Minerals 14 00285 g006
Figure 6. Solution data at each L/S ratio in phase diagrams for (a) the Ca-Si-H2O system and (b) the Mg-Si-H2O system at 60 °C. C0.8SH: C-S-H (Ca/Si = 0.8); C1.2SH: C-S-H (Ca/Si = 1.2); C1.6SH: C-S-H (Ca/Si = 1.6); M0.75SH: M-S-H (Mg/Si = 0.75); M1.5SH: M-S-H (Mg/Si = 1.5). For graphical reasons, only three C-S-H data points with different Ca/Si ratios are shown in (a).
Figure 6. Solution data at each L/S ratio in phase diagrams for (a) the Ca-Si-H2O system and (b) the Mg-Si-H2O system at 60 °C. C0.8SH: C-S-H (Ca/Si = 0.8); C1.2SH: C-S-H (Ca/Si = 1.2); C1.6SH: C-S-H (Ca/Si = 1.6); M0.75SH: M-S-H (Mg/Si = 0.75); M1.5SH: M-S-H (Mg/Si = 1.5). For graphical reasons, only three C-S-H data points with different Ca/Si ratios are shown in (a).
Minerals 14 00285 g006
Minerals 14 00285 g007
Figure 7. Mineral transition of cement pastes by seawater.
Figure 7. Mineral transition of cement pastes by seawater.
Minerals 14 00285 g007
Minerals 14 00285 g008
Figure 8. Comparison of calculated and experimental results for the pH and aqueous concentration of major elements (Ca, Si, Mg, Al, Fe) after seawater immersion at 60 °C: (a) pH; (b) Ca; (c) Si; (d) Mg; (e) Al; (f) Fe. Experimental results below the lower limit of quantification (LOQ) are not plotted.
Figure 8. Comparison of calculated and experimental results for the pH and aqueous concentration of major elements (Ca, Si, Mg, Al, Fe) after seawater immersion at 60 °C: (a) pH; (b) Ca; (c) Si; (d) Mg; (e) Al; (f) Fe. Experimental results below the lower limit of quantification (LOQ) are not plotted.
Minerals 14 00285 g008
Minerals 14 00285 g009
Figure 9. Mineral composition calculated by thermodynamic calculations under the same conditions as the seawater immersion experiment at 60 °C: (a) OPC pastes; (b) LAC pastes. C3AFS0.84H4.32: siliceous hydrogarnet; C1.6A0.01SH: C-A-S-H (Ca/Si = 1.6, Al/Si = 0.01); C1.6SH: C-S-H (Ca/Si = 1.6); C1.5SH: C-S-H (Ca/Si = 1.5); C1.1SH: C-S-H (Ca/Si = 1.1); C0.8SH: C-S-H (Ca/Si = 0.8), C0.7A0.025SH: C-A-S-H (Ca/Si = 0.7, Al/Si = 0.025); M8AsH13: Mg-Al-SO4 LDH (Mg/Al = 4.0); M4AsH9: Mg-Al-SO4 LDH (Mg/Al = 2.0); M1.5SH: M-S-H (Mg/Si = 1.5); M0.75SH: M-S-H (Mg/Si = 0.75).
Figure 9. Mineral composition calculated by thermodynamic calculations under the same conditions as the seawater immersion experiment at 60 °C: (a) OPC pastes; (b) LAC pastes. C3AFS0.84H4.32: siliceous hydrogarnet; C1.6A0.01SH: C-A-S-H (Ca/Si = 1.6, Al/Si = 0.01); C1.6SH: C-S-H (Ca/Si = 1.6); C1.5SH: C-S-H (Ca/Si = 1.5); C1.1SH: C-S-H (Ca/Si = 1.1); C0.8SH: C-S-H (Ca/Si = 0.8), C0.7A0.025SH: C-A-S-H (Ca/Si = 0.7, Al/Si = 0.025); M8AsH13: Mg-Al-SO4 LDH (Mg/Al = 4.0); M4AsH9: Mg-Al-SO4 LDH (Mg/Al = 2.0); M1.5SH: M-S-H (Mg/Si = 1.5); M0.75SH: M-S-H (Mg/Si = 0.75).
Minerals 14 00285 g009
Table 1. Chemical composition of starting materials.
Table 1. Chemical composition of starting materials.
ComponentComposition (wt.%)
CaOSiO2Al2O3Fe2O3MgONa2OK2OSO3L.O.I 1
Ordinary Portland cement
(OPC)		64.9721.664.763.181.080.340.432.040.97
Silica fume (SF)0.3395.840.520.170.430.140.700.090.80
Fly ash (FA)4.7055.0624.306.421.991.431.250.831.70
1 L.O.I is loss of ignition at 950 °C for 1 h.
Table 2. Mineral composition of FA.
Table 2. Mineral composition of FA.
MineralComposition (Weight %)
Quartz (SiO2)5.52
Mullite (3Al2O3·2SiO2)10.77
Magnetite (Fe3O4)1.12
Glass (4.8SiO2·Al2O3) 182.59
1 The composition of glass was estimated by subtracting the crystalline components (quartz, mullite, magnetite) from the bulk FA chemical composition.
Table 3. Concentration of artificial seawater (Aquamarine).
Table 3. Concentration of artificial seawater (Aquamarine).
ComponentConcentration (mol/L)
Na+4.5 × 10−1
Mg2+5.5 × 10−2
Ca2+1.0 × 10−2
K+1.0 × 10−2
Sr2+1.6 × 10−4
Cl−5.6 × 10−1
SO42−2.9 × 10−2
HCO3−2.4 × 10−3
Br−8.4 × 10−4
BO33−4.4 × 10−4
F−7.1 × 10−5
Table 4. Thermodynamic data of magnesium silicate hydrate (M-S-H), layered double hydroxide (LDH), and siliceous hydrogarnet.
Table 4. Thermodynamic data of magnesium silicate hydrate (M-S-H), layered double hydroxide (LDH), and siliceous hydrogarnet.
PhaseReactionLog10 Kso at 60 °CSources
M-S-H
M0.75SHMg1.5Si2O5.5(H2O)2.5 + 3H+ = 1.5Mg2+ + 2H4SiO411.76[28]
M1.5SHMg1.5SiO3.5(H2O)2.5 + 3H+ = 1.5Mg2+ + H4SiO4 + 2H2O16.23[28]
Mg-Al/Fe-CO3 LDH
Mg2Alc0.5OHMg2Al(OH)6(CO3)0.5(H2O)2 + 6.5H+ = 2Mg2+ + Al3+ + 0.5HCO3− + 8H2O28.72[29]
Mg2Fec0.5OHMg2Fe(OH)6(CO3)0.5(H2O)2 + 6.5H+ = 2Mg2+ + Fe3+ + 0.5HCO3− + 8H2O27.70[29]
Mg3Alc0.5OHMg3Al(OH)8(CO3)0.5(H2O)2.5 + 8.5H+ = 3Mg2+ + Al3+ + 0.5HCO3− + 10.5H2O43.43[29]
Mg3Fec0.5OHMg3Fe(OH)8(CO3)0.5(H2O)2.5 + 8.5H+ = 3Mg2+ + Fe3+ + 0.5HCO3− + 10.5H2O42.19[29]
Mg-Al-SO4 LDH
M4AsH9Mg4Al2(OH)12SO4(H2O)3 + 12H+ = 4Mg2+ + 2Al3+ + SO42− + 15H2O42.91[30]
M6AsH11Mg6Al2(OH)16SO4(H2O)3 + 16H+ = 6Mg2+ + 2Al3+ + SO42− + 19H2O72.25[30]
M8AsH13Mg8Al2(OH)20SO4(H2O)3 + 20H+ = 8Mg2+ + 2Al3+ + SO42− + 23H2O101.69[30]
Siliceous hydrogarnet
C3AFS0.84H4.32(AlFeO3)(Ca3O3(SiO2)0.84(H2O)4.32) + 12H+ = 3Ca2+ + Al3+ + Fe3+ + 0.84 H4SiO4 + 8.64H2O56.49[31]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Kobayashi, Y.; Sato, T. Mineralogical Evolution of High-pH/Low-pH Cement Pastes in Contact with Seawater. Minerals 2024, 14, 285. https://doi.org/10.3390/min14030285

AMA Style

Kobayashi Y, Sato T. Mineralogical Evolution of High-pH/Low-pH Cement Pastes in Contact with Seawater. Minerals. 2024; 14(3):285. https://doi.org/10.3390/min14030285

Chicago/Turabian Style

Kobayashi, Yutaro, and Tsutomu Sato. 2024. "Mineralogical Evolution of High-pH/Low-pH Cement Pastes in Contact with Seawater" Minerals 14, no. 3: 285. https://doi.org/10.3390/min14030285

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop