Introduction

Carbonation process involves the interaction of CO₂-rich fluids with host rocks containing minerals that are rich in reactive divalent cations (e.g., Mg2+, Fe2+, and Ca2+) to form solid carbonates1,2,3,4,5,6,7. This reaction is a remarkable phenomenon occurring in mafic and ultramafic rocks. Carbonate-rich minerals (e.g. magnesite, dolomite) in forms of surface coating and veins were reported in several ophiolites around the globe7,8,9,10 (Fig. 1). Formation of these minerals has played a key role in capturing CO2, thereby maintaining its stable levels in the atmosphere11. The process of CO2 carbonation includes a sequence of dissolution-precipitation reactions where: (i) carbonic acid, formed by the dissolution of CO2, lowers the pH and thereby promotes the dissolution of silicate minerals12,13,14,15; (ii) hydronium ions in the acidic water react with specific minerals, leading to the release of divalent cations into the solution (Fig. 1); (iii) the released cations in turn increase the solution’s alkalinity (commonly referred to as pH swing; Fig. 1) and react with excess CO2 in the solution to form solid carbonate minerals, such as magnesite (MgCO₃), siderite (FeCO₃), and calcite (CaCO₃)16,17,18. Under ambient conditions, the conversion efficiency of silicate minerals to carbonate phases remains generally low, typically ranging between 1% and 4%19. Achieving significantly higher degrees of carbonation necessitates elevated pressure and temperature regimes, which enhance the kinetics of mineral dissolution and subsequent carbonate precipitation17,20,21. For instance, an 84% conversion efficiency of olivine (containing 50% MgO) can be achieved at 185 °C and 65 atm pCO₂, using a particle size fraction of less than 38 μm20.

Fig. 1
figure 1

(a, b) Schematic illustrations highlight combined reactions of dissolution of silicate minerals by acidic (CO2-rich) water, and liberation of metal ions followed by precipitation of metal carbonates and a pronounced swing in pH during these processes (compiled from3,15,22. (c) Field photograph taken by the lead author in the Kempirsay ultramafic massif, Kazakhstan, showing naturally formed carbonate veins within serpentinized harzburgite. These veins provide direct evidence of in situ mineral carbonation processes occurring under ambient geological conditions.

Full size image

Mineral carbonation is thermodynamically favorable process, with a Gibbs free energy change (ΔG) of − 47 kJ/mol when using CO₂ gas and amorphous silica at 60 °C and a pCO₂ of 100 bar23. However, the controls on rate and extent of carbonation process are still poorly constrained. For example, the carbonation of basaltic glass is very sensitive to temperature, and the extent of carbonation at low temperature conditions requires a high fluid/rock ratio24. Indeed, a slight carbonation of olivine-basalt was observed after reacting with aqueous CO₂ at 95 °C and 10 bars for one month25. Similarly, magnesite and calcite were precipitated via reaction of ultramafic rocks with CO₂ dissolved in an alkaline solution at 95 °C and 10 bars for six hours, achieving approximately 85% conversion of olivine to magnesite26. The addition of acidic and alkaline ions commonly accelerates the carbonation of ultramafic minerals, resulting in a rapid carbonation within just a few hours27,28. Additionally, the presence of metal catalysts such as Fe, Ni, and Cr in ultramafic rocks may further speed up the carbonation reaction29. These findings reveal that near ambient carbonation is feasible at certain geochemical conditions in the presence of specific lithologies. For example, serpentinized brucite-rich ultramafic rocks are proven to be more prone to rapid and effective carbonation30,31. However, carbonation experiments at low-temperature conditions (< 50 °C) are very scarce, and very little is known about the impact of initial mineralogy on the carbonation rate.

Most of the published works focused on the reactivity of olivine, serpentine, and pyroxene in ultramafic rocks32,33,34,35,36,37. In contrast, there is a paucity of information on the carbonation of hydrotalcite, a common layered double hydroxide from the anionic clay group, structurally related to brucite and typically forming through the hydrothermal alteration of peridotite38,39,40. Hydrotalcite which has been identified at various ultramafic mine sites offering promising pathways for CO₂ sequestration41,42,43,44. However, hydrotalcite reactivity under ambient conditions is limited and the release of Mg2+ ions is largely dependent on the solubility of Cr3+, Al3+, and Fe3+ trivalent cations43. In this study, we perform direct aqueous carbonation of hydrotalcite-rich partly serpentinized ultramafic rocks at low temperature (~ 40 °C) and low pressure (60 bars) to investigate the differential carbonation of some mineral phases at low temperature and pressure conditions. Results of the present study will define the most potential ultramafic mineral phases for low temperature and pressure CO2 sequestration.

Materials and methods

Material

The peridotite sample utilized in this study was obtained from a chromite mine near Khromtau village, located within the Kempirsay ultramafic massif in Western Kazakhstan. This massif forms part of the Uralian ophiolites and is situated within the Sakmara allochthon of the Central Ural Uplift45,46. The Kempirsay ultramafic massif is dominated by lherzolite at its base. This is overlain by several kilometers of harzburgite, which hosts numerous chromite-rich dunite lenses47. The ultramafic rock mineralogy is listed in Table 1. The rock sample studied is highly-serpentinized harzburgite where serpentine (chrysotile), olivine, and hydrotalcite respectively represent the most abundant phases, while Fe and Al oxides were reported as minor phases (Table 1).

The whole rock elemental composition of the studied rock is illustrated in Table 2. The most abundant elements are Mg and Si, which is consistent with the dominance of serpentine and olivine. On the other hand, relatively high concentrations of Al are attributes to the presence of hydrotalcite. The fourth most abundant element is Fe, normally incorporated in magnetite and olivine.

Analytical methods

Petrography, mineralogy and whole rock geochemistry

Petrographic investigation was conducted on polished thin section prepared from the ultramafic rock sample which was examined using conventional plane polarized microscopy (PPL) as well as scanning electron microscopy (SEM). The latter was carried out at the core facilities department of Nazarbayev University, Kazakhstan using JSM-IT200 fully equipped with electron dispersive spectrometer (EDS). SEM measurements were conducted on the gold coated thin section achieving a resolution of 3.0 nm at an accelerating voltage of 30 kV.

Whole rock mineralogy and elemental geochemistry were obtained using X-ray diffraction (XRD) and X-ray fluorescence (XRF) analyses which were conducted on before and after the experiment at the core facilities department of Nazarbayev University, Kazakhstan. XRD analysis was performed on the powdered rock using SmartLAb (Rigaku) equipment following the analytical procedure described by Zhou et al.48. The whole rock mineralogy of the sample was determined before and after the experiment using the PDF 4 + spectra database and the Bruker DIFFRAC.EVA software suite. The relative quantification of identified minerals was based on reference intensity ratio (RIR) (more details in Green et al.49. Finally, the major element composition of the rock sample was obtained through X-ray fluorescence (XRF) which was carried out using Axios Max PANalytical equipment with analytical error ranges from ± 0.5 wt% to ± 2 wt%.

Experimental procedure

Powdered whole rock sample (5 g) was suspended in 100 ml of deionized water. A mixture of 0.64 M NaHCO3 + 1 M NaCl was added into the suspension which was stirred at room temperature prior to adding a universal pH indicator. The addition of NaHCO3 and NaCl enhances the reactivity of ultramafic mineral phases through retarding the formation of Si-rich layer around the reacted and non-reacted grains36. The suspension was moved into a closed PARR batch reactor vessel, and the CO2-water-rock batch experiments were conducted at pCO2 of 60 bars and a fixed temperature of 40 °C for 14 days (Fig. 2). The PARR reactor was equipped with a window to monitor the color changes of the added pH indicator which recorded chemical changes in the aqueous solution. The experiment was stopped multiple times to sample the aqueous solution (5 mL). The sampled solutions were analyzed for major cations using Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES) System - iCAP 6300 (Thermo Scientific) at core facilities department of Nazarbayev University, Kazakhstan. All major cations were measured with an analytical error of ± 0.1%.

Fig. 2
figure 2

A sketch shows the experimental setup for aqueous CO2 reaction with powdered ultramafic rock.

Full size image
Table 1 The XRD bulk mineralogy of the studied ultramafic rock sample before and after the experiment.
Full size table
Table 2 XRF elemental composition of the studied ultramafic rock sample before and after the experiment.
Full size table

Results

Chemistry of the aqueous solution

The pH of the aqueous solution increased progressively from slightly acidic to neutral, alkaline and towards the end of the experiment hyperalkaline (Fig. 3a-d). Rapid increase of concentration of the dissolved Mg over the other cations (e.g. Fe, Si, Na, and K; Fig. 4) presents strong corroborative evidence of increasing alkalinity. A progressive increase in the concentration of all cations after 3 days of the experiment is observed, however, the concentration of dissolved Mg was 10 times greater than other cations (Fig. 4). Changes in the chemistry of the aqueous solution may be divided into three distinct phases. The first phase (0–3 days) was characterized by pH and cation concentrations increase with time. During the second phase (3–7 days) all cations displayed a progressive decrease as pH swung from acidic/neutral to alkaline. The third phase (7–14 days), Mg slightly decreased, whereas Fe, and Si showed a slight increase at the end of the experiment. Notably, Na displayed a complex trend during the third phase with a sharp increase and decrease in the dissolved Na content with increasing pH.

Fig. 3
figure 3

Photographs of batch reactor showing the pH changes throughout the experiment from acidic at the beginning (a), to slightly acidic after 3 days (b), to alkaline after 7 days (c), and to hyper-alkaline after 14 days (d). Precipitation of white crystals at the end of the experiment (d).

Full size image
Fig. 4
figure 4

A cross-plot illustrates the variation in chemistry of the aqueous solution throughout the experiment.

Full size image

Petrography and mineralogy

Olivine in the pristine rock ranges from fresh to partly altered (Fig. 5a). Vein and mesh-textured serpentine are the most common phase (Fig. 5a). Olivine and other phases (e.g., chromite and magnetite) are embedded in a serpentinized matrix. Chromite occurs in the form of euhedral crystals (Fig. 5b), and has a composition dominated by Cr, Fe, and Al with traces of Si (Table 3). Olivine is a forsterite variety with Mg/Fe ratio > 5 (Fig. 5c; Table 3). Serpentine minerals are represented by Mg-rich chrysotile which contains significant contents of Fe (~9%; Fig. 5d). Hydrotalcite emerges as a fibrous phase in the veins, normally in association with chrysotile (Fig. 5d). Hydrotalcite consists mainly of Mg and Al with significant content of Fe and Si (Table 3).

Fig. 5
figure 5

Thin section and SEM microphotographs with EDS results demonstrate the textural and elemental composition of different mineral phases encountered in the studied serpentinized harzburgite rock.

Full size image

Fe and Si traces in hydrotalcite may point to its formation as alteration product of both serpentine and olivine. The XRD results show the complete disappearance of some phases and emergence of others after reaction with aqueous CO2 (Fig. 6). Hydrotalcite had completely disappeared, whereas nesquehonite and dolomite had emerged as newly formed phases (Fig. 6). Notably, nearly 75% of both chrysotile and olivine were carbonated after the end of the experiment (Table 2). The reduction of Mg, Fe, and Si contents of chrysotile and olivine after reaction confirms dissolution of these mineral phases (Table 3).

Total content of Fe-oxides decreased by about 50% which is in line with their expected dissolution- reprecipitation and likely contribution to the formation of Fe-bearing carbonates upon carbonatization. Analysis of reaction byproducts revealed the emergence of elongated, needle-like crystals consist mainly of nesquehonite (hydrated Mg-carbonate) (Fig. 7). Nesquehonite contains traces of Fe (~ 4%), Si (~ 6%), and Al (~ 2%) inherited from its olivine, chrysotile, and hydrotalcite precursors.

Table 3 Elemental composition (EDS results) of the main mineral phases encountered in the studied peridotite sample before and after experimental reaction with aqueous CO2.
Full size table
Fig. 6
figure 6

XRD patterns of the studied serpentinized harzburgite (a) showing the variation in its mineralogy before, after experimental reaction with aqueous CO2 (b). Mineral abbreviations: chrysotile (chr); hydrotalcite (htc); olivine (ol), magnetite (mag); dolomite (dol).

Full size image
Fig. 7
figure 7

SEM-BSE imagery and EDS data highlighting the morphology and phase chemistry of the newly formed nesquehonite.

Full size image

Discussion

Phase transformations and CO2 carbonation

Reaction of aqueous CO2 with ultramafic rock sample under low temperature and pressure conditions resulted in significant transformations in mineralogy and fluid chemistry indicative of efficient CO2 mineralization. The pH measurements show a rapid increase in pH values with a prominent increase in the content of divalent cations in the solution (Figs. 3 and 4). The pH and concentration of divalent cations clearly demonstrate rapid dissolution of the studied ultramafic rock and release of cations into the solution. This has been confirmed by the notable decrease in Mg content in olivine and chrysotile phases after the end of the experiment (Table 3).

The obvious drop in the concentration of divalent cations after 3 days of experiment highlights efficient sequestration of CO2 and formation of solid carbonates, such as nesquehonite. This is consistent with the acid-base dynamics and pH swing reported in most aqueous CO2 reactions with mafic and ultramafic rocks25,40,41,42,43,44,45,46,47,48,49,50,51 (Fig. 1). The pH swing observed in our experiment aligns with the theoretical expectations of dissolution of reactive minerals and subsequent mineral trapping of CO2 (Table 3). However, the observed pH swing is faster and more progressive compared to the trends reported for the reactions of CO2 with both basalt52,53 and olivine25. This observation suggests that the studied ultramafic rock sample serves as a highly reactive medium for CO₂ sequestration, exhibiting even greater reactivity than pure olivine grains.

XRF and XRD analyses provide insight into the elemental and mineralogical changes in the ultramafic rock sample studied before and after the experiment. The analyzed ultramafic rock is rich in Mg and displays OPE (olivine-pyroxene elemental ratio, Mg + Fetot + Ca)/Si value of 1.49 suggesting uncarbonated harzburgitic composition54. Exposure to CO2 results in a notable decrease in Si, Fe, Mg, and Al (Table 2), suggesting dissolution of olivine, serpentine, and hydrotalcite. The OPE of the reacted sample dropped to 1.2, evidence of a significant dissolution and carbonation of the mafic mineral phases. Deschamps et al.55 reported that the changes in major-element composition of ultramafic rocks during either serpentinization or carbonation are generally attributed to dilution by volatiles.

The most significant geochemical transformation is the prominent two-fold increase in the loss of ignition (LOI) (Table 2), indicative of enrichment of chemically bound volatiles after exposure to aqueous CO2. These volatiles are primarily in the form of CO₂ within newly formed carbonate phases, i.e. nesquehonite. Cutts et al.54 utilized LOI as a proxy for hydration and carbonation to report a significant increase in LOI in the serpentinized and carbonated ultramafic rocks. Accordingly, the significant increase in volatile content provides evidence for efficient CO2 sequestration and formation of carbonates. The XRD and EDS results confirm the mineralogical and geochemical changes in the post-exposure sample (Tables 1 and 3). The results indicated a reduction in chrysotile, hydrotalcite, and olivine, which are rich in divalent cations such as Mg2+ and Fe2+, typifying their consumption during CO2 sequestration and formation of carbonates. This is supported by the complete replacement of hydrotalcite by nesquehonite, as well as the 75% reduction of olivine and chrysotile that were replaced by nesquehonite and dolomite (Table 1).

Comparison with literature

Over the past few decades, the CO2 mineral trapping studies and experiments were focused on either basalt24,51,55,56,57 or isolated olivine and pyroxene grains35,36,46,58. However, very few studies have been performed on ultramafic rocks (Table 4), and none of these studies examined the carbonation of LDH-rich ultramafic lithotypes. The formation of carbonate minerals in the form of natural alteration products in mafic and ultramafic ophiolite rocks has been documented22,58,59,60. Moreover, in the reaction of aqueous CO2 with harzburgite-rich ophiolite rocks at room temperature and atmospheric pressure conditions, Gill et al.61 reported a slight increase in dissolved total inorganic carbon without precipitation of carbonate phases. Rigopoulos et al.34 observed precipitation of minor aragonite after two months of aqueous CO2 reaction with ball-milled serpentinized harzburgite and dunite under ambient conditions. However, no carbonate precipitation was observed when pulverized ultramafic samples were utilized as experimental starting materials.

Low-temperature carbonation experiments carried out by Rigopoulos et al.34 typified the incorporation of released Mg2+ ions in Mg-rich clay minerals rather than carbonates. In contrast, Mg-rich carbonates, such as magnesite and Fe-magnesite, were formed during carbonation experiments carried out at high temperature (> 70 °C) and pressure (70–100 bars)37,62. The current study reports the first low-temperature carbonation of pulverized ultramafic rocks where Mg2+ was incorporated within stable nesquehonite rather than Mg-silicate clay minerals. Similar byproducts have been reported at ultramafic mine sites, where near-surface, low-temperature precipitation of hydrated Mg-carbonates occurs because of reactions with atmospheric CO263,64. The mineral carbonation reactivity presented in this study highlighted that LDH could be more reactive than Mg-olivine and serpentine and could be comparable to the reactivity of brucite30,31. This finding is consistent with the flow-dissolution experiments on synthetic hydrotalcite which have demonstrated rapid reaction kinetics, with up to ~ 80% leaching of Mg2+65. This is largely attributed to the structural properties and dissolution kinetics of these anionic clays, particularly their high surface area and porous, layered arrangement, which facilitate rapid interactions with CO₂-bearing fluids. Furthermore, LDH exhibit faster dissolution rates in mildly acidic conditions66,67,68, thus readily releasing Mg²⁺ ions, which are crucial for carbonate precipitation. In contrast, olivine and serpentine are characterized by strong Si-O bonding, which makes their dissolution significantly slower and limits the availability of Mg2+ ions for carbonation reactions68,69. This is consistent with the experimental findings of Azizi and Larachi70, who reported that the reactive chemical structure of Mg-hydroxides facilitates their rapid dissolution during carbonation. For LDH, their geochemical composition plays a key role in controlling dissolution rates. Lu et al.43 demonstrated that LDH phases incorporating Fe2+ exhibit higher labile Mg2+ content compared to those containing Al2+ or Cr2+. Accordingly, pyroaurite and iowatite are the favorable LDH phases for CO2 carbonation43,44. The observed enhancement in hydrotalcite carbonation under moderately elevated pressure (60 bars) and temperature (40 °C) conditions may indicate a significant influence of experimental parameters on the reactivity of LDH minerals. While hydrotalcite is considered less reactive due to its aluminum content, which can impede ion exchange and structural rearrangements necessary for carbonation43, our findings suggest that specific conditions can markedly improve its carbonation kinetics. However, to establish a comprehensive framework for the reactivity of LDH phases and their utility in CO2 sequestration, further in-depth analyses are necessary. Specifically, studies should encompass a wide range of samples with varying mineralogical compositions to account for the heterogeneity inherent in ultramafic lithologies.

Table 4 Comparison between experimental conditions and findings of this study and other published works on the carbonation of ultramafic rocks.
Full size table

Geochemical reactions and controls on CO2 mineralization in ultramafic rocks

Deciphering the geochemical interactions within the ultramafic rocks is crucial for elucidating the mechanisms of CO2 mineralization and sequestration15,71. Experimental data and mineralogical analyses revealed several reaction pathways which led to the dissolution of Mg-rich mineral phases and precipitation of solid carbonates in the form of nesquehonite.

Mineral dissolution and carbonate precipitation

Olivine, chrysotile and hydrotalcite phases in the studied ultramafic rock sample react with aqueous CO2 to form nesquehonite as follows:

$${\text{Mg}}_{{\text{2}}} {\text{SiO}}_{{\text{4}}} \left( {{\text{olivine}}} \right){\text{ }} + {\text{ 2CO}}_{{\text{2}}} + {\text{ 6H}}_{{\text{2}}} {\text{O }} \to {\text{ 2MgCO}}_{{\text{3}}} \cdot {\text{3H}}_{{\text{2}}} {\text{O }}\left( {{\text{nesquehonite}}} \right){\text{ }} + {\text{ SiO}}_{{\text{2}}} + {\text{ 3H}}_{{\text{2}}} {\text{O}}.$$

This reaction is comprised of three main steps: (1) CO2 dissolution in water generates hydronium ions72; (2) the hydronium ions attack ultramafic minerals resulting in release of Mg2+ ions and formation of silicic acid (H₄SiO4) which either remains in solution or precipitates as amorphous silica36; (3) the released Mg2+ reacts with HCO₃⁻ and carbonate (CO₃²⁻) to form nesquehonite64. Evidence of this reaction is the observed released silica in the solution as well as the notable decrease in cation contents of the olivine, chrysotile phases after the reaction (Table 3).

Thom et al.73 reported dissolution and subsequent carbonation of chrysotile via these three steps-reaction to form hydro-magnesite as follows:

$$\begin{aligned} {\text{Mg}}_{{\text{3}}} {\text{Si}}_{{\text{2}}} {\text{O}}_{{\text{5}}} & \left( {{\text{OH}}} \right)_{{\text{4}}} ~\left( {{\text{chrysotile}}} \right){\text{ }} + {\text{ 2}}.{\text{4H}}^{ + } + {\text{ 2}}.{\text{ 4HCO}}_{{\text{3}}} ^{ - } {\text{ }} \to {\text{ }} \\ & \;\;\;0.{\text{6Mg}}_{{\text{5}}} \left( {{\text{CO}}_{{\text{3}}} } \right)_{{\text{4}}} \left( {{\text{OH}}} \right)_{{\text{2}}} .{\text{ 4H}}_{{\text{2}}} {\text{O }}\left( {{\text{hydromagnesite}}} \right){\text{ }} \\ & \;\;\; + {\text{ 2SiO}}_{{\text{2}}} + {\text{ 1}}.{\text{4H}}_{{\text{2}}} {\text{O}}. \\ \end{aligned}$$

Similarly, in this study, hydrotalcite is completely dissolved after the reaction with aqueous CO2 to form nesquehonite as follows:

$$\begin{aligned} {\text{Mg}}_{{\text{6}}} {\text{Al}}_{{\text{2}}} & \left( {{\text{CO}}_{{\text{3}}} } \right)\left( {{\text{OH}}} \right)_{{{\text{16}}}} \cdot {\text{4H}}_{{\text{2}}} {\text{O }}\left( {{\text{hydrotalcite}}} \right){\text{ }} + {\text{ CO}}_{{\text{2}}} + {\text{ 8H}}^{ + } \to {\text{ }} \\ & \;\;\;{\text{6MgCO}}_{{\text{3}}} \cdot {\text{3H}}_{{\text{2}}} {\text{O }}\left( {{\text{nesquehonite}}} \right){\text{ }} + {\text{ 2Al}}\left( {{\text{OH}}} \right)_{{\text{3}}} \left( {{\text{gibbsite}}} \right) \\ & \;\;\;{\text{ }} + {\text{ H}}_{{\text{2}}} {\text{O}}. \\ \end{aligned}$$

In our system, the observed decrease in Mg, Si, and Fe from chrysotile and olivine, coupled with the emergence of carbonate minerals confirm the carbonation of chrysotile. However, unlike hydromagnesite, nesquehonite was the dominant carbonate product, likely due to the experimental conditions (low temperature and increased HCO3– activity), which thermodynamically favors nesquehonite precipitation over hydromagnesite63,64,74. Additionally, Al3+ is released during hydrotalcite breakdown was likely precipitates as poorly crystalline gibbsite. All reactions demonstrate the onset of precipitation of nesquehonite when the pH swings from acidic to alkaline implying that optimal nesquehonite precipitation occurs at pH values around 8. This agrees with the pH measurements during the experiment which revealed a drop in the magnesium content after 7 days consistent with the acidic-alkaline pH swing. Notably, all reactions except the carbonation of hydrotalcite involved release of silica into the solution, thereby explaining the similar pattern observed for Mg and Si ions during the experiment.

Mineral reactivity and implications for low temperature CO2 mineralization

The documented mineral changes and emergence of new carbonate phases are explained by a reaction pathway shown in the previous section. Gibbsite formation suggests effective removal of Al3+ from solution, which can enhance Mg2+ availability and associated nesquehonite formation (Table 1). Moreover, complete dissolution of hydrotalcite compared to the partial dissolution of olivine and chrysotile strongly suggests that hydrotalcite is the most reactive in a CO2-rich aqueous medium. The pH buffering effect of the NaHCO₃–NaCl solution likely promoted faster dissolution of hydrotalcite compared to other silicate minerals36. In contrast, olivine and serpentine exhibit much slower dissolution rates, particularly at temperatures below 100 °C. Incomplete dissolution of both olivine and chrysotile is likely due to the formation of silica-layer passivation slowing further dissolution36. Also, the weak bound between interlayer carbonate and hydroxyl groups in hydrotalcite facilitates the rapid breakage of Mg–OH layers under acidic conditions resulting in an instantaneous release of Mg2+ ions. However, a detailed kinetic analysis is still required to fully elucidate the interplay between dissolution rates, surface area evolution, and the subsequent formation of carbonate phases in similar ultramafic rock systems. This includes understanding how variables such as pH, ionic strength (e.g., presence of NaCl), and bicarbonate concentration influence the rate-limiting steps in LDH dissolution and carbonation. Clarifying these reaction pathways is critical for assessing the long-term carbonation potential of LDH phases within ultrmafic massifs.

The preferential formation of nesquehonite (MgCO3·3 H2O) over magnesite (MgCO3) is primarily controlled by temperature, hydration state, and kinetic constraints. At low temperatures (< 50 °C), nesquehonite is thermodynamically more stable than magnesite because magnesite formation is sluggish due to slow Mg2+ hydration kinetics. The formation of hydrated Mg-carbonate minerals has been observed in natural environments (e.g., hydromagnesite playas)63,64,74, typifying that these carbonate phases can be a favorable sink for CO2 at specific geological conditions. The hydrated structure of nesquehonite aligns with the hydration state of dissolved Mg2+ at temperatures below 50 °C. At these temperatures, a strong hydration shell forms around Mg2+, preventing direct Mg-O bonding to form magnesite75. Accordingly, this study provides first insights into the long-term low temperature sequestration of CO2 in the form of nesquehonite. Therefore, for engineered storage solutions, adjusting temperature conditions below 50 °C (e.g., geothermal reservoirs or deep formations) could promote nesquehonite formation, which can be stable over geological timescales. However, further experiments under variable conditions are necessary on a range of ultramafic lithologies containing different LDH phases to comprehensively assess their suitability for CO2 mineral trapping. Moreover, time-dependent reactivity analysis should be carried out to investigate the reactivity of LDH compared to Mg-silicates (e.g. serpentine and olivine).

Summary and conclusions

Serpentinized harzburgite from the Kempirsay Massif (western Kazakhstan) was subjected to batch carbonation experiments using aqueous CO2 under controlled low-temperature (40 °C) and pressure (60 bars) conditions for 14 days. The starting material consists predominantly of serpentine (chrysotile), olivine, and layered double hydroxides (LDH), specifically hydrotalcite, collectively accounting for over 95% of the whole-rock mineralogy. During the first week of experiment, the aqueous solution showed a notable increase in pH and alkalinity, accompanied by progressive release and accumulation of Mg, Fe, and Si cations. Afterwards, a marked decrease in Mg2+ concentration was observed, corresponding to the neo-formation of Mg-carbonates, particularly nesquehonite. XRD analyses confirmed significant post-reaction mineralogical transformations. Nesquehonite emerged as the dominant carbonate phase, while the contents of olivine and serpentine were reduced by nearly 75%. Hydrotalcite, in contrast, had completely disappeared, indicating its high reactivity. A twofold increase in loss on ignition (LOI) supports the formation of volatile-rich phases, highlighting the efficiency of CO2 mineralization. Concurrently, a reduction in SiO2 content and the formation of gibbsite point to key reaction pathways involving the dissolution and carbonation of both silicate minerals and hydrotalcite. These results demonstrate that the carbonation efficiency of ultramafic lithologies increases with the abundance of LDH phases. The findings underline the importance of selecting suitable mineralogical compositions and geochemical conditions when designing engineered CO2 storage systems. Notably, this study provides evidence for effective low-temperature CO2 mineralization through the formation of stable nesquehonite and underscores the critical role of LDH as more reactive phases than Mg-rich silicates in such processes. However, to develop a robust framework for engineered CO2 mineralization within ultramafic lithologies, it is imperative to conduct further experimental investigations encompassing a broader range of LDH phases across diverse ultramafic rock compositions. Such studies should focus on elucidating the variability in carbonation kinetics and capacities among different LDH minerals, considering factors such mineralogical diversity and experimental conditions.