NaHCO3 as a carrier of CO2 and its enhancement effect … – Frontiers

1 Introduction

The urgent need to reduce the amount of carbon dioxide (CO2) for creating a carbon neutral society has driven the development of various CO2 capture, utilization, and storage (CCUS) techniques (Voigt et al., 2021). Among the CCUS techniques, mineralization of CO2 to water-insoluble carbonates using Mg or Ca-rich rocks such as basalt (Goldberg et al., 2008; Okoko and Olaka, 2021), peridotite (Oelkers et al., 2008) and wollastonite (Wang et al., 2021) is a promising method for large-scale and long-term CO2 storage with low risk of releasing stored CO2 into the atmosphere (Zhao et al., 2015) and probably a relative acceptable cost (National Academies of Sciences, 2018; Kelemen et al., 2019). Basalt is one of the optimum rocks for CO2 storage because of its chemical composition, abundant distribution and easy availability.

Bạn Đang Xem: NaHCO3 as a carrier of CO2 and its enhancement effect … – Frontiers

CarbFix Project in Iceland is an engineering example for large-scale CO2 mineralization, where 95% of the dissolved CO2 in freshwater is mineralized to carbonates in the basaltic groundwater setting which has a weakly alkaline pH and temperature of 20°C-50°C in 2 years (Alfredsson et al., 2013; Matter et al., 2016; Snæbjörnsdóttir et al., 2020). This successful application is inspiring other CO2 mineralization projects around the world to be developed. However, it has been estimated that 22 tons of water is required to capture 1 ton of pure CO2 for storage according to the operation in CarbFix (Snæbjörnsdóttir et al., 2020). To alleviate water shortage, recently, researchers suggested substituting freshwater with saline water (e.g., seawater) for long-term CO2 storage (Voigt et al., 2021), which makes the understanding of CO2-saline water-basalt interactions even more important.

There have been many studies discussing the mineral alterations in CO2-saturated NaCl saline systems at different pressures and temperatures (Ueda et al., 2021; Edem et al., 2022; Peter et al., 2022), however, many questions regarding reaction mechanisms are yet to be answered. NaCl is the most abundant salt component in saline aquifers, while also contains minor Ca2+, Mg2+, HCO3− and SO42−, etc., Due to the high ionic strength in the CO2-saturated NaCl saline system, the dissolution of silicates minerals to release Ca and Mg ions (Gudbrandsson et al., 2011; Wang et al., 2019; Bratcher et al., 2021), which subsequently react with CO2 to generate carbonate minerals, can be enhanced (Wang and Giammar 2013; Marieni et al., 2020). However, debates on the independent effects of NaCl and dissolved CO2 species (e.g., NaHCO3 in saline water) on CO2 mineralization are still ongoing. Although NaHCO3 is a relatively minor component in saline water, it is the main carrier for CO2 and can directly participate in CO2 mineralization. Gadikota et al. (Gadikota et al., 2014) proposed that NaHCO3 may be a more crucial factor in olivine carbonation promotion than NaCl because the carbonation process significantly influences the surface area and particle size of olivine, which induces further carbonation. Our previous research on hydrothermal olivine alteration also found enhanced olivine dissolution and carbonation in CO2-rich while Cl-free environments (Wang et al., 2019a). However, since the mineral composition of basalt is relatively complex, whether NaHCO3 has a potentially positive role in basaltic rock carbonation has yet to be confirmed.

Another important but less studied aspect of basalt carbonation is related to the carbonate formations under high temperature conditions. Although higher temperatures are favorable for mineral dissolution (Casey and Sposito, 1992; Wang et al., 2019b), the suggested temperatures for basalt carbonation were generally ≤200°C due to the competitive hydrothermal alteration reactions. Among the previous studies, Rosenbauer et al. (Rosenbauer et al., 2012) conducted basalt carbonation experiments using a CO2-rich NaCl brine at 50°C-600°C; however, the maximum carbonation rate occurred at 300°C with ferroan magnesite [(Mg, Fe)CO3] as the main carbonation product. On the other hand, Wolff-Boenisch and Galeczka (Wolff-Boenisch and Galeczka, 2018) reported the carbonation products were calcite and Ca/Mg-carbonates at 90°C in synthetic seawater. Due to these mutually inconsistent results, more studies are needed to establish the mechanism of basalt carbonation especially at high temperatures.

This paper aims to address the above research vacancy by evaluating CO2 storage efficiency and revealing the independent effects of NaHCO3 during basalt alteration in hydrothermal systems which is Cl-free through a series of batch experiments conducted at 600°C-100°C. Basaltic rocks obtained from Iceland were used for reaction. This study experimentally verified the hypothesis using a CO2-rich hydrothermal environment to promote CO2 storage through basalt carbonation and clarified how Na joins water-basalt-CO2 interactions. The study results are expected to quantify the CO2 mineralization potential of using such systems and to provide technical backup for the development of CO2 mineralization systems using basalt or other rocks.

2 Materials and methods

2.1 Materials

The basaltic rock used in this study was obtained from Iceland (MORB), which is one of the most widely studied areas for CO2 storage, the reactivity of these basaltic rocks has been confirmed by other studies (Matter et al., 2016; Snæbjörnsdóttir and Gislason, 2016; Snæbjörnsdóttir et al., 2020) and is thus easier for comparison. Its main mineral compositions were plagioclase (Ca0.7Na0.3Al1.7Si2.3O8), clinopyroxene (Ca0.8Mg0.7Fe0.5Si2O6), olivine (Mg1.6Fe0.4SiO4), glass, Fe-Ti oxides, and Cr oxide, determined using an electron probe microanalyzer (EPMA, JEOL JXA-8200) at Tohoku University. The average chemical composition of the rock is shown in Table 1, quantified via X-ray fluorescence (XRF, Rigaku, ZSX Primus IV) measurement. The basaltic rock was ground and sieved to obtain the <63 µm-sized fraction for experiments. Anorthite (CaAl2Si2O8), an endmember of plagioclase solid solutions, with only a small amount of Na, was also used for experiments in this study to reproduce the albitization process of plagioclase observed in experiments using basalt as well as to quantify the effects of Na in solution on promoting mineral alteration. The anorthite used was obtained from Hokkaido, Japan, with the compositions shown in Table 1. The particles selected for the experiments were relatively coarse, with sizes ranging from 63-125 µm. Sodium bicarbonate (NaHCO3) purchased from Kanto Chemical (Japan) was used to create Na-rich systems and used as the CO2 source for storage.

2.2 Experimental design and measurement

A closed batch reactor with a length of 150 mm, diameter of 10 mm, and internal volume of 13 mL was used in this study for hydrothermal experiments (Figure 1). The reactor body was made of SUS316, with the main compositions of Fe, Ni, Cr, as well as few Mo, Si, Mn, C, P, and S. The reactor was easy to operate and could be put in a muffle furnace (Yamato, FP31) for hydrothermal reactions. The tolerated temperature of the reactor was 600°C and the pressure was 10 MPa. For each experiment, 8 mL solution was injected into the reactor with the filling ratio of around 62%.

The experimental conditions for basalt alteration are shown in Table 2. In Runs 2, 3, 4, and 5, 0.5 mol/L NaHCO3 solution was prepared for basalt alteration, this concentration has been proved to be effective in promoting the carbonation of olivine in our previous study (Wang et al., 2019a). Run 1 used Milli-Q water, the pH of which was also adjusted to weak alkaline, similar to the value of NaHCO3 solution using NaOH solution. Runs 1 and 2 were conducted to verify the positive effects of NaHCO3 on basalt alteration. Runs 2, 3, and 4 were tested to study the effects of reaction temperature while Run 5 was conducted to compare with Run 2 to illustrate the effect of reaction duration. In all the above experiments, 0.3 g basalt powder was added into the reactor tube, followed by addition of 6 mL of 0.5 mol/L NaHCO3 solution prepared using Milli-Q water. The reactor was sealed and gently shaken to mix the solid and fluid, following which it was vertically put into a muffle furnace, whose internal temperature was assumed to quickly increased to a desired value (400°C-700°C); meanwhile, the pressure was increased to water-saturated pressure for reactions (<8.5 MPa). After keeping the reactor at a constant temperature for 5-10 days for reaction, it was taken out from the muffle furnace and cooled to room temperature (∼20°C) within 1 h using a fan. The reaction rate is considered slow at ambient temperatures; thus the reaction is assumed to have stopped. The reactor with solution inside was weighed before and after reaction to ensure that no liquid has leaked out of the reactor and that the reactor was highly gas tight. The solid and liquid samples were then separated for analysis. To eliminate if elements have been released from the autoclaves or not, blank experiments (remarked as 2b, 3b and 4b in Table 2) without rocks were also conducted at each temperature.

Xem Thêm : Nhóm hài Gãy Tv – Người nổi tiếng

All liquid samples taken after the reactions were analyzed via inductively coupled plasma-optical emission spectrometry (ICP-OES; Agilent 5100) to quantify the target elements (e.g., Mg, Al, Si, Ca, Fe). The concentration of Na was not analyzed in this study due to the initial inclusion of a high concentration of NaHCO3. Each fluid sample was measured for three times, showing good reproducibility within a margin of error of 3%. The pH of the suspension was measured at room temperature (around 20°C). The solid samples were washed with Milli-Q water and dried in an oven with a temperature of 60°C for 24 h before mineral composition measurement using X-ray diffraction (XRD; Multiflex, Rigaku, Japan) with Cu Kα radiation (λ = 1.54 Å) operated at 40 kV and 20 mA and with a 2θ step size of 0.02° from 5° to 50°. The surface morphologies of minerals were observed using scanning electron microscopy (SEM; SU-8000, Hitachi, Japan) equipped with energy dispersive spectroscopy. Thermogravimetric (TG) analyses of all reacted minerals were performed using a thermogravimetry (Thermo + + Plus EVO TG 8120, Rigaku, Japan). The temperature was increased from room temperature to 1000°C at a rate of 10°C/min. The carbonate fraction was determined from the weight loss due to CO2 release in a specific temperature range, for instance, calcite was decomposed to release CO2 at around 100-780°C (Wang et al., 2019).

3 Results and discussion

3.1 Effects of NaHCO3 as a function of reaction duration

The solution pH showed no obvious change in basalt-Milli-Q water experiment and in blank experiments without basalt, while it slightly increased after reaction of basalt in NaHCO3 solution (Table 2). This change was considered a result of silicate dissolution (Zhi and Ying, 1993), or the CO2 degassing/dissolution process during heating and cooling the autoclave (Shibuya et al., 2013). Solutions obtained from blank experiments (without basalt) contained only a few dissolved Fe, with a concentration ≤0.004 mmol/kg, while other elements are lower than the detection limitation, suggesting influence from the autoclaves is negligible. The concentrations of leached elements in black experiments with basalt addition are shown in Figure 2. Obvious increases in Si concentration were observed in experiments with NaHCO3 solution (Runs 2 and 5). Although elemental concentrations are not always consistent with the extent of dissolution, especially at hydrothermal conditions due to generation of secondary minerals, such drastic increase in Si concentration can be a result of the enhanced dissolution of basalt with the use of Na and CO2-rich water. In 5 days, the Si concentration in NaHCO3 solution was much higher than that in water. This high Si concentration then decreased with the reaction duration increased from 5 to 10 days, which was attributed to the generation of secondary silicate minerals. The concentrations of Mg in NaHCO3 experiment also decreased as the reaction duration increased from 5 to 10 days, suggesting the formation of Mg-bearing secondary minerals.

The XRD patterns of original basalt and solid materials collected after experiments are shown in Figure 3. The XRD pattern of solid sample did not show any obvious change after reaction with Milli-Q water for 5 days, indicating that the secondary mineral was not generated, or the generated amount was too less to be detected. However, the peaks of plagioclase, clinopyroxene, and olivine were smaller after reacting with 0.5 mol/L NaHCO3 solution, than those of original basalt, such as the obvious diminishing peak of plagioclase at 13.5°. The preferred dissolution of plagioclase over clinopyroxene from basalt was consistent with the result of Nesbitt and Wilson (1992) (Nesbitt and Wilson, 1992). Calcite is the main product of basalt carbonation, as evidenced by the characteristic calcite peak at 29.6° (2θ, CuKα) (Wang et al., 2021) after reaction of basalt in NaHCO3 solution, and the calcite peak showed higher intensity when the reaction time increased from 5 to 10 days, suggesting an increase in calcite concentration in the reacted sample. Meanwhile, peaks of analcime (e.g., 15.8°, 26.0°), a Na-Al silicate with the general chemical formula of NaAlSi2O6·H₂O, and smectite (in the range of 6.3°-8.0°), a clay mineral contains Mg, Fe, Si, as well as Al (Yamashita et al., 2019), were also detected. This finding was also evidenced by SEM images, as shown in Figures 4A, B. Rhombohedron calcite was observed in both experiments using NaHCO3 solution, but an increase in calcite particle size from 2 µm to >4 µm was observed when the reaction time increased from 5 to 10 days. Smectite with honeycomb texture was only observed on the surface of basalt after reaction in NaHCO3 solution for 10 days (Figure 4B). However, analcime was not found during SEM observations even though XRD suggested its generation, which might be covered by smectite. The generation of analcime and smectite during reaction between 5 and 10 days was consistent with the ICP-OES result, i.e., the decrease in Mg and Si concentrations.

Calcite in the reacted samples obtained from Runs 2 and 5 was quantified via TG analysis and the first derivative curves (Figure 5). The weight loss at 300-780°C was assigned to the decomposition of Ca carbonates (loss of CO2), based on previous studies (Wang et al., 2022). Therefore, weight losses of 0.6% and 4.0% for solid samples collected on Day 5 and 10 of experiments with NaHCO3 indicated that the calcite mass fractions reached 1.4 wt% and 9.1 wt%, respectively. On ignoring the changes in the total weight of solids during reactions, calcite formation in the later 5 days was 5.5 times higher than that formed during the initial 5 days. This kind of enhancement was unexpected because rocks generally dissolve faster in the initial stage due to the presence of finer rock particles and more fresh surface area for dissolution. The saturation index (SI) of calcite was then calculated using SOLVEQ based on the composition of fluid obtained after each experiment, the results are shown in Table 3. In the 10-day experiment at 500°C, SI of calcite (1.6) is slightly increased than that of 5-day experiment at 200°C (1.3), which is consistent with the experimental observation. It should also be noted that for Day 10 sample, an additional weight loss was observed at around 110-380°C, which might be attributed to the dehydration of analcime and smectite, as suggested by Liu et al. (2019); Frost and Ruan (2 ngàn). However, weight loss at 400-700°C, which usually represents the dehydroxylation of smectite (Frost and Ruan, 2 ngàn), was little, suggesting smectite fraction was minor in the reacted samples while most secondary silicate was analcime. Therefore, the acceleration of carbonation at the latter stage of the 10-day experiment might be related to both a higher SI of calcite and the formation of analcime with the change in fluid chemistry. This process consumed the dissolved elements and triggered further dissolution of basalt.

3.2 Effects of NaHCO3 as a function of temperature

Basalt was reacted in 0.5 mol/L NaHCO3 solution at 500°C, 230°C and 200°C, respectively, for 5 days to clarify the effects of temperature on CO2 storage. The chemistry measurement of fluids obtained after each experiment indicated that a higher temperature might favor basalt dissolution, as evidenced by the high concentrations of Si, Al, and Ca at higher reaction temperatures (Figure 6). This kind of enhanced dissolution was particularly notable when the temperature was ≥230°C. However, the same result was not observed in Mg and Fe concentrations, which, on the contrary, decreased when the reaction temperature increased from 230°C to 400°C, which might be related to the promoted generation of Mg and Fe bearing secondary minerals such as smectite at a higher temperature.

XRD measurement of the reacted solids collected after reacting in NaHCO3 solution at 230°C and 100°C for 5 days showed that in almost all characteristic peaks of plagioclase and olivine, the original compositions of basalt were diminished whereas clinopyroxene peaks were relatively stronger due to the consumption of other minerals (Figure 7). This observation suggested that during basalt dissolution in hydrothermal CO2-rich saline systems, the dissolution of plagioclase and olivine was relatively faster while clinopyroxene may be the last one to be dissolved. It should be noted that this kind of peak diminishment does not mean that all minerals have been dissolved or reacted because mineral alteration occurs from the surface and the penetration depth of XRD measurement is limited to dozens of microns. Meanwhile, strong analcime and calcite peaks were observed in experiments conducted at 230°C and 400°C. According to SEM observation, analcime crystals with typical deltoidal icositetrahedron habit grew up to 6 μm in size (Figure 8) at the reaction temperature of 400°C. Among analcime particles, smaller calcite particles were observed as compared to analcime particles. The SI of calcite increases from 1.3 to 2.3 with the increase of reaction temperature from 200°C to 200°C, which is consistent with the experimental observations that more calcite was generated at a higher temperature. In this study, although basalt was dissolved to simultaneously release Mg, Ca, and Fe, the only carbonate generated is calcite [Ca carbonation with few Mg, Ca/(Ca + Mg) = 0.95] in all experiments using NaHCO3 solutions (Runs 2-5). At such high temperature, very small amount of Mg has contributed to the formation of carbonates.

As expected from the fluid chemistry measurement results and confirmation by the TG analysis, at a higher temperature, more calcite was generated. The weight loss at 400-780°C for samples reacted in NaHCO3 solutions at 230°C and 100°C were 4.9% and 7.5% (Figure 9), respectively, suggesting the calcite fractions of 11.1 wt% and 17.0 wt%, which were 7.9 and 12.1 times higher than that generated at 400°C, respectively. If the weight changes during basalt alteration were ignored, the approximate calcite formation from basalt carbonation under different conditions was calculated and summarized in Figure 9C. In this study, the initial basalt contained 12.16 wt% CaO, which means if all Ca carbonated, CaCO3 formation efficiency was 0.217 g/g of basalt. Further, reaction with NaHCO3 solution at 500°C suggested that approximately 79% of Ca in basalt was carbonated to form calcite in 5 days and this ratio could be elevated if the reaction time was extended. The CO2 storage efficiency at 100°C was 75 g/kg of basalt in 5 days. With the increase in calcite generation, analcime generation also enhanced, as indicated by a larger weight loss at 110-380°C, which indicates a positive correlation between albitization and calcium carbonate formation process.

3.3 Albitization and calcium carbonate formation reproduction using anorthite

To reproduce the albitization process which has been observed in basalt experiments and to have a better observation of how NaHCO3 promote mineral dissolution and carbonation, anorthite (CaAl2Si2O8) with negligible amount of Na, was reacted in both Milli-Q water and 0.5 mol/L NaHCO3 solution at 100°C for 5 days. Thin sections of the solid samples collected after reaction were prepared and observed via SEM. In water system, the surface of anorthite (edge of particles shown in Figure 10A) was rarely altered and no secondary minerals were found whereas in the Na and CO2-rich system (i.e., NaHCO3 solution), nearly half of the anorthite was reacted, as revealed by a clear boundary between reacted and unreacted anorthite (Figure 10B). In the reacted region, EDS measurement suggested that the compositions of the base were mainly Na, Ca, Al, and Si while calcite particles distributed along the base. Results of XRD analysis of the reacted anorthite also suggested the generation of calcite and cancrinite in the experiment with NaHCO3 solution (Figure 10C). Cancrinite is a complex carbonate and silicate with a general composition of Na6Ca2[(CO3)2|Al6Si6O24]·2H2O (Seo et al., 2018). However, also due to the formation of cancrinite, the carbonation efficiency could not be determined according to TG analysis, because the weight losses due to cancrinite and calcite decompositions were overlapped (Hassan et al., 2006).

4 Discussion

Due to the urgent need of decreasing atmospheric CO2 concentration, mineralization of CO2 into carbonates using rocks, especially the abundant and widely distributed one, basalt, has raised a lot of interest. This study provided fundamental insight into the process of basalt carbonation in a CO2-rich hydrothermal system and experimentally confirmed the enhancement of basalt carbonation with generation of more calcite in such system with a higher temperature.

Xem Thêm : Bài 25. Sự nhiễm từ của sắt, thép – nam châm điện – Củng cố kiến thức

A positive relationship between calcite and secondary silicate minerals such as analcime and smectite formation was revealed by the experiments. According to TG analysis, weight loss occurred at 400-780°C and 110-380°C, which has been assigned to the decomposition of calcite and analcime, respectively, indicating a good linear relationship with coefficient factor R2 up to 0.968 (Figure 11). Assuming that calcite and analcime fractions are directly proportional to their weight losses during TG measurement, their generations show good linear relationship. In 400°C NaHCO3 solution, analcime was not generated in 5 days, possibly because the concentration of Si was not high enough, which increased with the reaction time and finally triggered analcime formation after 5 days reaction. Such kind of secondary mineral formation in turn promoted more basalt to be dissolved and generated more calcite. The formation of calcite from plagioclase may present as follows (Eq. 1), where the hydrothermal fluids and pyroxenes provided additional Na+ and Si sources.

This study suggested that NaHCO3 in hydrothermal systems plays a crucial role in enhancing CO2 storage as well as promoting the generation of secondary silicates such as analcime during the alteration of basalt (illustrated in Figure 12). The use of such Cl-free and weaker acid fluid may be helpful to minimize the corrosion of CO2 injection tubes during the engineering process of CO2 mineralization. However, although this study suggested hydrothermal conditions for basalt carbonation, operation management in such high-temperature reservoirs is usually difficult and CO2 storage efficiencies change with the unstable geothermal system. Instead, using the chemical engineering process and industrial waste heat for high-temperature basalt carbonation will be more controllable; at the same time, it contributes to the efficient utilization of basaltic slags. According to the experiments conducted in this study, 0.17 kg calcite could be formed during the alteration of 1 kg basalt in 20 L NaHCO3 at 200°C. If assuming the initial system temperature is 25°C, 9.72 × 10−2 GWh energy is required to heat the system to 100°C for storing 1 ton of CO2; and this energy can be supplied by waste heat produced in non-metallic mineral, chemical and petrochemical, iron and steel industry sectors (Bianchi et al., 2019). Another way of applying this finding is in the CO2 plume geothermal power generation process, during which basalt is one of target rocks and some amount of CO2 may be stored as minerals at the edges of the reservoir. The use of NaHCO3 hydrothermal fluid may contribute to keeping the bulk of CO2 inside the geothermal reservoir. Based on these findings, further studies should focus on developing a better understanding of the dissolution behaviors of basalt blocks in Na-rich hydrothermal systems and changes in subsequent rock properties as a function of reaction time with a view to expand the efficient utilization of basalt reservoirs for CO2 capture and storage. Moreover, the effect of basalt properties on CO2 storage efficiency should also be clarified.

5 Conclusion

The present study experimentally quantified the effects of NaHCO3 on basalt carbonation at 500-300°C in Cl-free environments. The presence of NaHCO3 was confirmed to drastically promote the alteration of basalt with accelerated CO2 storage, especially at higher temperatures. The CO2 storage efficiency at 600°C reached 75 g/kg of basalt in 5 days, which was 12.1 times higher than that at 400°C. In such hydrothermal systems, basalt was altered to generate mainly calcite, analcime, and smectite. Further, the dominant element for CO2 storage was Ca mainly from plagioclase, while released Mg and Fe were converted to clay minerals such as smectite. Na in the system participated in the formation of Na silicates such as analcime in the case of basalt. In this way, the main elements released during basalt dissolution were quickly consumed in secondary mineral generations, which in turn promoted basalt dissolution to release more Ca and CO2 storage. Fundamental effects of adding NaHCO3 for basalt alteration have been considered in this paper, however, detailed chemical mechanisms should be investigated as further studies.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

SK: conceptualization, methodology, data curation, formal analysis, writing-original draft. JW, NW and NT: conceptualization, methodology, investigation, funding acquisition, project administration, resources, writing-original draft and writing-reviewing and editing. OD, MU and NH: investigation, methodology, software, writing-reviewing and editing. All authors contributed to the article and approved the submitted version.

Funding

The present study was partially supported by the Japan Society for the Promotion of Science (JSPS) through Grant-in-Aid for Early-Career Scientists (No. 21K14571), Grant-in-Aid for Scientific Research (A) (No. 21H04937), and JST/JICA SATREPS (No. JPMJSA1703).

Acknowledgments

We would like to thank Shinichi Yamasaki at the Graduate School of Environmental Studies, Tohoku University, for their assistance during the ICP-OES analyses.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References