Conversion of CO2 into mineral carbonates using a regenerable buffer to control solution pH
Introduction
This paper investigates a novel concept to enable the conversion of CO2 into mineral carbonates using an aqueous route. The barrier that is currently stalling conversion is pH control. The first part of this introduction outlines why pH is critical, the second part outlines what work has been done in the field of converting CO2 into mineral carbonates using an aqueous route, and the third part introduces the novel concept behind this study.
In order for a mineral carbonate, such as MgCO3, to form the concentrations of and Mg2+ in solution must be high enough for Eq. (1) to be satisfied [1].
The solubility of CO2 in aqueous solution and the dissociation of carbonic acid to form bicarbonate and carbonate anions can be described by Eqs. (2), (3), (4), (5) [1], [2], [3], [4].where 29.36 is the Henry’s law constant (dm3 atm mol−1) for CO2 in water at 25 °C [2]. Here, [H2CO3] includes both dissolved molecular CO2 and molecular H2CO3. Eqs. (3), (4) shows that ions only start to become significant in solution when the pH is above about 8 (see Fig. 1). It could be thought that if the partial pressure of CO2 was raised the concentration of all species in solution (H2CO3, and ) would increase, and therefore a raised concentration of would result. However, solving Eqs. (2), (3), (4), (5) for various partial pressures of CO2 () shows that the concentration of in solution stays very low at around 5 × 10−11 M despite the partial pressure of CO2 rising to 50 atm (see Fig. 2). This is due to the H+ ions, forming from the dissociation of H2CO3, always suppressing the formation of (see Eqs. (3), (4)). Only the concentration of becomes appreciable at high CO2 partial pressures.
Given the solubility limit of MgCl2 in solution (55 g/100 ml), the maximum possible concentration of Mg2+ is approximately 6 M Mg2+ and the corresponding carbonate concentration required to precipitate MgCO3 is 5 × 10−9 M. Given that 6 M Mg2+ is difficult to achieve as it is close to the solubility limit, a more realistic concentration of 0.6 M Mg2+ would require a concentration of 5 × 10−8 M . Assuming a partial pressure of CO2 of 1 atm, Eqs. (2), (3), (4) can be solved to establish a relationship between carbonate concentration and pH. Fig. 3 shows this relationship and shows that when the carbonate concentration is 5 × 10−8 M the equilibrium pH is approximately 5.5. This means that in order to precipitate MgCO3 the pH needs to exceed 5.5.
Bond and co-workers [5], [6], [7] looked at the potential of using an enzyme, called carbonic anhydrase, to convert power station CO2 into CaCO3. Coral reefs use carbonic anhydrase to assist in the production of . The enzyme was found to catalyse the formation of , however, although the catalyst enables the rapid formation of , there is also a simultaneous rapid drop in solution pH to approximately 4, which prevents the precipitation of carbonates. In the work of Bond et al. [5], [6], [7] and in recent work by Mirjafari et al. [8], Ozdemir [9] and Rayalu and co-workers [10], [11], precipitation of carbonates was only possible if a buffer was added to the solution in order to complex H+ ions and keep the pH high. The buffer used was tris(hydroxymethyl)aminomethane or ‘Tris’, which is commonly used to maintain pH at around 8. This buffer could not be continuously used in a large scale process, as it cannot be regenerated, which presents a major obstacle to the further development of this approach to sequestering CO2. Coral reefs have the ocean as a giant buffer for the H+ that they generate. It is worthwhile to note that Mirjafari et al. [8] found that calcium carbonate did not precipitate when they used carbonic anhydrase with no buffer, but found precipitation to take place when they used the buffer with no carbonic anhydrase.
Instead of converting salt solutions into carbonates, researchers have also looked at converting Mg–silicate minerals into carbonates in order to use the neutralising capacity of the mineral [12]. O’Connor and co-workers [13], [14] found they could convert 34% of serpentine (Mg3Si2O5(OH)4) to MgCO3 using aqueous CO2 at 115 atm, 185 °C and a residence time of 24 h. The pH generated in solution would be around 3. The high temperature would be assisting the kinetics of Mg dissolution. Presumably the pH increases as dissolution proceeds and this aids precipitation of carbonate. However, as the pH increases the dissolution rate of Mg would drop to negligible levels. The silica left behind and carbonate forming would also hinder further Mg dissolution/carbonation. O’Connor et al. found that artificially adding NaHCO3 to shift the equilibria in favour of a higher to H+ ratio aided carbonation.
It follows that the best way to enable carbonation might be to have a two stage process where the first stage is optimised for Mg dissolution (high temperature and low pH) and the second is optimised for Mg carbonation (high pH). This need has been recognised as pH swing [15]. pH swing requires the removal of H+ ions from solution, which is usually achieved with soluble oxides/hydroxides, however, these come from the calcination of carbonates which is obviously not possible.
This paper investigates the use of weakly basic tertiary amines for complexing H+ ions generated when CO2 is added to solution and therefore enabling carbonate precipitation from a variety of salt solutions. The acid dissociation constant for protonated tertiary amines varies with temperature and so the approach is to use “temperature swing” to regenerate the amine, whereby the loaded amine is heated to strip the acid off. This concept has recently been patented [16].
Conventional CO2 capture technologies involve absorbing CO2 into a mixture of primary and tertiary amines, including monoethanolamine (MEA) and methyldiethanolamine (MDEA) respectively. The reason for this is that while the primary amine forms a strong carbamate bond and therefore enables a high CO2 loading due to the strength of the bond considerable energy is needed to break it and regenerate the MEA (approximately 3–5 MJ/kg CO2 [17]). In order to reduce the energy load tertiary amines are blended. Tertiary amines do not bond with CO2 but rather strip the solution of H+ ions thus driving the formation of and ions in solution. The loading of CO2 in solution as these ions is much lower but the energy needed to regenerate the MDEA is much less, such that the total energy needed to regenerate the MEA/MDEA mixture is around 1–3 MJ/kg CO2 [17]. The introduction of MDEA means that taller towers are needed for a given separation efficiency and so a balance between capital and operating (energy) cost must be struck.
The idea put forward here is to use tertiary amines alone to convert CO2 into mineral carbonates. It is thought that the loading of CO2 can be higher than that for conventional CO2 capture using MDEA because the ions that form leave the solution as solid carbonate, thus providing a stronger driving force for the capture of CO2 into solution.
The advantage of this concept for CO2 sequestration is twofold. Firstly, the energy needed for the process might be considerably lower than that needed for conventional CO2 capture because less energy is needed to regenerate the amine and compression of the CO2 (needed for storage) is not necessary. Secondly, the CO2 would be locked up in a mineral form that is known to be stable for millions of years, which is an important consideration given the scale of CO2 that needs to be stored.
Fig. 4 shows a simplified diagram of the novel concept that is being explored. In stage 1, acid loaded amine is heated to ∼100 °C and contacted with a Mg silicate such as serpentine ((Mg, Fe)3Si2O5(OH)4). At the high temperature the acid (HCl) dissociates from the amine, thereby providing a low pH capable of dissolving Mg out of the serpentine. The Mg depleted serpentine is separated by density and/or filtration and the solution, containing MgCl2 and regenerated amine, passes to stage 2. In stage 2, the solution is cooled and the flue gas containing CO2 is sparged through. At the low temperature acid generated by the dissociation of H2CO3 plus excess acid generated in stage 1 are complexed by the tertiary amine which causes the pH to rise and consequently the concentration to rise to a level that is sufficient to begin interacting with Mg2+ and precipitating MgCO3. The MgCO3 is separated by density and/or filtration and the solution, containing acid loaded amine is recycled to stage 1.
The overall reaction taking place in stage 1 is as follows:
And the overall reaction taking place in stage 2 is as follows:
The concept shown in Fig. 4 can be used in a variety of different ways by a variety of different industries, and is not locked into one industrial application.
The concept could be used for the treatment of magnesium silicate deposits as described above. It is worthwhile to note that many magnesium silicate deposits contain significant levels of valuable heavy metals such as Ni, called Ni laterites, and the process could therefore have the dual operation of Ni extraction combined with CO2 sequestration. It is known that treating Ni laterites with acid to dissolve the Mg enables a greater amount of Ni to be extracted into solution.
Secondly, the concept could be used for salt mining operations. Chloride and sulphate salt deposits are mined for KCl or K2SO4 to be used as fertiliser. By using solution mining the salt comes up hot and is cooled to separate the potassium salts. The refuse salt represents a waste that is generally deposited back into the reserve. The salt solution could be processed using the sequestration technology shown in Fig. 4 to form a mixture of carbonates and bicarbonates that are deposited back into the reserve for longterm CO2 storage. The heat that must be taken out of the solution as it comes to the surface could be used to provide the heat needed for stage 1. Using the concept for salt solutions means that a by-product of the process is acid, either hydrochloric or sulphuric acid. Therefore, the scale of the operation would need to be matched with HCl or H2SO4 needs in the oil, chemical and mineral sectors.
This paper presents our work to date on this novel concept.
Section snippets
Serpentine sample
The serpentinite sample (Mg3Si2O5(OH)4) used for this study was obtained from a naturally occurring deposit in northern Queensland, Australia. It was initially ground by hand in a pestle and mortar and then in a laboratory attrition mill. The ground serpentinite was sieved with ASTM standard sieves to obtain particles with a diameter of <57 μm. Australian Laboratory Services (ALSs) performed the elemental analysis via alkali fusion, acid digestion and inductively coupled plasma-atomic emission
Magnesium dissolution
The effect of HCl concentration on the dissolution of Mg from serpentine is shown in Fig. 6. All percentages are weight percentages. The residence time used was 3 h and the temperature was the boiling temperature of the solution (∼100 °C). The stoichiometric amount of acid needed to dissolve all of the Mg according to R1 is approximately 0.12 M which gives approximately 40% extraction, while a plateau of approximately 65% extraction is reached at around 0.5 M HCl. Fe and Al were also found to
Conclusion
The best conditions for the dissolution of Mg from serpentine have been found to be a solids to solution volume of more than 50 g/L to enable a high proton concentration. The amount of acid should be no more than twice the stoichiometric amount needed for reaction. Reaction temperature should be as high as possible, close to the boiling temperature of the solution or higher (100–150 °C) if using a pressurised vessel. These conditions combined with a residence time of 3 h are able to dissolve
References (19)
- et al.
Biomimetic sequestration of CO2 in carbonate form: role of produced waters and other brines
Fuel Process Technol
(2005) - et al.
Carbon dioxide disposal in carbonate minerals
Energy
(1995) - et al.
Chemistry of aqueous mineral carbonation for carbon sequestration and explanation of experimental results
Environ Progr
(2006) - et al.
CO2 mineral sequestration: physically activated dissolution of serpentine and pH swing process
Chem Eng Sci
(2004) - et al.
Integration of CO2 capture unit using single- and blended-amines into supercritical coal-fired power plants: implications for emission and energy management
Int J Greenhouse Gas Control
(2007) - et al.
Dissolution of natural serpentinite in mineral and organic acids
Int J Miner Process
(2007) - et al.
Critical stability constants volume 4
(1974) - et al.
The chemistry of metal carbonato and carbon dioxide complexes
Chem Rev
(1983) - et al.
The ionization constant of carbonic acid in water and the solubility of carbon dioxide in water and aqueous salt solutions from 0 to 50°
J Am Chem Soc
(1943)
Cited by (24)
A sustainable temperature-swing process for CO<inf>2</inf> capture and mineralization at below 100 °C using a recyclable chelating agent and bottom ash
2024, Journal of Environmental Chemical EngineeringDiluted methane mitigation by a co-culture of alkaliphilic methanotrophs and the microalgae Scenedesmus obtusiusculus towards carbon neutrality
2024, Biochemical Engineering JournalReview of contemporary research on inorganic CO<inf>2</inf> utilization via CO<inf>2</inf> conversion into metal carbonate-based materials
2022, Journal of Industrial and Engineering ChemistryAdvances in process development of aqueous CO<inf>2</inf>mineralisation towards scalability
2020, Journal of Environmental Chemical EngineeringCitation Excerpt :Besides the inorganic basic additives elevating the basicity of a solution, organic substances were investigated to increase the pH for carbonation. Steel et al. [77] implemented a tertiary amine to buffer the solution to raise the pH to 8 for carbonation. In order to accelerate the carbonation rates and improve the purity of products, crystal seeds was added in the research of Katsuyama et al. [93].
Development of a nickel extraction-mineral carbonation process: Analysis of leaching mechanisms using regenerated acid
2020, HydrometallurgyCitation Excerpt :Recycling acid may be a good approach to reducing acid requirements for Mg-rich laterites. The University of Queensland (UQ) is developing a process for Ni laterites that aims to improve the sustainability of Ni laterite operations by recycling acid, as well as recycling base and reducing carbon dioxide (CO2) emissions (Steel et al., 2013; Balucan and Steel, 2015; Zappala et al., 2019). The main input to the process is heat, which could come from renewable resources, and the process is designed to utilise low value Mg-rich ores with a very low Ni content.
Acceleration of CO<inf>2</inf> mineralisation of alkaline brines with nickel nanoparticles catalysts in continuous tubular reactor
2019, Chemical Engineering JournalCitation Excerpt :There are several methods to prevent a solution from getting overly acidic after injecting CO2. The most widely used is to add substances into the solution, such as buffer solution [19,20], basic inorganic chemicals (e.g. sodium hydroxide, ammonium hydroxide, industrial wastewater [21–23]), organic chemicals (e.g. methanol, ethanol, propanol [24]), and solid industrial wastes such as steelmaking slag [25], red mud and fly ashes [26,27]. The approach used in this research was to contact the brine with blast furnace (BF) slag, to transfer alkalinity to the brine, with sodium hydroxide solution being used as an additional alkali substance.