Electrochemical CO2 mineralization for red mud treatment driven by hydrogen-cycled membrane electrolysis

Heping Xie,Yunpeng Wang ,Tao Liu ,Yifan Wu ,Wenchuan Jiang ,Cheng Lan ,Zhiyu Zhao,Liangyu Zhu,Dongsheng Yang

1 School of Chemical Engineering,Sichuan University,Chengdu 610065,China

2 Institute of Deep Earth Science and Green Energy,Shenzhen University,Shenzhen 518060,China

3 Institute of New Energy and Low-Carbon Technology,Sichuan University,Chengdu 610065,China

4 State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation,Southwest Petroleum University,Chengdu 610500,China

Keywords:CO2 mineralization Red mud Electrolysis Waste treatment Flue gas

ABSTRACT CO2 mineralization as a promising CO2 mitigation strategy can employ industrial alkaline solid wastes to achieve net emission reduction of atmospheric CO2.The red mud is a strong alkalinity waste residue produced from the aluminum industry by the Bayer process which has the potential for the industrial CO2 large scale treatment.However,limited by complex components of red mud and harsh operating conditions,it is challenging to directly mineralize CO2 using red mud to recover carbon and sodium resources and to produce mineralized products simultaneously with high economic value efficiently.Herein,we propose a novel electrochemical CO2 mineralization strategy for red mud treatment driven by hydrogen-cycled membrane electrolysis,realizing mineralization of CO2 efficiently and recovery of carbon and sodium resources with economic value.The system utilizes H2 as the redox-active proton carrier to drive the cathode and anode to generate OH- and H+ at low voltage,respectively.The H+ plays as a neutralizer for the alkalinity of red mud and the OH- is used to mineralize CO2 into generate highpurity NaHCO3 product.We verify that the system can effectively recover carbon and sodium resources in red mud treatment process,which shows that the average electrolysis efficiency is 95.3% with highpurity (99.4%) NaHCO3 product obtained.The low electrolysis voltage of 0.453 V is achieved at 10 mA.cm-2 in this system indicates a potential low energy consumption industrial process.Further,we successfully demonstrate that this process has the ability of direct efficient mineralization of flue gas CO2(15%volume)without extra capturing,being a novel potential strategy for carbon neutralization.

1.Introduction

The accumulation of CO2emissions from the large-scale utilization of fossil fuels has resulted in an alarming rate of climate change,which has a huge impact on the ecological environment[1].The reduction in CO2emissions has become a global consensus[2].In recent years,considerable research has been performed on CO2emission reduction technologies,including CO2capture,utilization,and storage[3,4].Among them,mineralization is an effective means of industrial CO2utilization,which is one of the most promising approaches to reduce emissions,owing to its thermodynamic favorable properties [5–8].A variety of feedstocks,such as bulk calcium and magnesium salt ore [9,10] (serpentine,olivine,wollastonite,etc.) or industrial solid residue [11,12] (carbide slag,steel slag,fly ash,etc.) can be utilized in CO2mineralization processes to convert CO2to stable carbonates.Recently,mineralization using alkaline residues has become an attractive method for direct and indirect decreases in CO2emissions from industries and power plants.Despite the significant CO2capture capacity of natural ores,mineralization using alkaline solid wastes has other merits,such as low feedstock cost and availability near the source of CO2[3].

Annually,billions of tons of solid waste are discharged during industrial production[13].Red mud is a by-product of bauxite processing via the Bayer process [14].Approximately 35%–40% of the processed bauxite ore enters the waste in the form of alkaline red mud slurry,and 1–1.5 tons of red mud are generated per ton of alumina produced [15].An enormous quantity of red mud is generated worldwide annually,posing a consequential and alarming environmental challenge,such as contamination of soil and groundwater resources due to large stockpiles[16].Several studies have indicated that red mud has the potential for comprehensive utilization,such as the production of construction bricks and cement [17,18] and environmental governance [19,20].However,the strong alkaline content of red mud restricts its comprehensive utilization.Utilizing CO2to mineralize red mud can reduce CO2emissions and treat the alkali of red mud [21].The mechanisms of red mud mineralization by carbon dioxide gas,in which carbonic acid reacts with the basic components of the red mud have been studied.Because the mineralization reaction is relatively slow(similar to the natural weathering process),at the short contact time required by the industrial process rate,solely a fraction of the alkaline substances in the red mud are mineralized by gaseous CO2[22].Hence,the red mud needs to be activated and pretreated to accelerate the mineralization reaction process in hightemperature and high-pressure environments,resulting in high energy consumption for mineralization reactions.Moreover,owing to the uncontrollability of the reaction,it is difficult to accurately regulate the conventional method of mineralizing red mud.Although there are considerable usable metal resources in red mud,limited by its complex composition,the current mineralization technology can only obtain a mixture as its product [23],which is difficult to separate for further utilization or sell as a valuable product,and resource recycling cannot also be achieved.Therefore,improving the purity of the product and the efficiency of mineralization are technical bottlenecks that need to be addressed urgently.

To treat the alkalinity of red mud and realize the recovery of sodium resources,in this study,we propose a novel electrochemical mineralization CO2process for the treatment of alkali from red mud with resource recycling driven by hydrogen-cycled membrane electrolysis.As illustrated in Fig.1(a),the CO2mineralization electrolysis cell (CMEC) has an HOR in the anode zone and HER in the cathode zone.As H+is reduced at the cathode to generate H2,is converted to(Eq.(1)).Then,it was balanced by Na+,which comes from the soluble alkali in red mud,permeated from the anode to form Na2CO3.CO2was blown into the cathode chamber to produce sodium bicarbonate (Eq.(2)).H+is generated by the oxidation of hydrogen gas at the anode and further combines with OH-(Eq.(3)).The anode liquid containing Na2SO4was circulated back to the leaching step to re-immerse the red mud.The red mud obtained can be utilized for construction production.Electrochemical mineralization CO2technology using hydrogencycled membrane electrolysis enables reactions to be performed under mild and low energy consumption conditions with the recovery of sodium resources.

Fig.1.(a) Illustration of CO2 mineralization electrolysis cell.(b) Schematic representations of the process.

Aiming at the alkaline impurities in red mud,this study proposes an electrochemically efficient red mud resource recovery process.The system flowchart for CO2mineralization and treatment of red mud is illustrated in Fig.1(b),where the three sections can be considered.Alkaline leaching section:Obtain the red mud leaching solution while removing solid waste residue.Electrolysis section:Neutralize the alkaline leaching solution and obtain the alkali-rich catholyte to facilitate subsequent operations.Mineralization section:Absorb CO2to obtain the final product.

In this study,we verified the feasibility of utilizing such a process coupled with CO2mineralization to treat alkaline red mud.Subsequently,the process was systematically investigated to demonstrate its excellent red mud alkaline treatment ability and resource recovery capacity.Different operational parameters were studied to identify suitable conditions for high-efficiency leaching and low-energy electrolysis.Finally,the product purity and estimated energy consumption were discussed.

2.Experiment and Analysis

2.1.Material

The red mud samples utilized in this study were obtained from the Hongqiao electrolytic aluminum factory (Shandong,China).Analytically pure sodium carbonate,sodium bicarbonate,sodium sulfate,and hydrochloric acid were purchased from Kelon.CO2and N2gases with purities of 99.99%were purchased from Heping(Chengdu,Sichuan) gas company.The H2utilized in the experiment was produced by a hydrogen generator (SFH-300).The 40%Pt/C catalyst,5% Nafion solution,60% polytetrafluoroethylene(PTFE) emulsion,Nafion-115 membrane,and carbon paper with a microporous layer on one side (HCP120) were purchased from Shanghai Hesen Electric Co.,Ltd.A platinum-plated nickel mesh was self-prepared in the laboratory (Pt loading capacity was 1 mg.cm-2).All the solutions utilized in these experiments were prepared using deionized water.

2.2.Pre-preparation

2.2.1.Gas diffusion electrode (GDE) preparation

The catalyst slurry was configured with a certain proportion of Pt/C,Nafion solution,and PTFE solution,and dispersed in 50 ml of isopropanol ultrasonically for 30 min.The Nafion-115 membrane was preheated at 120 °C as the substrate.The catalyst was then sprayed evenly on the surface of the substrate to ensure the Pt load is 1 mg.cm-2.

2.2.2.Assembly of CMEC

An assembly drawing of the CMEC is illustrated in Fig.2.The cell is composed of H2,anolyte,and catholyte chambers.The H2and anolyte chambers are separated by a hydrogen GDE,while the anolyte and catholyte chambers are separated by a cation exchange membrane (CEM).The current collecting plate is a 10 mm thick graphite plate,on which a serpentine flow channel is designed to ensure the uniform distribution of hydrogen and catholyte on the entire electrode surface.The surfaces of the GDE and CEM were both 2 cm ×2 cm square.An elastic polymer mesh utilized between the anode and CEM supports the liquid flow space and increases the electrical conduction.The cathode was fabricated using a platinum-plated nickel mesh.Each component of the cell was hermetically sealed using rubber gaskets.

Fig.2.The structure of CMEC.

2.3.Experimental process

Leaching process:The purpose of leaching is to extract soluble sodium from red mud,which is convenient for subsequent comprehensive utilization.We took 10 g of already ground red mud,passed it through a 200-mesh sieve,and dried it in an oven at 110 °C overnight.Then,red mud and ultrapure water were added to a 200 ml beaker according to a certain liquid–solid ratio.After reacting in a water bath magnetic stirrer for a period of time,a vacuum suction filter was utilized for solid–liquid separation,and the Na+content in the extract was detected by AAS.The separated red mud residue was subjected to XRF (X-ray fluorescence) and XRD(X-ray diffraction) to determine its elemental content.

Membrane electrolysis process:The electrolytic process was processed in a CMEC.A flow of H2was introduced into the GDA at a flow rate of 30 ml.min-1.The leaching solution containing 0.5 mol.L-1Na2SO4and NaHCO3solution was pumped into the CMEC from two 100 ml round bottom flasks to serve as the anolyte and catholyte.The cells were placed in a water bath to maintain the temperature at 30 °C.The electrolyte was circulated at a rate of 80 ml.min-1.In the experiment,to reduce the change in the potential caused by the H2partial pressure of the anode and cathode,the catholyte was injected with H2before the reaction to maintain the balance of the H2pressure at both electrodes.

Three experiments were conducted:one was to measure the variation in cell voltage and pH by keeping the current density constant at 10 mA.cm-2,and sequentially withdrawing samples of anode and cathode solutions in aliquots of 3.0 ml at specified time intervals.The sample concentrations were then tested using acidbase titrations.The other was to measure the current–voltage relationship of the PCME cell by adjusting the voltage from 0 to 1.2 V at a rate of 10 mV.s-1.The final step was to verify the hydrogen circulation efficiency in the membrane electrolysis system by quantitatively detecting the amount of hydrogen produced and consumed by gas chromatography (GC).

Crystallization process:The catholyte was pumped into a 50 ml beaker,where the catholyte was carbonated by bubbling a stream of CO2at a flow rate of 100 ml.min-1,while the pH value of the catholyte was recorded over time.

Owing to its low solubility,the crystal product crystallized from the catholyte.When the pH value did not change,the absorption process was considered to have ended.The filtered product was vacuum dried in a vacuum drying oven at 45°C for 48 h.The dried product was subjected to XRD and TGA (thermogravimetric analysis).

2.4.Analysis and characterization

The AAS method was utilized to analyze the Na+content in the leaching solution,and the alkali dissolution rate of the water leaching process was calculated using Eq.(5):

where nsand nrare the sodium contents in the leaching solution and red mud,respectively.

By determining the concentration of OH-andin the anode and cathode respectively,the electrolysis efficiency in the anode and the conversion rate of CO2in the cathode can be calculated using the following equations (Eqs.(6)–(7)):

where ηanodeand ηcathodeare the anode and cathode electrolysis efficiencies,respectively;nexpand ntherepresent the experimental and theoretical amounts of OH-andions,respectively,and Chaand Vhaare the concentrations and volumes of hydrochloric acid utilized in the titration,respectively.Vsand Vrare the volumes of the samples and the reaction solution,respectively,F=96485 C.mol-1is Ferrari’s constant,and z (z=1) is the number of electrons transferred in the electrolytic process.In addition,I and t represent the current and time utilized in the electrolytic process,respectively.

The theoretical hydrogen circulation volume was calculated using Faraday’s law and the law of thermodynamics (Eqs.(8)–(9)).The hydrogen cycled efficiency was calculated using Eq.(10).

where Vexpis the actual volume of hydrogen measured by gas chromatography using N2as the carrier gas,n0is the number of moles of hydrogen generated theoretically,calculated by the Faraday formula,P is the gas pressure,R is the ideal gas constant,and T is the Kelvin temperature.

The input energy comes from three parts,energy for electrolysis(WE),pump work for CO2inlet (Wp) and thermal energy (Qin) for electrolyte heating up (Eq.(11)).

(a) The theoretical electrolysis energy consumption of the CMEC was estimated according to the thermodynamics of the electrode reaction.A redox reaction of H2/H+occurs on the electrodes of the cell.Previous studies [24,25] provided the theoretical potential of the electrode at 25 °C according to the Nernst equation (Eq.(12)).

where E0is the standard electrode potential from Eq.(1),(E0=0 V),F=96485 C.mol-1,and pHelectrode=-lg(αH+),where (αH+) is the activity of H+in the electrolyte.When the partial pressure of hydrogen gas was 100 kPa,lg()0.5=0.Therefore,the relationship between pH and cell voltage can be calculated using Eq.(13)below.

And the electrolysis energy consumption can be estimated by Eqs.(14)-(15).

where Peis the electrolytic power,and t is the time.U=0.45 V is the average cell voltage of the electrolytic process under 10 mA.cm-2,F=96485 C.mol-1is the Faraday constant,η is the average current efficiency,and n is the number of electrons transferred for soluble sodium ions in 1 ton of red mud.

(b) The pump work can be easily calculated with CO2mass and flow rate.The pump work is

(c) Qinis the heat required to heat up the electrolyte from the working temperature (T1=25 °C) to the regeneration temperature (T2=40 °C) (Eq.(17)).The specific heat capacity of the solution is taken as 3.979 J.K-1.g-1which is similar to other sodium based solutions.

The purity of the product was calculated from the mass loss between the temperatures of 30 °C and 450 °C by TGA.

where ηprepresents the purity of the product,Δm is the sample mass loss,and m is the mass of the sample utilized for TGA analysis.

Gas chromatography was utilized to quantitatively detect the production and consumption of H2by Shimadzu GC-2014C,with N2as the carrier gas.The temperature of the gas chromatography column box was set to 120 °C,the detector temperature was controlled at 150 °C,and the mass flow rate of nitrogen as the carrier gas was 100 ml.min-1.

The XRD analyses were performed using a Bruker D8 advance X-ray diffractometer with CuKα radiation (k=1.54056 Å,1Å=0.1 nm),operating at 40 kV and 30 mA voltage and anode current,respectively,at a scanning mode of 0.03° intervals,set time of 0.05 s and a scanning angle (2θ) from 10° to 90°,to collect the XRD patterns of the samples.

Thermogravimetric analysis (TGA,Beijing HCT-3) was utilized to characterize the decomposition characteristics of the product at a heating rate of 10 °C.min-1and an argon flow rate of 50 ml.min-1.

The SEM experiments were performed using a thermo scientific Apreo 2C scanning electron microscope at 5.0 kV.A polaronion sputtering gun was utilized to deposit a gold film with a thickness of 25 nm onto the samples to enhance their electrical conductivity.

3.Results and Discussion

3.1.The optimization of the immersion process conditions

Red mud is the residue after the digestion of bauxite mineral,and its complex composition was characterized by XRF.The results of the XRF analysis of the sample are presented in Table 1.As the analysis indicates,a typical red mud sample mainly comprises Fe,Al,Si,Na,Ti,Ca,O,and other trace metal elements with resource recovery values,especially sodium.

Table 1 Main chemical composition of red mud

The existing form of sodium in the red mud was studied based on the XRD analysis.Examination of the XRD peak locations and intensities for the red mud samples (Fig.3) illustrates that the sodium in red mud mainly exists in two forms:soluble sodium(NaOH.H2O and(Na2CO3)1.5.(H2O)2)and insoluble sodium(Na8Al6-Si6O24CO3) [26].NaOH.H2O belongs to the unreacted sodium hydroxide remaining in the red mud during the aluminum oxide production process of the Bayer process,and reacts with CO2in the air to produce Na2CO3.Na8Al6Si6O24CO3is produced by the alkali dissolution of bauxite under high pressure.Other components of red mud,such as hematite (Fe2O3),boehmite (AlO(OH)),and laumontite (Ca12Al14O33),are attributed to the complex composition of bauxite.Soluble alkali is also the main cause of harm to the environment by red mud.To quantitatively detect the leaching effect,the total amount of soluble sodium ions was measured.It is obvious that the Na+dissolution rate decreases significantly with an increase in the number of immersions(Fig.4(a)).The alkaline dissolution rate for the first leaching reached 13.19%,and the subsequent leaching rates reached only 5.81%,3.27%,1.16%,0.52%,and 0.32%,respectively.The pH value of the leaching solution also gradually decreased from 10.24 to 8.2.When the pH valueof the red mud leaching solution drops to 8,it can be used as a construction material that does no harm to the environment.Therefore,it can be considered that the red mud after leaching is nonhazardous usable material.

Fig.3.XRD pattern of red mud before and after leaching.

When red mud is in deionized water,NaOH.H2O and (Na2-CO3)1.5.(H2O)2are easily soluble in water and ionize to produce Na+,OH-,and.However,sodalite is a type of zeolite,and its pore structure has a certain physical adsorption effect on Na+,OH-,andproduced by hydrolysis[27].When deionized water is utilized for multiple piles of washing,the Na+,OH-,andin the pores continue to undergo an adsorption–desorption equilibrium process,so that Na+can be gradually and completely leached.Therefore,the soluble Na+in the red mud is the sum of the sodium ions leached each time,accounting for 24.28% of the total sodium content in the red mud.Another part of the insoluble sodium exists in the filter residue in the form of Na8Al6Si6O24CO3.The results of the XRF analysis of red mud after leaching indicate that the sodium content was reduced from 4.74%to 3.59%,which verified the feasibility of the leaching process (Table 2).The determination results of impurity ions in the leaching solution show that compared with sodium ions,the content of other metal ions is several orders of magnitude less (Table 3).Therefore,they have little effect on the purity of the product.

Table 2 Main chemical composition of red mud after leaching

Table 3 Main ion content of leachate

After studying the feasibility of the leaching process,the optimal extraction conditions were determined.Fig.4(b)-(d) illustrates the effects of temperature,leaching time,and liquid–solid ratio on the leaching rate and pH of the leaching solution.Irrespective of the change in conditions,the pH values of the leaching solution,which range from 10.2–10.4 did not change significantly.It is proved that the carbonate ion concentration in the leaching solution is in the range of 0.017 mol.L-1-0.019 mol.L-1through quantitative analysis.This indicates that the buffer capacity of sodium carbonate affects the pH change of the red mud extract.Increasing the temperature can increase the solubility of the alkaline substances in the red mud and move the adsorption–desorption equilibrium in the positive direction.The alkali dissolution rate increased slowly with increasing temperature(Fig.4(b)).Considering energy consumption and other side reactions that occur at high temperatures,a leaching temperature of 40 °C was selected.Na+adsorption–desorption is a dynamic equilibrium process.Similar to temperature,increasing the leaching time can also cause the adsorption–desorption equilibrium to move in the positive direction.However,when the adsorption–desorption process reached equilibrium (leaching time of 120 min),increasing the leaching time had negligible effect on the alkali dissolution rate (Fig.4(c)).Long-term leaching reduces the processing efficiency;hence,the leaching time was determined to be 60 min.Furthermore,the liquid–solid ratio also affects the dissolution balance of Na+.When the liquid–solid ratio increased from 1 to 5,the dissolution equilibrium moved forward,and the alkali dissolution rate reached 13.09% (Fig.4(d)).As the liquid–solid ratio continually increased,the alkali dissolution rate remained stable.Simultaneously,considering that increasing the alkalinity of the leaching solution has advantages for electrolysis,the liquid–solid ratio was set to 5.In conclusion,we verified the feasibility of the leaching section and obtained the best leaching conditions by an orthogonal experiment.

Fig.4.The effect of different leaching conditions on the Alkali dissolution rate and pH value of leaching solution.(a) Influence of leaching times.(b) Influence of leaching temperature.(c) Influence of leaching time.(d) Influence of leaching liquid–solid ratio.

3.2.Performance of the CMEC

The leaching solution obtained in the leaching section was pumped into the CMEC for membrane electrolysis.The electrolysis system based on PCME can split H2O into H+and OH-at low energy consumption (Eqs.(1)–(3)).During the constant current electrolysis process,the alkalinity in the cathode and anode zones was altered at a similar rate.Fig.5(a)illustrates that the amount of H+produced in the anolyte and OH-produced in the catholyte,both determined by acid-base titration,varied with the electrolysis reaction time.When red mud extract with 0.5 mol.L-1Na2SO4and 1 mol.L-1NaHCO3were used as the initial anolyte and catholyte,respectively,the amount of H+and OH-produced at the anode and cathode,respectively were approximately linear against time and with similar slopes.Using Faraday’s law,the current efficiency(η)of the anode and cathode zones after four hours of constant current (10 mA.cm-2) electrolysis can be calculated as 95.3% and 96.1%(Eqs.(6)-(7)),(Fig.5(b)).The reason for the decrease in electrolysis efficiency may be that a few protons generated by reduction do not react with OH-and directly penetrate into the cathode chamber instead of Na+through the membrane [28].The difference in electrolysis efficiency on the anode and cathode sides may be caused by side reactions in the anolyte and membrane permeation.

Hydrogen was consumed at the anode and released in the cathode equally,regardless of the type of cations (Na+,H+) passing through the membrane (Eqs.(1)–(3)).Therefore,there is no net production or consumption of hydrogen in the electrolysis system.To ensure that the electrolysis system runs stably for a long time,it is necessary to study the hydrogen consumption and production efficiencies of the anode and cathode,respectively.According to Faraday’s law,in the constant-current electrolysis process,the amount of hydrogen produced in the cathode region and the amount of hydrogen consumed by the anode depend solely on the total charge of the reaction,which is determined by the electrolysis current (I) and electrolysis time (t) (Eq.(8)).To verify the description of the process,the experiment quantitatively studied the relationship between the hydrogen generation and consumption efficiencies with time by GC (Eqs.(9)–(10)).The areas filled by the red and blue lines represent hydrogen consumption and hydrogen production,respectively.The efficiencies of hydrogen production and consumption fluctuated by approximately 95%(Fig.5(c)).The electrolysis efficiency does not affect the hydrogen cycled efficiency,which is consistent with our hypothetical theory.However,the hydrogen cycled efficiency deviates from the theoretical value.This may be influenced by the experimental error caused by air infiltration,hydrogen leakage in the device,or gas fluctuations in the pipeline.It is possible that the temperature affects the theoretical volume of the gas (Eq.(9)).

Fig.5.(a) Variation of alkaline concentration during electrolysis process at current density of 10 mA.cm-2.(b) Current Efficiency at the current density of 10 mA.cm-2.(c)Hydrogen cycled efficiency of CMEC at the current density of 20 mA.cm-2.The flow rate of the nitrogen carrier gas in the GC is 100 ml.min-1.The red filled area represents hydrogen consumption.The blue filled area represents the amount of hydrogen produced.(d)The U-I curves of CMEC using different NaHCO3/Na2CO3 solutions as catholyte and red mud extract with 0.5 mol.L-1 Na2SO4 as anolyte.NaHCO3/Na2CO3 were 1 mol.L-1/0 mol.L-1,0.8 mol.L-1/0.2 mol.L-1,0.6 mol.L-1/0.4 mol.L-1,0.4 mol.L-1/0.6 mol.L-1,0.2 mol.L-1/0.8 mol.L-1,and 0 mol.L-1/1 mol.L-1,respectively.(e) The U-I curves of CMEC under different flow rate operating conditions using leaching solution with 0.5 mol.L-1 Na2SO4 and 1 mol.L-1 NaHCO3 as the initial anolyte and catholyte,respectively.(f)The U-I curves of CMEC under different temperature operating conditions using leaching solution with 0.5 mol.L-1 Na2SO4 and 1 mol.L-1 NaHCO3 as the initial anolyte and catholyte,respectively.(g)The change of decomposition voltage,pH of anolyte,and pH of catholyte during electrolysis process at the current density of 10 mA.cm-2,using leaching solution with 0.5 mol.L-1 Na2SO4 and 1 mol.L-1 NaHCO3 as the initial anolyte and catholyte,respectively.(h)Comparison of voltage and pH value under different mineralization conditions.Solid lines and potted lines represent cell voltage and pH value of catholyte,respectively.Black,intermittent mineralization;blue,continuous mineralization (Using 15% volume CO2);red,continuous mineralization (Using 100% volume CO2).

Fig.5.(continued)

After verifying the reaction principle,we determined the optimal electrolysis parameters,including the concentration of the initial catholyte,flow rate,and temperature.Using NaHCO3/Na2CO3with different group distribution ratios as the catholyte for membrane electrolysis.From Fig.5(d),the curves shift to the right as the catholyte pH increases,indicating that the electrolysis energy consumption of the CMEC cell increases with the increase in ΔpH value.The law of potential growth conforms to the theoretical predictions (Eq.(12)).The initial catholyte concentration was 1 mol.L-1NaHCO3.However,there was a difference of approximately 0.38 V between the theoretical and experimental potentials.This may be affected by the overpotential of the electrodes or the membrane potential caused by the concentration difference.However,the voltage difference is still higher than that observed in other studies [25,29].There are accumulations of H+and OH-on the anode and cathode electrode surfaces during electrolysis,respectively.Therefore,the actual pH difference was much higher than that of the main body of the solution during electrolysis.Generally,the law of potential growth conforms to theoretical predictions.

Based on previous analysis results,the initial concentration of the catholyte in the CMEC was limited.Hence the experiment explored the effect of different flow rates on the performance of the electrolytic cell.As the flow rate of the anode and cathode increased,the performance of the electrolytic cell improved(Fig.5(e)).This is because increasing the flow rate accelerates the renewal of the material on the surfaces of the electrodes.Especially in the cathode zone,the hydrogen produced by electrolysis is removed in time as the flow rate increases and reduces the reaction resistance caused by the H2gas-phase barrier on the catalyst surface.In addition,given the narrow space of the CMEC flow channel,increasing the flow rate also reduces the contact resistance between the electrode and the current collector.However,the figure illustrates that the effect of increasing the flow rate on the decomposition voltage is not significant.There is also a risk of electrolytic cell leakage owing to excessive local pressure.Therefore,a flow rate of 80 ml.min-1was selected for the follow-up experiment,considering the resistance of the sealing part of the electrolytic cell to the fluid pressure.Increasing the temperature not only accelerates the reaction rate,but also speeds up the diffusion of substances during the electrolysis process.In addition,the temperature can significantly reduce the overpotential of the hydrogen evolution reaction,thereby improving the electrolysis performance.Accordingly,the effect of different temperatures on the performance of the electrolytic cell was experimentally studied(Fig.5(f)).As the reaction temperature increased,the electrolysis current increased at a similar voltage.However,the current variation is only obvious in the higher electrolysis voltage range(>0.8 V).It is important to note that the heating process requires additional energy input.Considering the energy produced by the waste heat recovery of the flue gas and the service life of the membrane in the system,it is reasonable to control the temperature in the range of 30 °C to 40 °C.

To evaluate the energy consumption in the electrolysis process,the decomposition voltage(U)and the pH of the anode and cathode with the electrolysis time(t)were measured.Refer to Faraday’s law(Eq.(8)) to calculate the time for the sodium ions in the leaching solution to be completely transferred to the cathode chamber.Then,using the electrolysis efficiency obtained in the previous experiment,the final electrolysis time was calculated to be 120 min.At the beginning of electrolysis,the initial voltage of the electrolytic cell was 0.345 V.Subsequently,the decomposition voltage rose slowly over time.Simultaneously,the pH dropped in the anolyte,and rose in the catholyte during electrolysis,which is consistent with the previous analysis results.The S-shaped fluctuation in the anolyte pH curve may have been caused by the disappearance of thebuffer system.When the pH difference increased to 2.57,the cell voltage correspondingly increased to 0.178 V.Further experiments indicated that the cell voltage increased to 0.519 V in 120 min(Fig.5(g)).There is a certain deviation between the voltage change and the theoretical value calculated by Eq.(12),which might be caused by the polarization of the electrode surface and the main body of the solution.The average voltage at current densities of 10 mA.cm-2is 0.453 V.Calculated by integrating the area of the decomposition voltage–time curve,the energy consumption corresponding to this voltage value is 130.464 J.

Moreover,the energy consumption difference between continuous and intermittent mineralization was studied.During electrolysis with bubbling CO2into the catholyte (continuous mineralization),the decomposition voltage is significantly lower than that of intermittent mineralization (bubbling CO2into the catholyte after electrolysis),especially when 100% (volume) CO2was introduced (Fig.5(h)),which is attributed to a decrease in the pH of the catholyte (Eq.(2)).When 15% (volume) CO2was blown into the catholyte,the electrolysis voltage was in the middle.Possibly,owing to the restriction of the reaction equilibrium,the pH value of the catholyte varied slightly,which is still lower than that of the intermittent mineralization process.The average electrolysis voltage at a current density of 10 mA.cm-2during the continuous mineralization process was 0.386 V (100% volume CO2) and 0.406 V (15% volume CO2),indicating that the process is more energy efficient.However,continuous mineralization requires more complex electrolysis equipment,which also increases the cost,and is difficult to control precisely.Therefore,our research focus is still on the intermittent mineralization discussed below.

3.3.CO2 absorption and NaHCO3 product

After the electrolysis was completed,the alkali-rich catholyte was utilized for the absorption of different concentrations of CO2.The pH curve of the catholyte that absorbed CO2over time was recorded (Eq.(2)).As the absorption reaction progressed,the pH of the solution gradually decreased.Within 10 min of the absorption reaction,the pH of the catholyte decreased significantly and then slowly decreased.Our analysis indicates that in the initial stage of the reaction,the OH-concentration in the solution was high,and it reacted quickly with CO2,causing a rapid drop in the pH value(Fig.6(a)).After 10 min,as the alkalinity of the catholyte decreased,the main reaction may be the process ofwater solution absorbing CO2.The reaction rate was relatively slow.It was observed that the solution bubbling with different concentrations of CO2gradually became turbid at 75 and 170 min,respectively,and a numerous solids precipitated.This process can be considered as NaHCO3reaches the supersaturated state in the solution first,and finally precipitates out in large quantities as the crystals continue to nucleate and grow.It was verified that 15%volume CO2(the concentration of CO2in the flue gas)can be utilized in the electrochemical CO2mineralization system,which is of great significance for industrial applications.Given the attractiveness of directly mineralizing flue gas CO2(no capture costs),the system feed gas is set to flue gas CO2.

To analyze the composition of the crystalline product,the precipitated solid was suction filtered and then vacuum dried at 45 °C for 72 h to remove water.After the calculation,the reaction yield was obtained.The characterization results for the precipitated product are presented in Fig.6.The XRD analysis results indicate that the product was highly pure NaHCO3(Fig.6(b)).The TGA results(Fig.6(c))illustrate that the purity of sodium bicarbonate is approximately 99.42%.SEM photographs of the final mineralization products indicate that the surface morphology of sodium bicarbonate is in the shape of a long strip (Fig.6(d)).

Fig.6.(a)The pH of the catholyte changes with time during the CO2 absorption process.(b)XRD patterns of solid product formed in the crystallizer and after drying under 45℃for 48 hours.(c) The thermo-gravimetric analysis of solid product.(d) SEM of mineralization products.

3.4.Electrolytic consumption evaluation

The energy consumption was estimated by Eqs.(11)–(17) as 0.039 GJ.t-1red mud under the operating conditions of 10 mA.cm-2,assuming that all soluble sodium ions were converted to sodium bicarbonate.The mass balance revealed that the energy consumption per ton of CO2for mineralization by this process is 1.76 GJ,simultaneously producing 1.91 tons of 99.4%pure sodium bicarbonate.Conventional mineralization of red mud generally needs to be performed in a high-temperature and high-pressure operating environment [22,30].The proposed process has the potential for large-scale applications owing to its low energy consumption and outstanding controllability.

4.Conclusions

In this study,a novel process employing electrochemical CO2mineralization for red mud treatment,driven by hydrogen-cycled membrane electrolysis is proposed,which exhibits excellent resource recovery ability and energy savings operating potential under mild conditions,compared with the conventional red mud mineralization method.The optimal leaching conditions were obtained by analyzing the alkali dissolution rate and existing forms of alkaline substances in the red mud.In the CO2mineralization electrolytic cell,H+and OH-were formed under a low cell voltage.H+was utilized to neutralize the basic leaching solution,and a basic catholyte was utilized to precipitate NaHCO3by absorbing CO2.As the electrolysis proceeded,the electrolysis efficiency dropped to 95.3%,but the proton cycled rate remained stable,which proved the cycle stability of the system.It is verified the soluble alkali in the red mud was successfully processed and converted into highly purified products utilizing flue gas CO2(15%volume).This process consumes 0.039 GJ per ton of red mud for mineralization treatment,and the treated red mud can be utilized as a construction material.In this process,1.91 tons of sodium bicarbonate is synergistically produced for every ton of CO2mineralized.However,owing to the complex composition of red mud,the influence of impurity ions in the leaching solution on the long-term stability of the electrode needs to be ascertained before commercialization is possible.Therefore,further studies on the structure–activity relationship between impurity ions and stability will make it more suitable for industrial scale-up.Innovative research provides a new strategy for direct efficient mineralization alkaline waste by flue gas CO2.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was funded by the Science and Technology Department of Sichuan Province(2020YFH0012).We also thank the Analysis and Testing Center of Sichuan University and Ceshigo Research for help in characterizations.