Oxygen‐Stable Electrochemical CO₂ Capture using Redox‐Active Heterocyclic Benzodithiophene Quinone

Abstract

Electrochemical carbon capture offers a promising alternative to thermal amine technology, which serves as the traditional benchmark method for CO₂ capture. Despite its technological maturity, the widespread deployment of thermal amine technologies is hindered by high energy consumption and sorbent degradation. In contrast, electrochemical methods, with their inherently isothermal operation, address these challenges, offering enhanced energy efficiency and robustness. Among emerging strategies, electrochemical carbon capture systems using redox‐active materials such as quinones stand out for their potential to capture CO₂. However, their practical application is currently limited by their low stability in the presence of oxygen. We demonstrate that benzodithiophene quinone (BDT‐Q), a heterocyclic quinone, exhibits high stability in electrochemical carbon capture processes with oxygen‐containing feed gas. Conducted in a cyclic flow system with a simulated flue gas mixture containing 13 % CO₂ and 3.5 % O₂ for over 100 hours, the process demonstrates high oxygen stability with an electron utilization of 0.83 without significant degradation, indicating a promising approach for real world applications. Our study explores the potential of new heterocyclic quinone compounds in the context of carbon capture technologies.

CO₂ capture was achieved with over 90 % capacity utilization in a 100‐hour cycle using a heterocyclic benzodithiophene quinone in a flow system containing 13 % CO₂ and 3.5 % O₂, demonstrating its potential for real‐world applications. This process was investigated through a combined experimental and computational approach to elucidate the kinetic and thermodynamic mechanisms underlying the carbon capture efficiency.

Introduction

Carbon dioxide (CO₂) emissions are known as a major contributor to the Greenhouse Effect, significantly driving climate change. ^[1] There is a clear need to develop CO₂ capture, utilization, and storage (CCUS) ^[2] technologies to mitigate anthropogenic CO₂ emissions. Traditional approaches to carbon capture such as thermal solvent scrubbing with aqueous amine solutions, ^[3] present high operating energy costs and environmental concerns due to the thermal decomposition of amine solutions (e.g., monoethanolamine (MEA)), limiting their wider application.[ ⁴ , ⁵ , ⁶ ] In contrast, electrochemically mediated carbon capture (EMCC) emerges as a promising alternative to conventional amine‐based thermal swing methods.[ ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ ] EMCC can be operated under isothermal conditions and integrated with low‐carbon high‐capacity energy sources as industries move towards electrification. However, challenges remain in developing electrochemical methods that are not only energetically efficient but also demonstrate long term stability under the operational conditions for real world application.

Among the different approaches employed in EMCC systems, the use of redox‐active compounds such as quinone,[ ¹² , ¹³ , ¹⁴ , ¹⁵ ] phenazine, ^[16] pyridine,[ ¹⁷ , ¹⁸ ] and disulfide derivatives ^[19] has shown considerable promise in the field due to their molecular flexibility that allows tuning to optimize the performance metrics of the electrochemical carbon capture system. The quinone compounds, in particular, are of considerable interest due to their exceptional tunability, and electrochemical and chemical reversibility involving two‐electron transfers. ^[20] However, their capture capacity degradation in the presence of oxygen ^[21] via reoxidation of the activated quinone dianion has prevented long‐term operation in the presence of oxygen. ^[22] Despite their potential, only a few studies have investigated electrochemical CO₂ capture and release from oxygen containing feed gas streams.[ ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ ] Given the presence of oxygen in both flue gas and air, the design of a practical system demands oxygen stability of the activated quinone dianion to develop an electrochemical system with enhanced performance metrics including energy efficiency, long lifetime, and cost‐effectiveness for real‐world applications.

In addition to achieving oxygen stability, fine‐tuning of the physicochemical properties of redox‐active materials is crucial for optimizing their performance by modulating the binding constant (K _CO2) for CO₂ capture via direct CO₂ binding. ^[26] The linear free‐energy relationship between redox potential and CO₂ binding strength shows a negative correlation, indicating that shifting the redox potential to more positive values weakens CO₂ binding, thereby sacrificing capture capacity and kinetics. ^[28] Therefore, optimization is essential to ensure aerobic stability, high capture capacity, and kinetics. ^[25] The calculation of K _CO2, as introduced by Bell et al., ^[29] quantifies the binding strength of CO₂ captured by sorbents under varying partial pressures of CO₂. To effectively capture the atmospheric CO₂ with the concentration of 426 ppm, the log of the K _CO2 value must surpass approximately 5.5. ^[30]

Although heterocyclic quinones, particularly benzodithiophene‐based compounds, have demonstrated promising performance in solar cell applications due to their electronic tunability and molecular stability, they have yet to be explored for CO₂ capture applications.[ ³¹ , ³² , ³³ ] In this study, we demonstrate electrochemical carbon capture systems that enhance oxygen stability by employing a redox‐active heterocyclic benzodithiophene quinone (BDT‐Q). We hypothesized that the sulfur group in BDT can extend the π‐conjugation system, aiding in the distribution and stabilization of charges or radicals during electroreduction, especially around the C=O bond. This interaction may enhance the molecule‘s ability to interact with CO₂, improve resistance to oxidative degradation, and contribute to more effective CO₂ capture.

Through detailed studies, we found that BDT‐Q demonstrates promising oxygen stability under simulated flue gas conditions with 3.5 % O₂ in a cyclic flow system for over 100 hours. A systematic comparison with a well‐studied anthraquinone (AQ)[ ¹⁴ , ³⁴ , ³⁵ , ³⁶ ] under identical conditions using an H‐cell indicates a high oxygen stability of BDT‐Q relative to that of AQ with an electron utilization of 0.83. Additionally, to gain insight into the reaction mechanism and the thermodynamic aspects of the CO₂ capture system, we further applied a combination of experimental and computational techniques to investigate the bimolecular rate constant (k _f=1.056 M⁻¹ s⁻¹), and equilibrium constant (log K _CO2=7.6) between CO₂ and BDT‐Q dianion species.[ ³⁷ , ³⁸ ]

Results and Discussion

The electrochemical mechanism of BDT‐Q for the CO₂ capture and release is described in Figure 1a. The initial one‐electron transfer to BDT‐Q generates a semiquinone radical (BDT‐Q⋅⁻), which is then further reduced through a second one‐electron transfer, resulting in the formation of quinone dianion (BDT‐Q²⁻). The quinone dianion serves as a CO₂ sorbent, leading to the formation of BDT‐Q(CO₂)₂ ²⁻.[ ¹² , ¹⁴ ] To corroborate the mechanism, the Gibbs free energies (ΔG) of electron transfer and the CO₂ binding steps were calculated using Density Functional Theory ($\omega$ B97M‐V/def‐2‐tzvpd) with SMD solvent model. A comparison between the calculated ΔG values suggests that CO₂ binding to BDT‐Q²⁻ (ΔG: −7.8 kcal/mol), is significantly more favorable than to BDT‐Q⋅⁻ (ΔG: +17 kcal/mol). The second CO₂ capture by BDT‐Q (CO₂)²⁻ with a ΔG of +3.1 kcal/mol is weaker than the initial capture (−7.8 kcal/mol). DFT investigations summarized in Figure S1 further examined the formation of an unbound semiquinone‐CO₂ complex (S2), indicating more favorable formation (+6.5 kcal/mol) compared to the CO₂ bound semiquinone (S1). However, the unbound state requires a more negative potential ($E_{0/-1}^0-0.4\mathrm{V}$ ) to be reduced to the dianion adduct form, BDT‐Q (CO₂)²⁻.

Figure 1.

(a) Schematic illustration of electroreduction reaction of BDT‐Q to BDT‐Q⋅⁻ and BDT‐Q²⁻ followed by CO₂ binding to BDT‐Q²⁻. Black arrows represent electron transfers, while the red arrows represent CO₂ binding. The associated free energy values for CO₂ binding during each step are indicated. (b) Cyclic voltammetry (CV) comparison of 10 mM BDT‐Q under N₂ (gray) and CO₂ (red) in 0.1 M TBAClO₄ as a supporting electrolyte dissolved in DMF, with a scan rate of 100 mV/s.

Experimental investigations were carried out using cyclic voltammetry (CV) (Figure 1b and Figure S2). The CV was measured using a three‐electrode system, including a glassy carbon working electrode, Ag|AgCl reference electrode, and Pt wire counter electrode. A solution of 10 mM of BDT‐Q in dimethylformamide (DMF) was utilized, with the addition of 0.1 M tetrabutylammonium perchlorate (TBAClO₄) as a supporting electrolyte. Under a nitrogen (N₂) atmosphere (gray curve), the CV shows characteristic two stepwise single‐electron transfers at E_red,peak,1=−1.02 V and E_red,peak,2=−1.81 V vs Fc⁺/Fc in DMF, resulting in two quasi‐reversible redox peaks corresponding to the formation of the semiquinone radical (BDT‐Q⋅⁻) and quinone dianion (BDT‐Q²⁻), respectively. Upon introducing 100 % CO₂ into the solution (red curve), a noticeable positive shift of the second reduction peak by 370 mV to −1.44 V vs Fc⁺/Fc was observed, while the position of the first reduction peak remained unaffected, albeit with an increase in current. The significant shift in the second reduction peak indicates that BDT‐Q²⁻ is the primary species responsible for capturing CO₂ at a more negative potential, which is well‐aligned with the computational study described in Figure 1a.

Further CVs were collected to probe the electrochemical behavior of BDT‐Q and AQ (Figures 2a–2c). BDT‐Q (black curve) under N₂ shows both quasi‐reversible one‐electron transfer redox peaks at a more positive potential than AQ (grey curve) (Figure 2a). Under a CO₂ atmosphere (Figure 2b), the second one‐electron transfer peak of BDT‐Q at E _red,peak,2=−1.44 V vs Fc⁺/Fc appears at a more positive potential than that of AQ at E _red,peak,2=−1.63 V vs Fc⁺/Fc highlighting that more energy is required to reduce AQ compared to BDT‐Q. For a comprehensive comparison, AQ was fully reduced at −10 mA to form AQ²⁻ (light red curve) and the results were compared with the fully reduced BDT‐Q²⁻ (dark red curve) as shown in Figure 2c. The AQ and BDT‐Q exhibit reduction peak potentials at −1.63 and −1.24 V vs Fc⁺/Fc, respectively, reaffirming the formation of CO₂ binding (BDT‐Q‐(CO₂)₂) at a more positive potential in the case of BDT‐Q compared to AQ. This may be due to the heterocyclic structure of BDT‐Q, where the presence of sulfur could enhance charge distribution within the molecule, potentially improving the CO₂ capture process and augmenting BDT‐Q′s effectiveness. This phenomenon was further investigated using DFT calculations. As summarized in Table S1, the reduction potential of the AQ semiquinone CO₂ adduct is 0.18 V more positive than AQ′s first formal redox potential. This explains the partial overlap of peaks seen in the CVs in Figures 2b and 2c. ^[39]

Figure 2.

Electrochemical carbon capture study of redox‐active benzodithiophene quinone (BDT‐Q). Cyclic voltammetry (CV) comparison of 10 mM (a) BDT‐Q and AQ under N₂; (b) BDT‐Q and AQ under CO₂; (c) BDT‐Q and AQ after complete electroreduction at −10 mA under CO₂; and (d) BDT‐Q under different CO₂ concentrations of 15 %–100 % at a scan rate of 100 mV/s. All the CV experiments were conducted using a 10 mM BDT‐Q/AQ in 0.1 M TBAClO₄ as a supporting electrolyte dissolved in DMF, with a scan rate of 100 mV/s. (e) Chronoamperometry (CA) of BDT‐Q under applied potentials V₁ under N₂ and CO₂; and V₂ under N₂. (f) Determination of k_f for the addition of CO₂ to BDT‐Q⋅⁻ under various CO₂ concentrations of 15 %, 50 %, and 100 %.

The binding constant between the dianion BDT and CO₂, K _CO2, was determined from the peak shift of the second reduction wave at increasing CO₂ concentration (Figure 2d), given by Eq. 1:

(1)

yielding a value of log K _CO2 of 7.61 (Figure S3). ^[38] Here, R is the gas constant, T is temperature and F the faraday constant; n=2 is the number of electrons transferred. Additionally, the rate of reaction between the semiquinone species and CO₂ was determined from chronoamperometry (CA) measurements of BDT‐Q saturated with N₂ and subsequently 15 %, 50 %, and 100 % CO₂ (Figure 2d–2e). ^[40] This method, the details of which are given in the Supporting Information, relies on the integrated Cottrell equation relating the time and charge delivered when operating at constant potential for a diffusion controlled process to determine the constant k, under N₂ at an applied voltage V ₁. Chronoamperometry at V ₁ reflects BDT‐Q reduction to BDT‐Q⋅⁻. When operating at V ₁ in the presence of dissolved CO₂, BDT‐Q(CO₂)₂ ²⁻ forms as the monoadduct semiquinone is reduced, resulting in larger currents than at V ₁ under N₂ (Figure 2e and Figure S4). As summarized in the Supporting Information, the current i and time t are related non‐linearly via Eq. 2:

$${\displaystyle -\ln \left(2-\frac{i{t}^{1/2}}{k}\right)=k_{f}t}$$ (2)

where k is obtained from charging at V ₁ under N₂. Thus, a plot of $-\ln \left(2-i{t}^{1/2}/k\right)$ against time can be used to extract the pseudo‐first order rate constant, $k_f$ , between CO₂ and the semiquinone (Figure 2f and Figure S2). The bimolecular rate constant, $k_b=k_f/\left[{CO}_2\right]$ , is estimated from the slope of a plot of k_f against to [CO ₂] be 1.056 M⁻¹ s⁻¹, in line with findings from prior research (Figure S5). ^[37]

To further investigate the binding of CO₂ to BDT‐Q, several analytical techniques, including ultraviolet‐visible spectroscopy (UV/Vis), nuclear magnetic resonance (NMR), and quartz crystal microbalance (QCM) were employed as depicted in Figure 3. For this purpose, 25 mM BDT‐Q in 5 mL of DMF was electrochemically reduced at a constant current of −10 mA for 32 minutes under N₂ and CO₂ individually, equivalent to transfer of 0.2 mmol of electrons. This reduction resulted in the major formation of BDT‐Q⋅⁻, causing the solution color to change from yellow to dark green (Figure 3a). Subsequently, upon oxidation at 10 mA for 32 min, the color reverted to its original light yellow. To examine the optical absorption changes of BDT‐Q solutions by electrochemical reduction and oxidation, a series of UV–vis absorption spectra were measured under N₂ and CO₂ (Figures 3b and Figure S6). The UV/Vis spectrum of BDT‐Q before electroreduction shows peaks at 290 and 344 nm.[ ⁴¹ , ⁴² ] After electrochemical reduction under N₂, these peaks decreased in intensity, and new peaks at 409 and 432 nm, along with a twin band at 577 and 623 nm, appeared, presumably from BDT‐Q⋅⁻ and BDT‐Q²⁻ formation. ^[42] These peaks red‐shifted to 579 and 625 nm, respectively, with a decrease in absorbance intensity under CO₂, could be attributed to the formation of new adducts, including BDT‐Q(CO₂)₂ ²⁻.

Figure 3.

(a) The observed colour changes during electrochemical reduction and oxidation reactions of 25 mM BDT‐Q in 5 mL of DMF with 0.25 M TBAClO₄ as a supporting electrolyte, using an H‐cell setup with a carbon felt working electrode and a zinc wire counter electrode. (b) UV–vis absorption spectrum of BDT‐Q comparison during electro‐reduction/oxidation processes under N₂ and CO₂. (c) Dynamics of CO₂ release from BDT‐Q (CO₂)₂ ⁻ monitored by in situ electrochemical quartz crystal microbalance (QCM) at the applied potential of −1.1 V vs Fc⁺/Fc. (d) Stacked ¹³C NMR spectra comparison of 10 mM BDT‐Q in 0.1 M TBAClO₄ in d₆‐DMSO from top to bottom in d₆‐DMSO: blank BDT‐Q before the electrochemical reaction, BDT‐Q purged with CO₂ under no reaction, full electroreduction of BDT‐Q under CO₂ to form BDT‐Q (CO₂)₂ ²⁻, and full electroreduction of BDT‐Q under N₂ to form BDT‐Q²⁻ (a trace amount of HCl was added to the solution before the NMR study to facilitate the peak splitting).

The reaction behavior of BDT‐Q during electrochemical reduction was further elucidated using ¹³C NMR (Figure 3d and Figure S7) and ¹H NMR (Figure S8 and S9) spectroscopy. As shown in Figure 3d, the peak at 174 ppm corresponds to the carbon of −C=O (red asterisk) of BDT‐Q, observed in the starting materials before reduction. Notably, bubbling the solution with CO₂ did not shift the carbon peak, indicating no capture of CO₂ by BDT‐Q in its neutral form. Upon complete electroreduction of quinone under CO₂, the peak at 174 ppm disappeared, while a new peak at 162 ppm appeared (blue asterisk), which belongs to the carbon of CO₂ captured by the quinone in BDT‐Q (CO₂)₂. This peak is absent in the case of fully reduced BDT‐Q under N₂, which confirms the conversion of BDT‐Q²⁻ to BDT‐Q(CO₂)₂ ²⁻. Notably, in ¹H NMR in Figure S9, the partially electrochemically reduced BDT‐Q solution, clearly shows the mixture of BDT‐Q²⁻ and BDT‐Q(CO₂)₂ ²⁻ formation under electroreduction, which confirms the binding of CO₂ with quinone dianion (BDT‐Q²⁻).

Additionally, QCM ^[43] was employed to detect the mass changes on the quartz crystal surface (Figure 3c and Figure S10). The change in the resonant frequency of the QCM (ΔF) correlates directly with the change in mass (Δm), which can be calculated using the Sauerbrey equation ^[7] (see Supporting Information). In Figure 3c, the frequency changes of the QCM at −1.1 V vs Fc⁺/Fc for a period of time showed a decrease in mass (~54 ng/cm²), with the curve reaching a plateau upon removal of the potential; the frequency changes again on resumption of the applied potential. This consistent trend confirms the release of captured CO₂ from BDT‐Q (CO₂)₂ ²⁻ under the relatively oxidizing conditions at −1.1 V vs Fc⁺/Fc which not only validates BDT′s ability to capture CO₂ but also highlights its potential as a CO₂ capture agent.

CO₂ Capture and Release using an H‐cell

The CO₂ capture/release properties of BDT‐Q/BDT‐Q²⁻ redox cycle were evaluated through electrochemical reduction/oxidation reactions, as illustrated in Figure 4. The reaction took place in a sealed H‐cell, featuring two 10 mL chambers (Figure 4a). The cathodic chamber was equipped with graphite felt as a working electrode and 25 mM BDT‐Q in 5 mL of DMF containing 0.25 M TBAClO₄ as a supporting electrolyte. In the anodic chamber, a zinc (Zn) wire served as the counter electrode placed in 5 mL of DMF containing 0.1 M TBAClO₄. Zinc oxidizes at the anode, releasing two electrons necessary for generating BDT‐Q²⁻ and BDT‐Q⋅⁻, thus facilitating the electrochemical reactions.[ ⁴⁴ , ⁴⁵ ] To ensure the optimal electrolyte for our system, we first investigated the effect of electrolyte salts, focusing on the cation's influence on thermodynamics and kinetics (Figure 4b, and Figure S11–S14). ^[46] While the impact of mono‐cations on quinone‐mediated CO₂ capture is well‐studied,[ ⁴⁰ , ⁴⁴ ] evaluation of di‐cation effects remains relatively unexplored.⁴²

Figure 4.

The H‐cell experiment illustrates the electrochemical CO₂ capture capacity comparison of BDT‐Q under CO₂‐O₂ gas mixtures. (a) Schematic of the electrochemical H‐cell experimental setup for CO₂ release. The cathodic compartment was connected to a gas flow meter and an FT‐IR CO₂ sensor, featured 25 mM BDT‐Q and 0.25 M TBAClO₄ in 5 mL DMF, with a carbon felt working electrode. The anodic compartment included 0.25 M TBAClO₄ in 5 mL DMF with Zinc (Zn) serving as a counter electrode. (b) The Cyclic Voltammetry (CV) comparison of 10 mM BDT‐Q in DMF with 0.1 M electrolytes: Ba (ClO₄)₂ (blue), LiClO₄ (black), and TBAClO₄ (pink) at 100 mV/s scan rate under CO₂. (c) CV comparison of 10 mM BDT‐Q under pure 15 % CO₂ (dark red), 13 % CO₂ and 3.5 % O₂ (light red), and O₂ (green). The CO₂ captured and released of BDT‐Q under (d) 15 % CO₂; (e) 13 % CO₂ and 3.5 % O₂; (f) 9.2 % CO₂ and 8.1 % O₂, and (g) under 9.2 % CO₂ and 8.1 % O₂ for AQ.

The electrochemical behavior of BDT‐Q in the presence of various mono and di‐valent cations was compared with the response when TBA⁺, a largely non‐interacting cation that served as the reference case, was used (Figure S11–S14). Li⁺ and Ba²⁺ were found to be the most strongly associating among of those examined, exhibiting the most anodic shift of the second reduction under nitrogen. Ca²⁺ and Mg²⁺ significantly shift the quasi‐reversible second reduction peak of BDT‐Q under nitrogen, but the oxidative features were suppressed with Mg and CO₂,[ ⁴⁷ , ⁴⁸ ] and all features are absent with Ca and CO₂. Li and Ba cations also strongly associate with the CO₂ adduct of BDT‐Q²⁻, as indicated by the positively shifted oxidation wave under CO₂.[ ⁴⁸ , ⁴⁹ , ⁵⁰ ] We hypothesize that these strong interactions between supporting cations and the reduced quinone could diminish the driving force for CO₂ adduct formation, thereby impeding reaction kinetics. ^[49] While the presence of strongly associating cations can mitigate oxygen sensitivity, it also hinders CO₂ absorption kinetics and increases the minimum thermodynamic separation work due to significant shifts in the oxidation features.[ ⁵⁰ , ⁵¹ , ⁵² ] Given these considerations, we chose TBA⁺ as the benchmark cation to further evaluate the electrochemical behavior of BDT‐Q in the context of electrochemically mediated carbon capture for the remainder of this study. As shown in Figure 4c and Figure S16, using TBAClO₄ electrolyte, BDT‐Q undergoes reduction under CO₂ at the E _1/2 of −1.24 V vs. Fc+/Fc while O₂ reduction occurs at a more negative potential of −1.29 V vs. Fc⁺/Fc. Although the difference is relatively small, it may enhance the compound‘s stability in the presence of oxygen during CO₂ capture. To explore this further, we conducted a systematic study to examine the capture capacity of BDT‐Q in the presence of varying concentrations of O₂.

The amount of CO₂ captured and released during the reduction and oxidation cycles was achieved through chronopotentiometry under a constant current of $\pm$ 10 mA in the H‐cell. Each capture/release cycle lasted 32 minutes, followed by a 30‐minute rest period between the redox steps. Initially, the electrochemical reaction was carried out with bubbling of 15 % CO₂ excluding any air or oxygen involvement over five cycles (Figure 4d). To gain preliminary results of the oxygen stability of BDT‐Q, additional experiments with exposure to 3.5 % O₂ and 13 % CO₂ (Figure 4e), and subsequently increasing the O₂ concentration to 8.1 % while decreasing the CO₂ concentrations to 9.2 % (Figure 4f). The average values of molecular capture capacity (number of CO₂ per BDT‐Q) over a period of 10 hours were 1.81 (15 % CO₂), 1.67 (13.3 % CO₂+3.5 % O₂) and 1.43 (9.2 % CO₂+8.1 % O₂). The corresponding electron utilizations for each condition were 0.9, 0.83, and 0.7, respectively (Figure S17). For a systematic comparison, the well‐studied AQ[ ²¹ , ²⁷ , ⁵³ , ⁵⁴ ] was subjected to the same conditions at the current of $\pm$ 10 mA for the electroreduction and oxidation process (Figure 4g). In this case, the amount of CO₂ captured and released exhibited a notable decline from the second cycle onwards, attesting to its well‐known sensitivity to the presence of oxygen.

While BDT‐Q exhibits high oxygen stability under 13.3 % CO₂ and 3.5 % O₂, increasing the O₂ concentration to 8 % results in a slight decrease in overall CO₂ capture capacity. This suggests that although BDT‐Q demonstrates high oxygen resistance, it is somewhat sensitive to higher O₂ concentrations. Following these preliminary observations, we advanced our study to simulate flue gas conditions with 13 % CO₂ and 3.5 % O₂ to further evaluate the system‘s performance under realistic operating conditions.

CO₂ Capture and Release using a Cyclic Flow System

Encouraged by results obtained from experiments in an H‐cell, we extended our investigation to probe the capture and release of CO₂ under aerobic conditions in a cyclic flow cell system to gauge the long‐term stability of the BDT‐Q based system for an extended period of operation. This setup involved a cyclic flow of liquid phase, with continuous measurement of CO₂ capture and release (Figure S18). The symmetric cyclic flow cell featured electrolyte flow paths with anodic and cathodic chambers connected to two 8 mL septum‐sealed vials, each containing a solution of BDT‐Q (25 mM) and TBAClO₄ (0.25 M) in 6 mL DMF, separated by a Nafion 115 membrane (Figures 5a). A Toray carbon paper electrode (TGP‐H‐060, 5 % wet Proofing from Fuel Cell Earth) was used as the working electrode for each of anodic and cathodic reactions.

Figure 5.

The cyclic flow system experiment illustrates the electrochemical CO₂ capture capacity of 25 mM BDT‐Q under 13 % CO₂ and 3.5 % O₂ gas mixtures. The setup includes two septum‐sealed vials, each containing a solution of BDT‐Q (25 mM) and TBAClO₄ (0.25 M) in 6 mL DMF, separated by a Nafion membrane. (a) Schematic of flow cell experimental setup for CO₂ capture and release. (b) CO₂ reading and voltage curve at the exit of the headspace of the reaction over 20 cycles. Electrochemical CO₂ capture and release capability of BDT‐Q in 13 % CO₂ and 3.5 % O₂ mixtures, measured over 100 hours: (c) Electron utilization values based on the amount of CO₂ per electron transferred during the capture and release process. (d) Molecular capture capacity of BDT‐Q over 20 cycles.

The 6 mL solution in each vial was continuously bubbled with the feed gas containing 13 % CO₂ and 3.5 % O₂ (5 : 1 in‐line mix of 15 % CO₂ and ambient air), directed individually to the cathodic and anodic chambers. Each capture/release cycle lasted 72.5 minutes at 10 mA, followed by a 120‐minute rest period between cycles. The polarity of the electrodes was reversed from one cycle to the next so that what was the cathode during the previous cycle became the anode in the current cycle, and vice versa. The CO₂ concentration in the exiting gas stream shown in Figure 5b was monitored using an FTIR CO₂ sensor. The consistent CO₂ response observed during repetitive electrochemical cycling throughout the entire reaction period (Figure S19) serves as additional evidence reinforcing the high stability of BDT‐Q for carbon capture under aerobic conditions

An average CO₂ molecular capture capacity of $\sim$ 1.6 and electron utilization of $\sim$ 0.80 was achieved over the 20 cycles (Figures 5c and 5d). These numbers are close to the results from the H‐cell which suggests that the solution retained its efficiency and effectiveness over 100 hours of cyclic operation, highlighting the robustness of the system during the CO₂ capture/release process over the 20 cycles. The electrochemical work of 108 kJ_e/mol was calculated from the electron utilization ${ϵ}_{{CO}_2}(\sim 0.80)$ and the peak potential difference $E_{cell}\sim 0.93V$ in the CV using Eq. 3:

$${\displaystyle W=\frac{F}{{ϵ}_{{CO}_2}}{E}_{cell}}$$ (3)

These results highlight the reversible capture and release of CO₂ by BDT‐Q with no significant degradation in capacity, confirming its promising stability under aerobic conditions.

Summary and Conclusion

In this study, we introduced a heterocyclic quinone derivative for a direct carbon capture system with relatively high oxygen stability. Along with fundamental studies and advanced techniques, including NMR and QCM, as well as computational studies, we sought to understand the reaction mechanism. We also employed both H‐cell and custom‐designed cyclic systems to investigate the CO₂ capture capacity of BDT‐Q. We established the oxygen stability of BDT‐Q over 20 cycles at a current density of 32 mA/cm² under simulated flue gas conditions containing 13 % CO₂ and 3.5 % O₂, with an electrochemical work of 108 kJ/mol. These promising outcomes, coupled with the demonstrated stability of the BDT‐Q/BDT‐Q²⁻ redox cycle, underscore the need for further investigation. Future work should focus on extending the evaluation to a larger number of cycles, probing long‐term stability, and examining the oxygen stability of BDT‐Q under more rigorous conditions, including higher oxygen concentrations. Our ongoing studies, utilizing both in situ and ex situ techniques, aim to provide a comprehensive understanding of the reaction intermediates and their interactions with reduced BDT‐Q, thereby elucidating the mechanisms underlying its oxygen stability. By expanding this investigation to include a wider range of conditions and techniques, we anticipate gaining deeper insights into the oxygen stability and performance of BDT‐Q. Overall, this work positions BDT‐Q as a leading candidate within the heterocyclic quinone family for advanced electrochemical CO₂ capture, particularly in oxygen‐rich environments.

Conflict of Interests

The authors declare no conflict of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

Abdinejad M., Massen-Hane M., Seo H., Hatton T. A., Angew. Chem. Int. Ed. 2024, 63, e202412229. 10.1002/anie.202412229

Data Availability Statement

The data supporting the findings of this study are available within the paper and its Supporting Information.