New Molecular Materials for Direct Air Capture of Carbon Dioxide Using Electro-Swing Chemistry

Featured Application

Capture of carbon dioxide directly from the atmosphere using electricity-driven processes.

Abstract

The increasing amount of carbon dioxide (CO₂) in the atmosphere is the main factor contributing to climate change. Recent studies have determined that simply reducing emissions is insufficient to restore the Earth’s atmospheric system—negative emissions are therefore necessary. Current carbon (i.e., CO₂) capture technologies use thermal or pressure swings. These approaches suffer from low energy efficiency, high cost, and geographic constraints. Electro-swing chemistry-based carbon capture has emerged as a promising potential solution to these challenges. However, strong CO₂-binding sorbents, not susceptible to oxygen interference, remain elusive. In this study, three electron-deficient quinones were designed and tested as CO₂ capture molecular sorbents. Cyclic voltammetry (CV) on these novel quinones reveals that 2,3-dicyano-1,4-benzoquinone (DBQ) has a second reduction potential positive of that of oxygen reduction. Moreover, this sorbent binds to CO₂ with a free energy ΔG_bind of −5.39 kcal/mol when activated by electrochemical reduction. These results suggest that DBQ may be a sorbent candidate that can capture >70% of CO₂ in the current atmosphere using electro-swing chemistry without the interference of oxygen in the air. This novel sorbent can be further developed for large-scale carbon capture and combating global warming and its associated impacts.

1. Introduction

From the pre-industrial era to modern-day, atmospheric carbon dioxide (CO₂) concentrations have increased by 50%, from 280 ppm to 420 ppm, and climate scientists predict further increases over the next two decades [1,2]. Climate change leads to severe weather events, disruptions of human life, and threats to fragile ecosystems [3,4,5]. To mitigate these consequences, 196 parties signed the Paris Agreement in 2015, establishing a goal to limit the global temperature rise to no more than 2 °C above pre-industrial levels. However, recent studies suggested that a 1.5 °C rise represents a critical threshold before the impacts of climate change become progressively severe. It was determined earlier this year that simply reducing emissions is insufficient to restore Earth’s atmospheric systems—negative emissions are now necessary [6,7,8].

The current carbon capture, utilization, and storage (CCUS) industry typically uses a temperature and/or pressure swing (Figure 1a) to achieve CO₂’s capture and release. Within these systems, amine-based sorbents [9] or strong base sorbents [10,11] can capture CO₂ at high pressure or low temperature, then release the CO₂ under reduced pressure or high temperature. In addition, sorbents are often deployed near concentrated emission sources, since high concentrations of CO₂ are needed for them to work; the use of thermo/pressure swings consumes large amounts of energy, so these processes are considered “energy intensive”. Hence, these types of sorbents are not suitable for direct air capture (DAC), and different chemistries must be considered to address the capture of CO₂ from the air [12,13].

Learning from the ideas surrounding industrial electrification, in which traditional energy-intensive processes are powered by renewable (solar, wind, and hydro) energy, new systems based on electricity-driven chemical reactions allow us to tune the energetics of the process using molecular design. Recent reports showed that an electro-swing process (Figure 1b) using organic molecules (e.g., quinones and isoindigos) can capture and release CO₂ in a highly energy-efficient manner [14,15,16,17]. For example, quinone sorbents (Figure 2a) can capture CO₂ when reduced and release it when oxidized. However, strong sorbents for capturing low concentrations of CO₂ from the air using electrochemistry without the interference of O₂ (E_1/2 ≈ −1.2 V) remain elusive, preventing the scaleup of this technology. Our previous studies estimated that, if the goal is to remove a fraction of atmospheric CO₂ (since air is plentiful) rather than all of it, sorbents with binding energies as low as −4.70 kcal/mol can be used [18,19]. This threshold value allows us to identify novel molecules that bind CO₂ from a fundamental physical organic chemistry perspective. Herein, we report the design of a novel, potentially oxygen-stable quinone sorbent that may be suitable for DAC of CO₂ using electro-swing chemistry.

2. Molecular Design of New Materials

Our recent studies on seven quinones showed that there is a linear correlation between the second reductive potential of the quinones and their free energy of binding to CO₂ [20]. This observation provides a predictive estimation of free energies of CO₂ binding. In addition, the second reduction potential is dependent on the electronic property of the substitutions on the quinone (Figure 2b). To overcome the oxygen interference problem for aerobic operation, we decided to continue to investigate additional quinones and identified three quinones with well-defined structures (Figure 2c) based on the previously acquired knowledge on their structure-function relationship [20]. This study is an extension of the findings of [20]. 2,3,5,6-Tetrafluoro-1,4-benzoquinone (F₄Q) was selected due to the increased electronegativity of F compared to Cl, possibly leading to a positive shift in the second reduction potential of 2,3,5,6-tetrachloro-1,4-benzoquinone (Cl₄Q). 2,3-Dicyano-1,4-benzoquinone (DBQ) and 2,3-dicyano-1,4-naphthoquinone (DNQ) possess two fewer electron-withdrawing groups (i.e., Cl) compared to 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ), and they have more electron-withdrawing groups (i.e., CN) than 2,3-dichloro-1,4-benzoquinone (Cl₂Q) and 2,3-dichloro-1,4-naphthoquinone (Cl₂NQ) (i.e., Cl). Therefore, it is anticipated that the second reduction potentials of DBQ and DNQ are between those of Cl₂Q or Cl₂NQ and DDQ (Figure 2).

3. Materials and Methods

3.1. Synthesis of DNQ and DBQ

These two quinones were synthesized by following Klein’s procedure with minor modifications, as shown in Figure 3 [21]. To characterize the synthetic materials, we determined their melting points using a DigiMelt micromelting-point apparatus (Stanford Research Systems, Sunnyvale, CA, USA). ¹H NMR (300 MHz) and ¹³C NMR (75 MHz) spectra were recorded on a Bruker AV-300 spectrometer (Billerica, MA, USA) with tetramethylsilane (TMS) as an internal standard. The chemical shifts (δ) are given in parts per million (ppm), and the coupling constants (J) are in hertz (Hz). IR spectra were recorded using an Agilent Cary 630 FTIR Spectrometer (Santa Clara, CA, USA). Unless otherwise noted, all reagents, including F₄Q, were obtained from commercial suppliers and were used without further purification. Organic solvents used were dried by standard methods when necessary. All reactions were monitored by thin-layer chromatography (TLC) with GF254 silica gel plates.

2,3-Dicyano-1,4-naphthoquinone (DNQ) [21]: 2,3-Dicyanonaphthalene-1,4-diol (700 mg, 3.33 mmol) was suspended in acetic acid (10 mL) and heated to 100 °C. While heating, 30% nitric acid (1.5 mL) was added. Addition of ice to the cold solution and collection of the precipitate afforded the desired product (416 mg, 2.00 mmol, 60%) as a yellow solid.

2,3-Dicyanobenzoquinone (DBQ) [22,23]: 2,3-Dicyanohydroquinone (160 mg, 1.0 mmol) was suspended in acetic acid (2.5 mL) and heated to 100 °C. While heating, 30% nitric acid (0.38 mL) was added. The reaction mixture was cooled by the addition of ice and extracted with CH₂Cl₂. The combined organic layers were washed with brine, dried over Na₂SO₄, filtered, and concentrated under vacuum to afford the desired product (40 mg, 0.25 mmol, 25%) as a yellowish-brown solid.

3.2. Cyclic Voltammetry (CV) Measurements [20]

The experiments were performed according to our previously developed procedure in [20]. The details were described in Supplementary Information. In brief, solutions of 1 mM analytes were assayed in 0.1 M NBu₄PF₆ in acetonitrile at glassy carbon (GC) working electrodes, platinum counter electrodes with a single junction Ag reference electrode referenced externally vs. Fc/Fc⁺. For carbon capture assays, CV data were collected under argon (Ar), 5% CO₂ in Ar, 30% CO₂ in Ar, and 100% CO₂ with the sparging hose connected to each tank in succession, sparging between runs. The CV data for each quinone under various atmospheric conditions were imported to OriginPro 2025 (Origin Lab) for visualization and analysis and collected in triplicate.

3.3. Data Analysis

To estimate the CO₂ binding equilibrium constant (K) of each activated quinone, we adapted our previously developed procedure [20]. In brief, atmospheric pressure at the time of data collection was determined from the National Center for Atmospheric Research (NCAR) Foothills Lab Weather Archive [24]. When necessary, corrections from sea-level values were removed to obtain the true pressure (Boulder’s elevation is 1655 m; 5.35 in Hg was used as the standard subtraction). Next, the temperature of the system was measured, and the vapor pressures of the solutions made according to the procedure were subtracted from the total headspace pressure. The CO₂ molar concentrations were calculated by subtracting the vapor pressure of acetonitrile at this temperature from the atmospheric pressure, followed by applying the fraction/percentage of CO₂. For each quinone, the E_2,red, and E_2,ox were extracted from the CV graph under the specific atmosphere. The E_1/2 value was calculated as the average of E_2,red, and E_2,ox. The shift of E_1/2 (∆E_1/2) under each CO₂ atmosphere condition was calculated as its E_2,1/2 minus the E_2,1/2 under argon atmosphere. A plot of ∆E_1/2 vs. [CO₂] was then generated using OriginPro 2025 (Origin Lab) and subsequently fitting the curve to Equation (1)

∆E_1/2 = (RT/F)ln(K[CO₂]ⁿ + 1),

(1)

where R is the ideal gas constant 8.314 J·mol⁻¹·K⁻¹, T is the temperature in K, and F is the Faraday constant. The fitting derived the CO₂ binding equilibrium constant, K and the number of CO₂ molecules (n) that are captured by the activated sorbent. The CO₂ binding free energy (ΔG_bind) for each quinone was calculated using Equation (2)

ΔG_bind = −RTln(K).

(2)

4. Results and Discussion

4.1. Preparation of Three Designed Quinones

Among the three quinones, DNQ was synthesized by following Klein’s procedure as shown in Figure 3 [21]. In brief, commercially available 2,3-dicyanonaphthalene-1,4-diol was oxidized by nitric acid in acetic acid at 100 °C. After cooling the reaction to 0 °C, DNQ precipitated and was collected by filtration and dried under high vacuum overnight at room temperature. It was characterized by melting point, IR, ¹H, and ¹³C NMR analysis, which match the reported data in the literature. DBQ was synthesized in a similar fashion to DNQ, but it was extracted with dichloromethane after the reaction. Synthetic DBQ was also characterized, and the analytical data matches reported results. The third quinone in the series, F₄Q, was commercially available and used without any purification.

4.2. Cyclic Voltammetry Studies of Three Designed Quinones

With the quinones in hand, we analyzed their electrochemical properties and CO₂ capture ability using our previously developed procedure. For each quinone, the cyclic voltammetry study was first performed from 1.0 to −2.5 V under an argon atmosphere using Potentiastat (BioLogic SP150, Seyssinet-Pariset, France). The CV plots (Figure 4) under Ar showed that, among three novel quinones that we designed, DBQ has the desired second reduction potential (−0.934 V), which is more positive than that of oxygen (c.a., −1.20 V). The second reduction potential of F₄Q was more negative (i.e., −1.250 V), which could be attributed to the electron-donating effect through conjugation of fluorine on the benzoquinone scaffold. Compared to DBQ, DNQ has a more negative second reduction potential (−1.366 V), likely due to its extended conjugation that stabilizes the radical anion intermediate and therefore decreases the second reduction potential.

Next, we estimated the ability of each quinone to capture and release CO₂ by performing the CV experiments under atmospheres containing various concentrations of CO₂ (i.e., 5%, 30%, and 100% CO₂ in argon, Figure 4). As expected, the CV plots of all three quinones showed no changes in the first reduction peaks and significant shifts in the second reduction peaks in the presence of CO₂. The second reduction potentials (E_2,1/2) of each quinone under various concentrations of CO₂ were calculated. The potential shifted positively when CO₂ was present in the atmosphere. This suggested rapid and spontaneous binding of CO₂ after the second reduction in the quinones, and CO₂ was released upon the first oxidation, as described in Figure 2a. The changes in second reduction potential (ΔE_1/2) from that under argon atmosphere were then calculated and summarized in

Table 1. The second reduction potentials and their shifts ΔE_1/2 under an atmosphere with various CO₂ concentrations for the quinones F₄Q, DBQ, and DNQ ¹.

Sorbent	Temperature (K)	[CO2] (%)	[CO2] (M)	E2,red a (V)	E2,ox a (V)	E2,1/2 a (V)	ΔE1/2 a (V)
F4Q	296	0	0.000	−1.338	−1.162	−1.250	0.000
		5	0.011	−1.297	−1.088	−1.193	0.058
		30	0.070	−1.288	−1.041	−1.165	0.086
		100	0.233	−1.288	−1.005	−1.147	0.104
DBQ	295	0	0.000	−0.903	−0.966	−0.935	0.000
		5	0.012	−0.943	−0.737	−0.840	0.095
		30	0.072	−0.942	−0.580	−0.761	0.174
		100	0.239	−0.935	−0.601	−0.768	0.167
DNQ	294	0	0.000	−1.532	−1.200	−1.366	0.000
		5	0.012	−1.405	−0.904	−1.155	0.212
		30	0.074	−1.331	−0.893	−1.112	0.254
		100	0.245	−1.350	−0.840	−1.095	0.271

^{1 a}: E_2,red and E_2,ox were extracted from CV plots directly; E_2,1/2 is the average of these two; ΔE_1/2 is the shift in potential and calculated as ΔE_1/2 = E_2,1/2 − E_2,1/2(Ar).

. Their corresponding CO₂ molar concentrations were calculated based on CO₂ percentages in argon and the vapor pressure of the solvent, acetonitrile, at the experimental pressure and temperature (see details in Supplementary Information).

4.3. Calculation of the Reduction Potentials and CO₂ Binding Affinity of Three Quinones

For each quinone, the ΔE_1/2 values were plotted against CO₂ concentration ([CO₂]) in OriginPro (Figure 5). The data were fitted to Equation (1) to estimate the CO₂ binding equilibrium constant (K) and the number of CO₂ molecules each quinone binds to (n). The results showed that each of the three quinones binds to 1 molecule of CO_2, and their binding constants (Ks) vary quite significantly, from 2.90 × 10⁵ M⁻¹ for F₄Q to 9.76 × 10³ M⁻¹ for DBQ and 594 M⁻¹ for DNQ. Their CO₂ binding free energies (ΔG_bind) were also calculated using Equation (2). Both F₄Q and DBQ are CO₂ binders with ΔG_bind values of −7.40 and −5.39 kcal/mol, respectively, while DNQ is a weaker binder with a ΔG_bind of −3.74 kcal/mol. Using the K values, we estimate that DBQ and F₄Q can capture at least 70% and 98% of CO₂, respectively, at low concentrations (i.e., 420 ppm in the current atmosphere) at room temperature [18,25].

4.4. Conclusions

In summary, three structurally well-defined quinones (i.e., F₄Q, DNQ, and DBQ) were designed as sorbents for DAC of CO₂ using electro-swing chemistry. DNQ and DBQ were synthesized according to a previous literature procedure and characterized using standard procedures and assayed using CV. Among these three quinones, DBQ has a desired second reduction potential of −0.935 V, which is more positive than oxygen (−1.2 V), suggesting that it may be activated/reduced for carbon capture at approximately −1.0 V, thereby outside of the oxygen reduction window. In addition, DBQ, upon activation, is a good CO₂ binder with an estimated ΔG_bind of −5.39 kcal/mol, suggesting that DBQ could capture >70% of CO₂ directly from the current atmosphere at room temperature based on thermodynamics alone. However, kinetic measurements and device optimization will be required to test these materials in real DAC scenarios. Given that this electro-swing process requires relatively low electric energy and can be easily coupled with renewable energy, the discovery of novel quinone-based sorbents holds promise for real-world applications in carbon capture, contributing to efforts to combat global warming and its associated impacts.