Optimizing Direct Air Capture Solvents to Minimize Energy Consumption of CO2 Release in a Carbonate Electrolyzer

Addressing climate change by carbon management is critical to achieving the goal of net zero carbon emissions by 2050. In this work, we examined the electrochemically-driven recovery of CO2 during alkaline solvent regeneration for solvent-based direct air capture. A mathematical model was developed by incorporating carbonate chemistry with water electrolysis to predict the energy consumption per unit of CO2 released. The predicted results were consistent with the experimental data, in which the experimental work was achieved by characterizing alkalinity and carbon loading values of solvent collected from a flow carbonate electrolyzer. Through this study, we learned that minimizing the energy expended on CO2 release can be achieved by using an anolyte with a lower alkalinity, increasing the electric charge input to the electrolyzer, and reducing the ohmic resistance of the electrolyzer. Furthermore, using a supporting electrolyte, e.g., Na2SO4 in the present work, effectively compensates for the higher ohmic resistance from using an anolyte with a lower alkalinity.

; and subsequently the solutions are chemically and thermally reconditioned to release CO 2 gas while producing the fresh alkaline solutions.6][7] This implies the potential for rapid scalability within a relatively short timeframe to combat climate change and carbon neutral by 2050.
The operational concept of the DAC process explored in this study is based upon both the solvent-based approach and mature technology of alkaline water electrolysis.The process involves regenerating the solvent through pH swings generated by a carbonate electrolyzer, [8][9][10][11][12][13] providing an alternative to the solid sorbent and liquid solvent processes previously introduced.In contrast to those previous methods, our approach, especially for the regeneration step, potentially brings two significant advantages: (1) the production of H 2 as a value-added product through water electrolysis and (2) the elimination of heat-related steps for the regeneration process.
As illustrated in Fig. 1, consuming OH − ions toward O 2 evolution decreases the pH level at the anolyte loop, while producing OH − ions resulting from splitting water causes the pH level to increase at the catholyte loop.Consequently, the reduced pH environment facilitates the release of CO 2 gas from the dissolved carbon species such as CO 3 2− , whereas the elevated pH environment encourages the formation of alkaline hydroxides for carbon capture in an air absorber.While similar concepts involving water electrolysis, electrodialysis, or pH swing for regenerating carbon capture solvent using a flow electrochemical cell have been documented in the literature, [14][15][16][17][18][19][20][21][22][23][24][25] most of these works, to the best of our knowledge, have primarily focused on demonstrating proof of concept and exploring materials by disclosing the initial findings rather than providing intrinsic relationships between carbonate chemistry and water electrolysis.
Efforts in this study were made to investigate the energy consumption per mole of CO 2 released from the anolyte loop of a flow carbonate electrolyzer.Through both theoretical predictions and experimental verifications, valuable insights have been gained in optimizing the DAC solvent to minimize energy consumption by adjusting the alkalinity levels and utilizing supporting electrolytes.

Experimental
Process introduction.-Boththe anolyte and catholyte were composed of an approximately 2 M Na 2 SO 4 solution as the supporting electrolyte to reduce the ionic resistance for achieving a lower operating voltage of the carbon electrolyzer.To attain a desired alkalinity of anolyte, Na 2 CO 3 and NaHCO 3 solids were dissolved into the 2 M Na 2 SO 4 solution, in which the mass of each solid was pre-calculated to simulate the co-existence of CO 3 2− and z E-mail: xin.gao1@uky.eduECS Advances, 2024 3 024501 HCO 3 − in equilibrium with respect to the 0.4 mbar CO 2 gas in air by solving Eqs.1-6.All the chemicals were purchased from Sigma-Aldrich.Prior to initiating the charging process, to wet the cationexchange membrane placed in the electrolyzer, 1 L of each anolyte and catholyte were continuously circulated between the PVC electrolyte tanks and electrolyzer via PVC plastic tubing using Masterflex peristaltic pumps at 1 L min −1 .This circulation process lasted for about 24 h before the charging process was performed.During testing, the electrolyzer was continuously charged at 4 A (400 A m −2 ) by a Keysight 3634 A DC power supply under the typical laboratory conditions, e.g., 23-25 °C, during which 3 anolyte samples were collected from the anolyte tank at the electric charge to total alkalinity ratio of 50% and 90%, in addition to an anolyte sample collected before the charging process.Herein, the electric charge in mole was calculated in accordance with Faraday's law for water electrolysis, and the total alkalinity was measured through acid-base titration.The liquid samples were later characterized by measuring the carbon loading to calculate the amount of CO 2 release.Changes in the pH values associated with the anolyte and catholyte were monitored in real-time by Hanna digital pH probes coupled with Cole-Parmer transmitters.All the data, including voltage and pH values, were recorded by a Graphtec midi GL220 logger.The above-described testing process was employed to study the energy consumption per CO 2 released from the anolyte by varying the alkalinity at 3 distinct levels, i.e., 0.24 M, 0.51 M, and 0.76 M.
Electrolyzer.-Theconfiguration of the carbonate electrolyzer, which was designed and fabricated in-house for this study, is similar to a typical alkaline electrolyzer with a single electrode pair, as sketched in Fig. 1 and depicted in Fig. S1.Instead of using an anionexchange membrane or a ceramic-based separator, 26 the carbonate electrolyzer used an Aquivion cation-exchange membrane with the liquid-membrane contact area of about 100 cm 2 to isolate 2 identical Inconel 625 plates that served as the anode and cathode.The Inconel plate was partially cut by a digital waterjet machine, creating flow channels that guide the departure of the liquid-gas mixture from the electrolyzer.(Fig. S1b).Two stainless steel endplates were employed to securely hold the electrolyzer in place using evenly distributed bolts encircling the endplates.Auxiliary components, including O-rings, chemical-resistant gaskets, and Teflon bushings, were employed to facilitate the sealing of the electrolyzer during liquid circulation.All the raw materials were purchased from McMaster-Carr.
Carbon loading measurement.-Toquantify the value of CO 2 release, inorganic carbon loading of the anolyte sample was measured by an in-house fabricated apparatus.(Fig. S2) In this measurement, about 3 g of solvent sample was injected into the glass vessel holding about 50 mL of concentrated H 3 PO 4 liquid.CO 2 gas was released and then carried by 0.5 L min −1 of N 2 gas to a Horiba VIA-510 CO 2 gas analyzer for quantitative analysis.The carbon loading value in mol kg −1 was automatically computed by integrating the CO 2 release with time by a built-in computational system.To convert the value to mol l −1 , the density of each sample was determined using a manufacturer-calibrated 5 ml pipette and a ThermoFisher digital balance with four decimal places of precision.The standard deviation for repeated measurements in this study was less than 7%.
Total alkalinity measurement.-Thealkalinity measurement has been detailed in our previous work in Ref. 8. Briefly, the total alkalinity of the anolyte sample was measured by acid titration using a Metrohm Titrando 836 automatic titrator.During testing, about 1 g of the sample was continuously titrated with 0.09 M H 2 SO 4 solution until its pH was 2.5.The volume of the acid at the final equivalent point was used to calculate the total alkalinity of the sample.The standard deviation for repeating measurements was smaller than 1% in the present study.

Theoretical Approach
Model introduction.-Oncethe electric charge flows across the carbonate electrolyzer, water electrolysis occurs while moving Na + ions across the cation-exchange membrane from the anode to the cathode, as illustrated in Fig. 1.Simultaneously, the pH of the anolyte decreases due to consuming OH − ions toward O 2 gas evolution via 4OH − → O 2 ↑ + 2H 2 O + 4e − , whereas the pH of the catholyte increases because of creating OH − ions driven by H 2 gas evolution via 2H 2 O + 2e − → H 2 ↑ + 2OH − .Besides O 2 evolution, the anode also releases CO 2 gas during water electrolysis via acid-base equilibria, including CO 3 2− + H 2 O ↔ HCO 3 − + OH − followed by HCO 3 − ↔ CO 2 ↑ + OH − .The reduced pH environment at the anode counteracts the presence of OH − , thereby promoting the forward shift in the equilibrium, leading to the release of CO 2 .Please also note that the pK value associated with SO 4 2− ↔ HSO 4 − (∼1.8) is much lower than that for HCO 3 − ↔ H 2 CO 3 containing dissolved CO 2 (∼6.3); 18 and therefore, the impact of [SO 4 2− ], due to the addition of Na 2 SO 4 as the supporting electrolyte, on the carbonate chemistry is likely negligible in the following equation derivations.Herein, the CO 2 release can be elucidated mathematically by incorporating the alkalinity-pH relationship in conjunction with the relevant equilibrium constants.The total alkalinity of anolyte, C Alk , is which is the ionic charge difference between the total negative charges and [H + ].Furthermore, the concentration of the total dissolved inorganic carbon species, C T , is [ * ] and OH [ ] − can be represented in a similar format with the equilibrium constants, [H + ], and C T , e.g., A schematic of the testing setup.This setup comprises a flow carbonate electrolyzer and two electrolyte tanks in addition to auxiliaries, including online pH probes coupled with transmitters, pumps, power supply, and datalogger.During testing, 1 L of each electrolyte was circulated at 1 L min −1 and was charged at 4 A. The operation of DAC process using this electrolyzer in conjunction with a CO 2 absorber is detailed in the supplementary information, alongside the setup images provided in Fig. S1.The carbon lean solvent produced in the catholyte loop is used to capture CO 2 from air in an air contactor while the carbon rich solvent after carbon capture is reproduced in the anolyte loop.
ECS Advances, 2024 3 024501 where K 1 , K 2 , and K w are the equilibrium constants for the equilibrium of HCO 3 In the present work, SO 4 2− is chemically and electrochemically inert; therefore, [SO 4 2− ] in Eq. 7 is unchanged.The reduction in C Alk is caused by moving Na + from the anolyte by passing the electric charge, q, which can be expressed by where eff α means the electric charge utilization efficiency for moving Na + ions from the anolyte, and the remaining part, (1eff α ), is the efficiency for moving H + ions from the anolyte.Herein, eff α = 1 was used throughout the calculations, assuming that no H + ions were transferred across the membrane.Moreover, q = i t, × in which i represents the current applied to charge the electrolyzer, and t denotes the charging period.
By holding C T constant, Eq. 7 is solved using the parameters from Table I to plot pH as a function of electric charge normalized to total alkalinity, q/(VFC Alk ), where V represents the volume of the anolyte, and F denotes the Faraday constant.The plot shown as the red dashed line in Fig. 2a represents the anolyte's pH reduction during transferring Na + ions from the anode to the cathode, where the two-stage pH decline involves converting Moreover, by defining the thermoneutral voltage, U tn , of 1.48 V for water electrolysis, 28 the voltage arising from the ohmic resistance of electrolyte, membrane, and gas bubbles, U ohmic , and the overpotential spent on gas evolution reactions, U reaction , the total voltage of the electrolyzer, U total , can be expressed as  Please note CO 2 * represents the dissolved CO 2 before the transition to CO 2 gas depending on the partial pressure of CO 2 in the gas atmosphere.Both the plots are plotted against the ratio of electric charge to total alkalinity, q/(VFC Alk ), where V and F are the anolyte volume and Faraday constant, respectively.The plots are constructed using the parameters listed in Table I to solve Eqs.3-8.
ECS Advances, 2024 3 024501 To streamline the results comparisons from the different cases, q is usually normalized to C Alk throughout the text unless specified otherwise.
Quantifying the amount of CO 2 release is achieved theoretically by studying the variance of [CO 2 * ] before and after water electrolysis.During the solvent regeneration, the electric charge attracts Na + ions to make the anolyte C Alk decrease, thereafter, altering [CO , * ] will increase to exceed its solubility limit, eventually becoming CO 2 gas escaping from the anolyte.Such a description is visualized in Fig. 2b by solving Eqs. 3 to 8 with the parameters depicted in Table I.The ability to release CO 2 from the solvent, [CO 2 ] theory , depends upon the partial pressure of CO 2 in the gas phase, P , CO2 which relates to Henry's law, as follows: where K h is the Henry's constant with a typical value of 0.034 M bar −1 .Herein, by setting P CO2 = 0.4 mbar in the presence of air, Eq. 12 estimates that, for any CO 2 [ * ] exceeding 1.36 × 10 -5 M, computed as 0.034 M bar −1 × 0.0004 bar, the difference will be considered as the value of CO 2 theory [ ] in the present study.It is also important to note that, for a practical process design, creating a vacuum/degassing environment in the anode tank can improve gas removal, making the term K P h C O2 × negligible, followed by a cryogenic distillation separation unit to concentrate CO 2 .Nevertheless, the additional energy consumption associated with the concentration of CO 2 via cryogenic separation while leveraging H 2 sales and incentives for cost offsets falls outside the scope of this study.Lastly, the energy consumption of the electrolyzer from Eq. 11 divided by the CO 2 release from Eq. 12, gives the energy consumption per mole of CO 2 released using the carbonate electrolyzer.
Predicted result.-Theuse of a theoretical approach predicts E CO2_theory at various levels of anolyte alkalinity, electric charge passed across the carbonate electrolyzer, and overvoltage (or overpotential) loaded to the carbonate electrolyzer.The plots in Fig. 3a, derived from considering (U pH + U tn ) only, show that as q/(VFC Alk ) increases, each E CO2_theory curve exhibits a consistent downward trend in the energy expended for CO 2 release.For instance, taking the blue solid line representing the initial C Alk of 0.1 M, E CO2_theory decreases by about 23% when moving q/(VFC Alk ) from 0.5 to 0.9.Moreover, the plots indicate that opting for an anolyte with lower alkalinity can substantially reduce E .CO2_theory (This conclusion is further supported by examining the plot of E CO2_theory across a broader range of alkalinity, ranging from 0.01 M to 2.5 M, in Fig. S3a).For example, when q/(VFC Alk ) = 0.5, the blue solid line corresponding to C Alk = 0.1 M demonstrates nearly a 2.3-fold reduction in E CO2_theory compared to C Alk = 0.9 M, which is depicted by the black dash-dotted line.The plots in Fig. 3a are further developed by varying (U ohmic + U reaction ) to 0.75 and 1.5 V, respectively.A similar trend of E CO2_theory but with larger magnitudes are presented in Figs.3b and 3c.The summary of E CO2_theory at the different (U ohmic + U reaction ) is  4d, stating that increases in (U ohmic + U reaction ) can cause significant increases in the energy consumption for CO 2 release.For instance, as shown in the blue solid line denoting the starting C Alk of 0.1 M, increasing (U ohmic + U reaction ) to 1.5 V from 0 V, results in an almost 100% of the increase in the energy required to release 1 mole of CO 2 when q/(VFC Alk ) = 0.9.Moreover, lowering the charge input, q/(VFC Alk ), from 0.9 to 0.5 makes E CO2_theory more sensitive to (U ohmic + U reaction ).From this predictive study, such comparisons reveal that minimizing energy consumption for CO 2 release is possibly achieved by selecting a solvent with lower alkalinity, increasing charge input, and reducing the overpotential contributions for the gas evolution reactions and ohmic resistance of electrolyte, membrane, and gas bubbles.
The plots associated with Fig. 3d also evoke the use of a supporting electrolyte to mitigate the U ohmic variance.When testing anolytes with varying alkalinity levels and without the use of a supporting electrolyte, the ohmic resistance of the electrolyzer is significantly influenced by the fluctuations in ionic concentration associated with the alkalinity level, i.e., the higher alkalinity corresponds to the lower ohmic resistance of the electrolyzer.As a consequence, U ohmic will become the dominating variable to calculate E CO2_theory according to Eqs. 9 and 10, as U reaction can be independent of the alkalinity levels via sufficiently electrocatalyzing the gas evolution reactions.The dependency of U ohmic on alkalinity makes the comparison of E CO2_theory across a variety of alkalinity levels difficult because deceasing alkalinity increases U ohmic (E CO2_theory ) as demonstrated in Fig. 3.As a result, to align the U ohmic contributions stemming from the anolytes with different levels of alkalinity, using a supporting electrolyze is proposed in the experimental work.

Results and Discussion
Voltage behavior.-Aconcentrated acid solution is often used in aqueous batteries, e.g., H 2 SO 4 in lead-acid batteries or vanadium redox flow batteries, 29 which serve as a supporting electrolyte to reduce the voltage arising from the ohmic resistance of the working electrolyte and membrane.Such an experimental technique has been adopted to counteract the U ohmic variances caused by the anolyte with different levels of alkalinity.As illustrated in Fig. 4, when a 2 M Na 2 SO 4 solution was employed as the supporting electrolyte, the U exp lines for all cases nearly overlapped with an average deviation of less than 0.05 V.This observation strongly indicates that introducing a significant influx of Na + ions from the supporting electrolyte, at a minimum of 5-fold higher concentration with respect to the alkalinity of anolyte in the present work, effectively mitigates the voltage increases caused by increased ohmic resistance due to lower alkalinity.Consequently, this approach provides a fair comparison for evaluating the impact of different alkalinity levels on the energy needed for CO 2 release since the total energy value, as described in Eqs. 10 to 13, can be adjusted to be approximately the same for each case.Furthermore, SO 4 2− ions are chemically and electrochemically inert under the testing conditions used in the present work, meaning that SO 4 2− ions will not intervene with the total alkalinity and carbon loading measurements.
In addition to U exp , both U tn and U pH are overlaid in Fig. 4, where U pH is calculated via Eq. 9 using the experimental pH values depicted in Figs.5a and 5c.The plots depict that U pH reaches the highest value of about 0.4 V at the end of the testing period, i.e., q/(VFC Alk ) = 90%.The plots also show that a significant portion of U exp , 1.3-1.6V within the range of 3.1-3.5V, is attributed to (U ohmic +U reaction ), the overpotential used to overcome the activation barrier of gas evolutions reactions and ohmic resistance of the electrolyzer, encompassing membrane and gas bubbles.Such an observation means that U pH merely constitutes a maximum of 13% of U exp allocated to CO 2 release, much less than (U ohmic + U reaction ) accounting for 40%-50% of U exp .Developing the methods for mitigating (U ohmic + U reaction ) will be one of the research areas while scaling the electrolysis-based DAC processes.
pH Response.-Asdepicted in Fig. 5, the trend of the experimental pH lines generally agrees with the predicted pH lines.For example, in Figs.5a and 5b, both pH values begin decreasing from 10.2-10.5, followed by a transition region at q/(VFC Alk ) of 0.2 to 0.4.However, it is also noted that the experimental lines show a less pronounced reduction in pH starting at q/(VFC Alk ) of 0.4 compared to the calculated lines.This difference may be accounted for by H + ions transport from the anode chamber during the operation of the electrolyzer.As electric charge passes through the electrolyzer, O 2 evolution at the anode continuously decreases the pH of the anolyte, meaning a continuous increase in the quantity of H + ions in the anolyte.Under such a scenario, H + ions, considering a finite rate for acid-base neutralization toward CO 2 evaluation at the anode, can likely migrate from the anode to the cathode alongside Na + ions, as both the physical and chemical properties of the Aquivion membrane are close to the Nafion products possessing no cation selectivity. 30urthermore, since H + ions have a smaller hydrated radius than Na + ions, 31 the likelihood of H + ions crossover increases.These explanations may be supported by examining the electric charge utilization efficiency, , eff α in Table II, which was calculated based upon where C Alk_0 represents the initial alkalinity of the anolyte, and C Alk_x denotes the alkalinity of the anolyte when the anolyte was collected at q/(VFC Alk ) of 50% and 90%.As shown in the table, some of the eff α values are apparently not 100%, suggesting that a portion of electric charge might be diverted to move the H + ions across the membrane rather than Na + ions.Subsequently, those H + ions neutralize the OH − ions produced at the cathode, resulting in lower pH values from the experiments than the calculated pH values, as depicted in Figs.5c and 5d.This conclusion is further supported by our calculation of the effect of eff α on the pH of both the anolyte and catholyte, as depicted in Fig. S4.Here, a reduction in eff α Figure 4. Plots of the experimental voltage, U exp , minimal voltage, (U pH + U tn ), and thermoneutral voltage, U tn , as a function of the ratio of q/(VFC Alk ).Herein, U exp is obtained by applying the current at 400 A m −2 , U pH is calculated using Eq. 9 based upon the pH difference between the anolyte and catholyte shown in Figs.5a and 5c, and U tn equals 1.48 V. Table II.Carbon loading and alkalinity values when the anolyte was collected at q/(VFC Alk ) of 0, 50%, and 90%.Each sample was tested three times to attain the average value.Both values are expressed in mole, with the initial anolyte volume of 1 liter, assuming no change in volume during the tests.The standard deviation of carbon loading is less than 7% of the average value, and the standard deviation of alkalinity is approximately 1% of the average value.eff α means the electric charge utilization for moving Na + ions from the anolyte, e.g., (0.241 -0.124)/(0.5× 0.241) × 100 = 97.3% at q / (VFC Alk ) = 50%.To calculate the ratio of CO 2 to O 2 , the amount of CO 2 release was the difference of carbon loading values corresponding to q/(VFC Alk ) = 0.5 or 0.9 with respect to q/(VFC Alk ) = 0; and the amount of O 2 evolution was calculated based upon the Faraday's law for water electrolysis, in which the charge efficiency of O 2 evolution was assumed as 100%.correlates with a diminished pH difference between the anolyte and catholyte.Another plausible interpretation is the low perm-selectivity of the membrane, resulting in the discrepancy between the measured and calculated pH values.Nevertheless, the discrepancy between the experimental and predicted pH results will be specifically studied in one of the subsequent works using the cationexchange membranes purchased from different producers.

Case
Energy consumption.-Theanolyte samples were collected from the anolyte tank at q/(VFC Alk ) of 0, 50%, and 90% to measure both the carbon loading and alkalinity.These results in Table II were used to calculate the experimental E CO2 alongside the voltage obtained in Fig. 4. The amount of CO 2 released from the experiment, m , CO2 is defined as the reduction in dissolved carbon species in reference to the initial carbon loading, C T_0 , i.e., where C T_x denotes the carbon loading of the anolyte when the anolyte was collected at q/(VFC Alk ) of 50% and 90%.For example, in the case of C Alk = 0.24 M, the total CO 2 release is 0.126 mole, computed as 0.158-0.032(where the volume is 1 L).To quantify the energy expended on CO 2 release in the experiment, E exp , Eq. 10 is modified to where U avg represents the voltage mean over the duration of electric charge passing through the electrolyzer.For instance, for the case of C Alk = 0.24 M shown in Fig. 4, U avg used in Eq.

= [ ]
The calculated data points from the experiments are plotted in Fig. 6 with overlaid predicted lines.All the predictions are consistent with the experimental data points with and without (U ohmic + U reaction ).For example, E CO _ exp 2 for C Alk = 0.24 M exhibits a lower value than the other energy values.Furthermore, E CO _ exp 2 associated with q/(VFC Alk ) = 0.9 has the lower value than that for q/(VFC Alk ) = 0.5.lastly, the data points for q/(VFC Alk ) = 0.5 are more scattered compared to those for q/(VFC Alk ) = 0.9.Such observations emphasize using an anolyte with a lower alkalinity, increasing electric charge passed, and reducing ohmic resistance to minimize the energy consumption of CO 2 release in a carbonate electrolyzer.Furthermore, the last column of Table II represents the ratio of CO 2 to O 2 , in which the amount of CO 2 was calculated via Eq.15, and the amount of O 2 , by assuming 100% charge efficiency, was determined by Faraday's law for water electrolysis.Reducing the alkalinity of the solvent is observed to increase the partial pressure of CO 2 in the gas mixture, which may aid in the subsequent separation of CO 2 from O 2 .
In this work, without consuming H 2 , we achieved an energy consumption as low as 500 kJ per mole of CO 2 released at 400 A m −2 .This energy value is higher than the similar works that utilize the pH swings to refresh the alkaline carbon capture solvents, as reported in Ref..14, 18 H 2 recovery can provide $1-10 kg −1 in sales to offset the DAC process cost.To further minimize the energy consumption, potential research explorations may include depolarizing the anode by utilizing and sacrificing the H 2 gas produced at the cathode and investigating methods to rapidly remove gas bubbles from the electrodes' chambers. 32,33Herein, we believe that, as per Ref. 18, depolarizing the anode reduces the operating voltage of the carbonate electrolyzer while elevating CO 2 purity by surpassing O 2 evolution.Furthermore, efforts will be devoted to studying the techno-economic assessments of CO 2 separation from O 2 through cryogenic distillation with H 2 sales for cost offset, and the resultant energy consumption of CO 2 release will be compared to the scenario where H 2 is employed for anodic depolarization.Lastly, the design of an air contactor should be advanced in optimizing mass transfer to compensate for the diminished capture rate resulting from the reduction in alkalinity.

Conclusions
The goal of this study is to reduce the energy consumption for CO 2 release in a carbonate electrolyzer for DAC applications.Initially, we developed a mathematical model based on the pHalkalinity-voltage relationship to simulate the energy consumption for CO 2 release at the anode as a function of the electronic charge passed through the carbonate electrolyzer.Subsequently, we carried out the constant-current charging tests on a flow electrolyzer to quantify the amount of released CO 2 by measuring the carbon loading value of the anolyte.Through both the theoretical and experimental studies, the key findings for achieving a reduced energy consumption for CO 2 release are: (1) decreasing the alkalinity of the anolyte, e.g., more than 10% of energy saving was achieved by reducing the alkalinity from 0.76 M to 0.24 M, (Based upon our experience, we recommend employing an alkalinity level ranging from 0.3 to 0.5 M for water electrolysis-based DAC.Within this alkalinity range, it is unlikely to significantly compromise the carbon capture efficiency, based on our findings.)and (2) increasing the electric charge input to the electrolyzer, e.g., increasing the electric charge input by almost double at an alkalinity of 0.24 M led to an energy reduction of approximately 40%.Furthermore, reducing both the voltage loss in water electrolysis and the ohmic resistance across the electrolyte, membrane, and gas bubbles is expected to yield additional energy savings for CO 2 release.These improvements can be achieved by employing catalytic electrodes to decrease the activation energy for gas evolution reactions and incorporating a supporting electrolyte to improve the ionic conductivity of a carbonate electrolyzer.Calculations of E CO2_ exp is described via Eqs.15 to 17.The predicted E CO2_theory lines are overlaid to assist the validation of using an anolyte with a lower alkalinity, increasing charge input, and reducing ohmic resistance to reduce the energy consumption of CO 2 release in a carbonate electrolyzer.

*U
CO 3 2− to HCO 3 − followed by HCO 3 − to CO .2At the cathode, the amount of OH − ions produced by splitting H 2 O is the same as Na + ions transported at ; eff α and therefore, the plot indicated by the black solid line in Fig.2ashows the catholyte's pH elevation, suggesting the production of NaOH solution.Under such a situation of pH variation between the anolyte, pH a , and catholyte, pH c , the pH-induced voltage, U pH , will vary in accordance with27

Figure 2 .
Figure 2. Predicted results for a carbonate electrolyzer during water electrolysis when the initial C Alk is set to 0.76 M. (a) pH of the anolyte and catholyte, and (b) changes in inorganic carbon species in the anolyte.Please note CO 2* represents the dissolved CO 2 before the transition to CO 2 gas depending on the partial pressure of CO 2 in the gas atmosphere.Both the plots are plotted against the ratio of electric charge to total alkalinity, q/(VFC Alk ), where V and F are the anolyte volume and Faraday constant, respectively.The plots are constructed using the parameters listed in TableIto solve Eqs.3-8.
in the anolyte.Under such a scenario, [CO 2

Figure 3 .
Figure 3. Theoretical energy consumption per of CO release, E .CO _theory (a) E CO _theory 2 was calculated using only the minimal voltage (U pH + U ), (b) and (c) E CO _theory 2 was calculated by considering (U ohmic +U reaction ) = 0.75 V and 1.5 respectively, and (d) increases in E CO _theory

Figure 5 .
Figure 5. Comparisons of experimental and predicted pH values of the anolyte and catholyte as a function of q/(VFC Alk ).(a) and (c) experimental pH for anolyte and catholyte, respectively, and (b) and (d) predicted pH for anolyte and catholyte, respectively.The tests were performed at 400 A m −2 .

Figure
Figure Experimental E CO2_ exp as a function of (U ohmic + U reaction ).Calculations of E CO2_ exp is described via Eqs.15 to 17.The predicted E CO2_theory lines are overlaid to assist the validation of using an anolyte with a lower alkalinity, increasing charge input, and reducing ohmic resistance to reduce the energy consumption of CO 2 release in a carbonate electrolyzer.

Table I .
Parameters used in the theoretical approach.
pH + U tn ) is known as the minimal or thermodynamic voltage required to release CO 2 gas in a carbonate electrolyzer.Following the voltage description, the theoretical energy consumption of CO 2 release, E theory , is calculated as eff α 1 -