Electrochemical Characterization of Electrodeposited Copper in Amine CO2 Capture Media

This study explores the stability of electrodeposited copper catalysts utilized in electrochemical CO2 reduction (ECR) across various amine media. The focus is on understanding the influence of different amine types, corrosion ramifications, and the efficacy of pulse ECR methodologies. Employing a suite of electrochemical techniques including potentiodynamic polarization, linear resistance polarization, cyclic voltammetry, and chronopotentiometry, the investigation reveals useful insights. The findings show that among the tested amines, CO2-rich monoethanolamine (MEA) exhibits the highest corrosion rate. However, in most cases, the rates remain within tolerable limits for ECR operations. Primary amines, notably monoethanolamine (MEA), show enhanced compatibility with ECR processes, attributable to their resistance against carbonate salt precipitation and sustained stability over extended durations. Conversely, tertiary amines such as methyldiethanolamine (MDEA) present challenges due to the formation of carbonate salts during ECR, impeding their effective utilization. This study highlights the effectiveness of pulse ECR strategies in stabilizing ECR. A noticeable shift in cathodic potential and reduced deposit formation on the catalyst surface through periodic oxidation underscores the efficacy of such strategies. These findings offer insights for optimizing ECR in amine media, thereby providing promising pathways for advancements in CO2 emission reduction technologies.


Introduction
Mitigating climate change necessitates urgent action to reduce CO 2 emissions.Achieving net zero CO 2 emissions by 2050 is critical to limit the temperature rise to below 1.5 • C, in line with the Paris Agreement (COP21) objectives [1].To address this challenge, the development of CO 2 capture technologies is a priority.Currently, the most advanced industrial technology employs aqueous amine solutions such as monoethanolamine (MEA), methyldiethanolamine (MDEA), and 2-amino-2-methylpropanol (AMP) for CO 2 capture.This process entails direct CO 2 capture from the flue gas stream and its subsequent release by heating the CO 2 -rich amine capture media in a desorber unit at approximately 120 • C [2].The regenerated amine then re-enters the capture cycle.However, the regeneration step, accounting for up to 30% of the total plant energy output, is highly energy-intensive [3].This issue may be overcome by regenerating the CO 2 -rich amine media using electrochemical CO 2 reduction (ECR) instead.This approach features the direct transformation of amine-CO 2 adducts, such as carbamates, into valuable chemicals [4].The seamless integration of amine-based CO 2 capture and ECR demonstrates good synergy.This approach not only bypasses challenges related to CO 2 transportation and storage but also eradicates the necessity for regenerating capture media through the thermal release of molecular CO 2 [5].
ECR is an emerging field, evolving independently from carbon capture technologies.It presents significant synergies with other climate change challenges, notably in storing

Specimen Preparation and Characterisation
Copper was electrodeposited onto a copper substrate using a method detailed in the Supplementary Materials.Following electrodeposition, specimens were selectively masked with resin provided by Belzona Ltd. (Hawarden, UK, reference 1395) resin, resulting in an exposed surface area of approximately 1 cm 2 for each specimen.To characterize the surfaces of these specimens, an optical microscope BX41M-LED from Olympus (Hachioji, Japan) and a Sigma 1455EP scanning electron microscope (SEM) from Ziess (Oberkochen, Germany) were employed.
The anolyte for these experiments was 0.5 M potassium chloride (KCl) solution.Additionally, 0.5 M KCl was incorporated into the amine solutions to enhance their electrical conductivity.All chemicals used in solution preparation were of laboratory grade and supplied by Merck Life Science UK Ltd. (Gillingham, UK).
In subsequent discussions and analyses, these solutions are referred to as MEA, MDEA, MDEA/PZ, and AMP for simplicity.

Electrochemical Tests
Electrochemical experiments were conducted using a Biologic (Grenoble, France) VMP-300 potentiostat and a 100 mL H-cell sourced from Dek research (Hong Kong), as represented in Figure 1.The experimental setup featured anodic and cathodic compartments, separated by a Nafion 117 proton exchange membrane (Dupont, Wilmington, DE, USA) which underwent appropriate activation before use.The reference electrode comprised an Ag/AgCl in 3.5 M KCl, positioned in a Luggin capillary filled with 3.5 M KCl to isolate it from the amine electrolyte.The capillary's tip was situated 5 mm from the specimen.A platinum mesh served as the counter electrode.The cell design included gas inlet/outlet provisions for purging the solution with CO 2 .Each specimen was kept at −0.5 V relative to the Open Circuit Potential (OCP) for 5 min before testing to remove the air-formed oxide layer.

Specimen Preparation and Characterisation
Copper was electrodeposited onto a copper substrate using a method detailed in the Supplementary Materials.Following electrodeposition, specimens were selectively masked with resin provided by Belzona Ltd. (Hawarden, UK, reference 1395) resin, resulting in an exposed surface area of approximately 1 cm 2 for each specimen.To characterize the surfaces of these specimens, an optical microscope BX41M-LED from Olympus (Hachioji, Japan) and a Sigma 1455EP scanning electron microscope (SEM) from Ziess (Oberkochen, Germany) were employed.
The anolyte for these experiments was 0.5 M potassium chloride (KCl) solution.Additionally, 0.5 M KCl was incorporated into the amine solutions to enhance their electrical conductivity.All chemicals used in solution preparation were of laboratory grade and supplied by Merck Life Science UK Ltd. (Gillingham, UK).
In subsequent discussions and analyses, these solutions are referred to as MEA, MDEA, MDEA/PZ, and AMP for simplicity.

Electrochemical Tests
Electrochemical experiments were conducted using a Biologic (Grenoble, France) VMP-300 potentiostat and a 100 mL H-cell sourced from Dek research (Hong Kong), as represented in Figure 1.The experimental setup featured anodic and cathodic compartments, separated by a Nafion 117 proton exchange membrane (Dupont, Wilmington, DE, USA) which underwent appropriate activation before use.The reference electrode comprised an Ag/AgCl in 3.5 M KCl, positioned in a Luggin capillary filled with 3.5 M KCl to isolate it from the amine electrolyte.The capillary's tip was situated 5 mm from the specimen.A platinum mesh served as the counter electrode.The cell design included gas inlet/outlet provisions for purging the solution with CO2.Each specimen was kept at −0.5 V relative to the Open Circuit Potential (OCP) for 5 min before testing to remove the airformed oxide layer.Linear polarization resistance (LPR), potentiodynamic polarization (PDP), and cyclic voltammetry (CV) tests were performed after recording and stabilizing the OCP for 1 h.To ensure reproducibility, each measurement was conducted in triplicate.LPR results are presented in Supplementary Materials Figure S6.The test solutions were maintained at ambient laboratory conditions (20 ± 2 • C) and could be CO 2 -purged as required.LPR tests utilized a potential offset of ±20 mV relative to OCP at a scan rate of 0.125 mV s −1 .PDP involved potential sweeping from −0.2 V to +0.2 V versus OCP at a rate of 10 mV min −1 .CV scans were conducted at 20 mV s −1 scan rate, reversing the scan when the current density reached |10| mA cm −2 for both cathodic and anodic polarization.
Chronopotentiometry experiments were designed to simulate 60 h of ECR.Three strategies, which included alternating anodic and cathodic segments, were implemented as detailed in Figure 2. Anodic segments were conducted at +1 mA cm −2 , while cathodic segments operated at −10 mA cm −2 , cumulatively totaling 60 h for cathodic segments.Pulse mode 1 consisted of a 24 s anodic segment every 45 min, and pulse mode 2 involved a 2 s anodic segment every 50 s.Linear polarization resistance (LPR), potentiodynamic polarization (PDP), and cyclic voltammetry (CV) tests were performed after recording and stabilizing the OCP for 1 h.To ensure reproducibility, each measurement was conducted in triplicate.LPR results are presented in Supplementary Materials Figure S6.The test solutions were maintained at ambient laboratory conditions (20 ± 2 °C) and could be CO2-purged as required.LPR tests utilized a potential offset of ±20 mV relative to OCP at a scan rate of 0.125 mV s −1 .PDP involved potential sweeping from −0.2 V to +0.2 V versus OCP at a rate of 10 mV min −1 .CV scans were conducted at 20 mV s −1 scan rate, reversing the scan when the current density reached |10| mA cm −2 for both cathodic and anodic polarization.
Chronopotentiometry experiments were designed to simulate 60 h of ECR.Three strategies, which included alternating anodic and cathodic segments, were implemented as detailed in Figure 2. Anodic segments were conducted at +1 mA cm −2 , while cathodic segments operated at −10 mA cm −2 , cumulatively totaling 60 h for cathodic segments.Pulse mode 1 consisted of a 24 s anodic segment every 45 min, and pulse mode 2 involved a 2 s anodic segment every 50 s.Electrochemical tests underwent iR compensation utilizing the potentiostat's built-in feature, which is based on high-frequency impedance measurements.The average cell resistances determined for each media are systematically tabulated in Supplementary Materials Table S1.

Electrochemical Characterisation of Electrodeposited Copper in Amine Media
Potentiodynamic polarizations were performed in triplicate under both CO2-free and CO2-purged conditions across various amine media and are shown in Figure 3.The Tafel extrapolation method was employed to estimate the corrosion current density (jcorr) [37]; the full set of Tafel extrapolation is provided in Supplementary Materials, Figures S2-S5.The average jcorr values, together with the corresponding corrosion potentials (Ecorr), are compiled in Table 1.Electrochemical tests underwent iR compensation utilizing the potentiostat's built-in feature, which is based on high-frequency impedance measurements.The average cell resistances determined for each media are systematically tabulated in Supplementary Materials Table S1.

Electrochemical Characterisation of Electrodeposited Copper in Amine Media
Potentiodynamic polarizations were performed in triplicate under both CO 2 -free and CO 2 -purged conditions across various amine media and are shown in Figure 3.The Tafel extrapolation method was employed to estimate the corrosion current density (j corr ) [37]; the full set of Tafel extrapolation is provided in Supplementary Materials, Figures S2-S5.The average j corr values, together with the corresponding corrosion potentials (E corr ), are compiled in Table 1.In environments devoid of CO2, jcorr values for all amines were lower compared to 0.5 M KCl.Specifically, MDEA and MDEA/PZ exhibited reduced jcorr values relative to MEA and AMP, with the latter two showing jcorr values comparable to KCl.Upon introducing CO2, a notable shift in Ecorr towards more anodic potentials occurred for all amines, attributable to the dissolution of CO2 and the formation of carbonate/bicarbonate species.Concurrently, the introduction of CO2 resulted in an increase in jcorr across all amines.The relative order of jcorr values remained consistent with the CO2-free cases: MEA and AMP exhibited higher jcorr, exceeding that of KCl, while MDEA and MDEA/PZ maintained significantly lower values.Polarization resistance (Rp) data, derived from LPR measurements, were consistent with jcorr.Lower Rp values, indicative of higher corrosion rates, Table 1.Average corrosion potential (E corr ), corrosion current density (j corr ), and polarization resistance (R p ) for electrodeposited copper specimens in different amine media.In environments devoid of CO 2 , j corr values for all amines were lower compared to 0.5 M KCl.Specifically, MDEA and MDEA/PZ exhibited reduced j corr values relative to MEA and AMP, with the latter two showing j corr values comparable to KCl.Upon introducing CO 2 , a notable shift in E corr towards more anodic potentials occurred for all amines, attributable to the dissolution of CO 2 and the formation of carbonate/bicarbonate species.Concurrently, the introduction of CO 2 resulted in an increase in j corr across all amines.The relative order of j corr values remained consistent with the CO 2 -free cases: MEA and AMP exhibited higher j corr , exceeding that of KCl, while MDEA and MDEA/PZ maintained significantly lower values.Polarization resistance (R p ) data, derived from LPR measurements, were consistent with j corr .Lower R p values, indicative of higher corrosion rates, were observed for CO 2 -loaded MEA and AMP, whereas CO 2 -free MDEA and MDEA/PZ demonstrated higher resistances.

Media
CV scans, presented in Figure 4, reveal distinct electrochemical behaviors of copper in various amine media.For comparative purposes, CV experiments were also conducted in 0.5 M KCl.In CO 2 -lean amine media, two oxidation peaks (A and B) and two reduction peaks (C and D) were identified.Peak onset potentials are tabulated in Table 2.It is well established that copper oxidizes to form either Cu(I) or Cu(II) [38].During the anodic sweep, the initial peak (A) is attributed to the formation of Cu(I), while the subsequent peak (B) corresponds to Cu(II) oxidation, both manifesting as copper oxides/hydroxides and their hydrated forms.The oxidation current increase beyond peak B signifies the formation of soluble copper ions such as CuO 2 − until the onset of oxygen evolution at more anodic potentials.The onset potential for peak A remains relatively consistent across all CO 2 -free amine media, approximately at 0.45 V.However, the onset for peak B varies significantly, indicating a more complex oxidation mechanism, which is in line with previous findings in KOH solutions [39].In MEA, the lowest onset potential leads to the overlapping of peaks A and B. In contrast, MDEA, MDEA/PZ, and AMP exhibit distinct B peaks at more anodic potentials, at −0.27 V, −0.15 V, and −0.27 V, respectively.Peaks C and D are associated with the reduction in Cu(I) and Cu(II) species, while the highest cathodic current surge is attributed to the hydrogen evolution reaction (HER).In CO 2 -rich amine media, only two peaks, labeled A CO2 and B CO2 , are evident.Peak A CO2 could represent the overlap of peaks A and B, indicating the concurrent formation of both Cu(I) and Cu(II) species, or it might signify the exclusive formation of Cu(II) species.The former scenario seems more probable, as the overlapping of peaks A and B has been previously reported in high pH conditions [39], likely influenced by CO 2 purging.The onset oxidation potentials in various amines are closer in the CO 2 -rich environment, recorded at −0.21 V, −0.27 V, −0.17 V, and −0.25 V for MEA, MDEA, MDEA-PZ, and AMP, respectively.This observation underscores the significant impact of the presence of CO 2 on copper oxidation and supports previous PDP and LPR data.Copper oxidizes more readily in CO 2 -rich media, as evidenced by a steeper increase in oxidation current at more cathodic potentials, except for MEA, which exhibits a slightly higher oxidation onset potential in CO 2 -rich conditions.In CO 2 -rich amines, the most anodic current is either due to HER or the CO 2 reduction reaction (CO 2 RR) and is shifted towards more anodic potentials.This shift suggests that CO 2 RR may have a more anodic onset potential than HER, leading to an observable increase in current density on the CV scan for CO 2 -purged media.In CO2-rich amine media, only two peaks, labeled ACO2 and BCO2, are evident.Peak ACO2 could represent the overlap of peaks A and B, indicating the concurrent formation of both Cu(I) and Cu(II) species, or it might signify the exclusive formation of Cu(II) species.The former scenario seems more probable, as the overlapping of peaks A and B has been previously reported in high pH conditions [39], likely influenced by CO2 purging.The onset oxidation potentials in various amines are closer in the CO2-rich environment, recorded at −0.21 V, −0.27 V, −0.17 V, and −0.25 V for MEA, MDEA, MDEA-PZ, and AMP, respectively.This observation underscores the significant impact of the presence of CO2 on copper oxidation and supports previous PDP and LPR data.Copper oxidizes more readily in CO2-rich media, as evidenced by a steeper increase in oxidation current at more cathodic potentials, except for MEA, which exhibits a slightly higher oxidation onset potential in CO2-rich conditions.In CO2-rich amines, the most anodic current is either due to HER or the CO2 reduction reaction (CO2RR) and is shifted towards more anodic potentials.This shift suggests that CO2RR may have a more anodic onset potential than HER, leading to an observable increase in current density on the CV scan for CO2-purged media.

Chronopotentiometry
Specimens underwent cathodic polarization for a cumulative duration of 60 h, employing varied pulse strategies.These strategies encompassed continuous cathodic polarization, pulse mode 1, and pulse mode 2, as detailed in Figure 2. The potential as a function of time was closely monitored, with the corresponding cathodic chronopotentiometry results displayed in Figure 5. Anodic chronopotentiometry results are similarly detailed in Figure 6.
polarization, pulse mode 1, and pulse mode 2, as detailed in Figure 2. The potential as a function of time was closely monitored, with the corresponding cathodic chronopotentiometry results displayed in Figure 5. Anodic chronopotentiometry results are similarly detailed in Figure 6.polarization, pulse mode 1, and pulse mode 2, as detailed in Figure 2. The potential as a function of time was closely monitored, with the corresponding cathodic chronopotentiometry results displayed in Figure 5. Anodic chronopotentiometry results are similarly detailed in Figure 6.Chronopotentiometry results for continuous cathodic polarization over 60 h in MEA, MDEA, and AMP are illustrated in Figure 5a.In MDEA/PZ, this experiment was hindered by the formation of carbonate salts, which obstructed the CO 2 gas inlet after only 2-3 h.Similar carbonate salt precipitation was observed in MDEA, leading to an unstable cathodic potential that progressively shifted towards more cathodic values, as depicted in Figure 5a.In contrast, the potentials in AMP demonstrated greater stability, albeit with a slight decrease over time.Notably, MEA exhibited the most stable potential, concluding the experiment at a potential 20 mV higher than the initial value.
The influence of pulse strategies on potential evolution for the various amine solutions is highlighted in Figure 5b-d.Employing a lower frequency pulse strategy (pulse mode 1) consistently improved potential stability across all amine solutions.In the case of MDEA, this strategy resulted in a stable potential decrease beyond 55 h, contrasting with the trend observed under continuous polarization.For MEA and AMP, pulse mode 1 induced a gradual electropositive shift in potential, reaching increments of +70 mV and +65 mV from their initial values, respectively.The adoption of a higher frequency of anodic segments (pulse mode 2) demonstrated diverse outcomes.In MEA, pulse mode 2 more significantly improved potential stability compared to pulse mode 1, achieving an increase of +125 mV compared to the initial value.In contrast, for AMP, the potential under pulse mode 2 initially remained stable but began to decline after 45 h, aligning more closely with the behavior observed under continuous polarization.For MDEA, the potential during pulse mode 2 rapidly deteriorated and became extremely unstable upon the onset of carbonate salt precipitation.
Anodic chronopotentiometry results for the various pulse strategies are shown in Figure 6.Three distinct behaviors emerged: stable, erratic, and decreasing.A stable potential was noted in all specimens subjected to pulse mode 1, as well as in MEA under pulse mode 2. The potential required to reach a current density of +1 mA cm −2 hovered around −0.15 V for pulse mode 1 and −0.2 V for pulse mode 2 in MEA.This variance is attributed to surface charging/discharging phenomena, necessitating more anodic potentials to offset the declining discharging current with an elevated Faradaic oxidation current for longer anodic durations.Since pulse mode 2 features shorter durations, the proportion of Faradaic current is reduced, hence less anodic potentials are sufficient to sustain the target current density.
In the case of MDEA and AMP under pulse mode 2, the anodic potential exhibited instability, albeit with markedly different behaviors.In MDEA using pulse mode 2, an erratic potential was observed beyond 15 h, likely caused by the intermittent formation and detachment of carbonate salts on the catalyst surface.In contrast, AMP during pulse mode 2 exhibited a continuous cathodic shift in anodic potential over time.
The post-chronopotentiometry visual appearance of copper catalysts exhibits marked variations contingent on the polarization mode, as illustrated in Figure 6.Continuously polarized specimens display darker surfaces, whereas those subjected to pulse modes generally manifest 'cleaner' copper surfaces, with the exception of AMP under pulse mode 2. When correlating the visual observations of the catalysts after chronopotentiometry with the anodic potential data from Figure 6, it becomes evident that periodic oxidation facilitated by pulse modes effectively mitigates the formation of dark deposits.The pronounced dark deposit observed in AMP under pulse mode 2 indicates inadequate oxidation of the copper catalyst.Indeed, several factors can contribute to a cathodic shift in potential when maintaining a target anodic current density, such as a reduction in active surface area, a shift in oxidation onset potential, or an increased charging of the interface reducing the proportion of Faradaic currents required to achieve the target.The observed potential diminution for AMP pulse mode 2 is presumably attributed to the latter, as post-test surface characterization revealed a dark deposit akin to that observed following continuous polarization, as shown in Figure 7.This suggests that the intended oxidation of copper was not effectively achieved, pointing to surface charging as the underlying cause for the diminished oxidation.Under these specific test conditions, an anodic duration of 2 s proved insufficient for achieving copper oxidation in CO 2 -rich AMP.These findings underscore the importance of cyclic copper oxidation in preventing the formation of such dark deposits.
ing continuous polarization, as shown in Figure 7.This suggests that the intended oxidation of copper was not effectively achieved, pointing to surface charging as the underlying cause for the diminished oxidation.Under these specific test conditions, an anodic duration of 2 s proved insufficient for achieving copper oxidation in CO2-rich AMP.These findings underscore the importance of cyclic copper oxidation in preventing the formation of such dark deposits.Artefacts resulting from a gas generation at the catalyst surface, characterized by black spots with occasional dark trails directed upward, are discernible in specimens under continuous and pulse mode 1 polarization.However, these artefacts are absent in samples subjected to pulse mode 2, where no such phenomena are observed.
Higher magnification images of the catalyst surface are provided in Figure 8. Prior to chronopotentiometry, as shown in Figure 8a,b, the surface primarily consists of pure copper, characterized by an uneven topology with sporadic overgrown protrusions, ranging from 15 µm to 35 µm in diameter.After continuous polarization, the formation of a dark deposit is apparent, leaving only a thin network of exposed copper (Figure 8c,d).Conversely, in MEA under pulse mode 2 (Figure 8e), the surface displays no discernible dark deposits, and the integrity of the surface is notably preserved.An instance of the gas generation artefact, observable in MEA under pulse mode 1 (Figure 8f), exhibits a dark deposit in the shape of a ring, approximately 100 µm in diameter, with a center of exposed copper corresponding to the cathodic site.Artefacts resulting from a gas generation at the catalyst surface, characterized by black spots with occasional dark trails directed upward, are discernible in specimens under continuous and pulse mode 1 polarization.However, these artefacts are absent in samples subjected to pulse mode 2, where no such phenomena are observed.
Higher magnification images of the catalyst surface are provided in Figure 8. Prior to chronopotentiometry, as shown in Figure 8a,b, the surface primarily consists of pure copper, characterized by an uneven topology with sporadic overgrown protrusions, ranging from 15 µm to 35 µm in diameter.After continuous polarization, the formation of a dark deposit is apparent, leaving only a thin network of exposed copper (Figure 8c,d).Conversely, in MEA under pulse mode 2 (Figure 8e), the surface displays no discernible dark deposits, and the integrity of the surface is notably preserved.An instance of the gas generation artefact, observable in MEA under pulse mode 1 (Figure 8f), exhibits a dark deposit in the shape of a ring, approximately 100 µm in diameter, with a center of exposed copper corresponding to the cathodic site.

Corrosion of Copper Catalyst in Amine Media
Corrosion rates were calculated from the corrosion current density (j corr ) in accordance with ASTM G102 guidelines [40].The derived corrosion rates, along with the time necessary to dissolve a 50 µm copper deposit (the estimated thickness based on weight measurements), are tabulated in Table 3 These calculations reveal that the complete dissolution of the copper deposit, under the presumption of exclusive Cu(I) formation and in the absence of any passivation, would necessitate 56 days in the most severe scenario (CO 2 -enriched MEA).In the case of Cu(II) formation, the time extends to 112 days.It is critical to recognize that the hypothesis of no passivation is valid only immediately after the termination of the cathodic current, as an oxide protective layer is expected to form subsequently.Thus, these calculations are pertinent mainly for approximating the cumulative switch-off duration tolerable by the catalyst.Accordingly, a cumulative switch-off time threshold of 56 days is considered more than adequate.Therefore, the free corrosion of copper in the amine media does not constitute a limiting factor for the catalyst's operational lifespan.The influence of corrosion-induced surface alterations on the catalyst's efficiency, including aspects such as selectivity, FE, and EE, still requires further investigation.However, existing literature, including findings by [41], indicates that copper oxides favor the reduction in CO 2 into C 2 + products, suggesting that corrosion may not pose immediate concerns for catalyst functionality.

Choice of Amine Capture Media for ECR
CO 2 capture in amine media is primarily governed by two processes.The initial process involves the formation of carbamate, as described by reaction (1).Subsequently, the formed carbamate undergoes hydrolysis into bicarbonate, delineated in reaction (2).The prevalence of these reactions is contingent on the amine type.Non-sterically hindered primary amines, such as MEA, tend to form stable carbamate; thus, reaction (1) predominates, leading to almost complete CO 2 capture in the form of carbamate [42].While this mechanism offers rapid capture kinetics, it necessitates higher energy for CO 2 release.
In sterically hindered amines like AMP, the carbamate bond is less stable [36,43], enhancing the carbamate hydrolysis reaction (reaction 2).This results in CO 2 being stored as both carbamate and bicarbonate, combining the fast capture kinetics of reaction (1) with an increased CO 2 absorption capacity due to the 1:1 stoichiometry between CO 2 and the amine molecule in reaction (2), in contrast to the 2:1 ratio in reaction (1).CO Tertiary amines, such as MDEA, do not form carbamate with CO 2 .Instead, they function as catalysts for CO 2 hydration [44], leading to exclusive bicarbonate capture as per reaction (3).This mechanism offers a higher absorption capacity, albeit at a slower capture rate.CO Historically, capturing CO 2 as bicarbonate has been favorable due to its greater absorption capacity and reduced thermal energy requirement for CO 2 release.The findings presented here indicate that tertiary amine can lead to significant carbonate salt precipitation during ECR, as observed in MDEA and MDEA/PZ.In contrast, primary amines do not facilitate carbonate salt formation, as evidenced in MEA and AMP.Based on these observations, primary amines are emerging as more promising candidates for direct ECR applications.This is confirmed by chronopotentiometry results which showed greater stability for MEA.

Pulse ECR in Amine Media
Pulse chronopotentiometry findings clearly demonstrate that incorporating short anodic segments counteracts the cathodic shift in potential over time, often resulting in an anodic shift in cathodic potentials.This trend suggests an enhancement in the efficiency of reduction reactions, as evidenced by lower overpotentials needed to maintain the target current density.The formation of copper oxides during anodic segments is likely responsible for this anodic potential shift.
Beyond the advantageous potential shift, pulse ECR exhibits a pronounced ability to preserve the catalyst surface, effectively minimizing the formation of dark deposits.These outcomes highlight the necessity for optimization in pulse ECR to enhance results, such as maximizing potential shifts and reducing deposit formation, as exemplified by pulse mode 2 in MEA.The optimization process varies with the media used, as evidenced by the differences in outcomes for MEA and AMP under pulse mode 2. Key considerations for effective pulse ECR include ensuring adequate oxidation of the copper surface by setting a sufficiently long anodic step duration, counter to the approach seen in AMP under pulse mode 2. Additionally, the cathodic step duration should be short enough to prevent excessive deposit formation at gas generation sites, such as observed in AMP under pulse mode 1, which could be problematic to eliminate during subsequent anodic steps.
Carbonate salt formation arises from localized HO − ion enrichment near the catalyst surface, a byproduct of the CO 2 reduction reaction (CO 2 RR) [20].Previous studies have suggested that pulse ECR can mitigate carbonate salt formation by curbing HO − accumulation [21,29].However, the pulse strategies evaluated were ineffective in preventing carbonate salt precipitation in MDEA-based capture media.This outcome implies that tertiary amines favor carbonate salt formation.

Conclusions
The investigation into the stability of electrodeposited copper catalysts for Electrochemical Reduction (ECR) in various amine media offers valuable insights into critical aspects of this process.The study underscores several key findings that advance our understanding and pave the way for future research endeavors:

•
Corrosion is not a significant impediment to the catalyst's longevity in amine media.Both computational modeling and experimental data corroborate that the inherent corrosion of copper in these conditions does not critically limit the operational lifespan of the catalyst.This insight alleviates concerns regarding the durability of copper catalysts in practical ECR applications.

•
Primary amines, particularly MEA, demonstrate higher compatibility with ECR processes, characterized by the absence of carbonate salt precipitation and more stable potentials over time.This observation emphasizes the importance of considering the

Figure 1 .
Figure 1.H-cell three-electrode setup for electrochemical testing.Figure 1. H-cell three-electrode setup for electrochemical testing.

Figure 1 .
Figure 1.H-cell three-electrode setup for electrochemical testing.Figure 1. H-cell three-electrode setup for electrochemical testing.

Figure 6 .
Figure 6.Anodic segments from chronopotentiometry of electrodeposited copper specimen in 30 wt.% MEA, 5 m MDEA, and 30 wt.% AMP solutions purged with CO 2 .Target current density was +1 mA cm −2 .Only the last data point of each cycle is represented for clarity.

Figure 8 .
Figure 8. Optical micrographs and SEM of electrodeposited copper catalyst (a,b) prior to chronopotentiometry; (c,d) after continuous chronopotentiometry in MEA and AMP, respectively; (e) after pulse mode 2 in MEA; (f) after pulse mode 1 in AMP.

Figure 8 .
Figure 8. Optical micrographs and SEM of electrodeposited copper catalyst (a,b) prior to chronopotentiometry; (c,d) after continuous chronopotentiometry in MEA and AMP, respectively; (e) after pulse mode 2 in MEA; (f) after pulse mode 1 in AMP.

Table 3 .
Corrosion rates and time required to dissolve a 50 µm thick copper layer, calculated from corrosion current density data in Table1.