Sequential deposition of FeNC–Cu tandem CO2 reduction electrocatalysts towards the low overpotential production of C2+ alcohols

Tandem CO2 reduction electrocatalysts that combine a material that selectively produces CO with Cu are capable of producing hydrocarbons at low overpotentials and high selectivity. However, controlling the spatial distribution and the catalytic activity of the CO-making catalyst remains a challenge. In this work, a novel tandem electrocatalyst that overcomes limitations of simple Cu catalysts, namely selectivity and efficiency at low overpotential, is presented. The tandem electrocatalysts are prepared through a sequential spray coating protocol, using a single atom Fe in N-doped C (FeNC) as the selective CO-producing catalyst and commercial Cu nanopowder. The high faradaic efficiency towards CO of FeNC (99% observed at −0.60 V vs. RHE) provides a high CO coverage to the Cu particles, leading to reduced hydrogen evolution and the selective formation of ethanol and n-propanol at a much low overpotential than that of bare Cu.


Introduction
The effective utilization of CO 2 by electrochemical reduction has become an attractive strategy for mitigating greenhouse gas emissions while simultaneously producing valuable chemicals [1,2].This technology aims to convert CO 2 into a variety of high value carbon-based products, such as hydrocarbons and alcohols, which can be used as fuels and feedstocks in various industrial processes [3][4][5].To achieve this aim, a significant amount of research has focused on creating efficient catalysts that are able to activate the CO 2 molecule (where the electrolyte cation plays a crucial role) [6], convert it via proton-coupled electron transfer and release the desired products.The selectivity of the reaction strongly depends on the binding energy of the catalyst material to reaction intermediates [7].Metal-nitrogen-carbon single atom catalysts and Au or Ag are highly selective to the production of CO owing to their weak binding affinity to CO [8][9][10][11].For instance, Yan et al reported high-density Ag nanosheets that displayed a faradaic efficiency (FE) towards CO of 91.2% [12].Meanwhile, only Cu-based materials have thus far been shown to be able to further reduce CO via C-C coupling reactions and form C 2+ products [13][14][15].However, the precise control over the selectivity in Cu-based materials is challenging, leading to the production of over 16 different carbon-based chemicals limiting the selective production of alcohols and hydrocarbons.A tandem system where CO 2 is reduced to CO, and CO employed as feedstock for electrochemical C-C coupling reactions can increase the efficiency towards the formation of multi-carbon products by facilitating CO transfer and spillover [16].Spillover plays a crucial role in enhancing the catalytic activity and selectivity in the electrochemical reduction of CO 2 [17,18] by transferring intermediates or reaction products from one catalyst surface to another.Spillover occurs when the adsorption energies of the intermediates on one catalyst surface are significantly different from those on the other catalyst surface [19,20].Ag-Cu and Au-Cu tandem catalysts have been recently described, where either Ag or Au act as the CO-producing electrocatalyst which is then further reduced on the Cu moieties [21][22][23].While these core-shell structures have been proven effective to increase the production of C 2+ chemicals, whose nature is determined by specific Cu facets, controlling the spatial distribution of the CO-making electrocatalyst, and therefore the spillover mechanism remains a challenge.Additionally, the structure and catalytic activity in this kind of alloyed particles can change under applied potential in the commonly employed CO2RR electrolytes [24][25][26].Heterogeneous tandem systems with single atom catalysts, such as atomic metal in nitrogen-doped carbons, or macrocycles (porphyrins or phthalocyanines), allow an easier tuning of the spatial distribution of the catalyst that produce the CO compared to alloyed nanoparticles, facilitating the spillover mechanism.These systems also allow the establishment of correlations between the intrinsic activity of the single atom and the FE towards C 2+ products of the tandem system [27].Recently, Wang et al showed that the FE towards ethylene in tandem metal porphyrin-Cu nanoparticle catalysts strongly depends on the turnover frequency for CO (TOF CO ) production of the porphyrin motif [28].By molecular tuning of Fe-porphyrins, they achieved a 22-fold increase of the TOF CO , which on the surface of Cu nanocubes with exposed (100) facets resulted in enhanced ethylene production.While these reports opened new avenues in the synthesis of hybrid electrocatalysts for high-added value chemicals production, the low stability and TOF CO of molecular electrocatalysts compared to that of Au or carbon-based single atom catalysts hinders the overall performance of the tandem system [29,30].Additionally, the fine control of their spatial distribution remains a challenge.
To address this issues and facilitate effective spillover, we show a sequential spray coating method to synthesize tandem electrocatalysts where we employed our recently reported porous FeNC as the CO-producing electrocatalyst, which possesses record high electrochemical active site utilization (>50%) and one of the highest TOFco reported in literature to date [29,31].To target the production of alcohols, we employed commercial shape-selected Cu nanoparticles with exposed Cu (111) facets, which have been shown to form ethanol rather than ethylene due to the favored CH X -CO coupling [22].The tandem electrocatalyst are prepared through sequential deposition of a CO-making catalyst and commercial copper nanoparticles in a gas diffusion electrode; where CO 2 flows through the gas diffusion layer, is reduced initially to CO in the surface of FeNC and the efficient CO spillover to the Cu nanoparticles results in C-C coupling towards the formation of ethanol and propanol.By screening different electrode configurations, we observed that the higher FE towards C 2+ alcohols is achieved by the structure that provides a higher CO coverage in the Cu nanoparticle, in line with previous findings.The prepared materials and tandem catalysts were characterized by scanning electron microscopy (SEM), x-ray diffraction, x-ray photoelectron spectroscopy (XPS) and transmission electron microscopy (TEM) to enable optimization of the process and clear correlation between the Cu distribution and performance.

Catalyst preparation
A high surface area nitrogen-doped carbon material was prepared as recently reported by our group by pyrolysis at 900 • C of a mixture comprised of 2,4,6-triaminopyrimidine (TAP) and magnesium chloride hexahydrate [31].Firstly, TAP (97%, Sigma Aldrich) and MgCl 2 .6H 2 O (99%, Sigma Aldrich) in a weight ratio of 1:8 was ground together with a pestle and mortar.The resulting mixture was pyrolyzed under N 2 atmosphere at 900 • C for an hour using a heating rate of 5 • C min −1 and a N 2 flow rate of 0.3 l min −1 .The obtained material was ground to fine powder and subjected to an overnight washing treatment using 2 M HCl (prepared by diluting fuming 37% HCl, Merck).The purpose of this washing step was to remove any remaining MgCl 2 and MgO species [32].Subsequently, the powders were filtered, thoroughly rinsed with distilled water (500 ml) to reduce the acidity on the materials surface, and dried under at 80 • C and labeled as TAP900.
Fe coordination within TAP900 material was carried out using a low-temperature wet impregnation method under methanol reflux [33,34].Namely, a 250 ml round bottom flask was used to disperse 60 mg of TAP900 material in 75 mL of MeOH.After that, 75 ml of FeCl 2 solution in methanol (25 mM, 98%, Sigma Aldrich) was added to the mixture, and the resulting mixture was refluxed for 24 h at 90 • C.After the metal coordination reaction, the products were filtered, rinsed with methanol, and treated with 0.5 M H 2 SO 4 (95%, Sigma Aldrich) overnight to remove any aggregated species.The TAP900@Fe materials were then thoroughly rinsed with DI water (500 ml) to reduce the acidity on the materials surface and dried at 80 • C overnight.

Electrode preparation
TAP900@Fe and Cu nanopowder (99.5% and 25 nm particle size, Sigma Aldrich) were deposited onto hydrophobic carbon paper (carbon-based gas-diffusion, Sigracet 39 BB) by means of spray-coating.The active catalyst (12 mg of either FeTAP900 material or Cu nanopowder) was ground with 40 mg of polytetrafluoroethylene (PTFE) with a pestle and mortar and the mixture was diluted with 20 ml of isopropyl alcohol (IPA) and subjected to tip-sonication for 5 min.The ink was then air-sprayed manually onto a 4 × 4 cm 2 carbon paper.Finally, 800 µl of 1 wt% Sustainion (Dioxide Materials XA-9 5% in ethanol, CO 2 recycling company) was drop-casted on the surface of the spray-coated electrode and dried overnight.To create tandem electrocatalysts with different configurations, Cu nanopowder was subsequently spray-coated on top of a TAP900@Fe electrode (following the same approached disclosed previously), TAP900@Fe was spray coated on top of a Cu nanopowder electrode, and a mixture of TAP900@Fe and Cu nanopowder (12 mg of each) was spray coated on bare carbon paper.The electrodes where then labeled as TAP900@Fe-Cu, Cu-TAP900@Fe, and TAP900@Fe + Cu.Benchmark Cu electrodes were prepared by mixing 2.7 mg of Cu nanopowder, 11.25 mg of PTFE, 1.35 mg of carbon black (Vulcan XC 72), 45 uL of Nafion, 10 ml of IPA, followed by spray-coating onto a 4 × 4 cm 2 carbon paper, resulting in a mass loading of 0.15 mg cm −2 [35].

Electrochemical reduction measurements
Electrochemical reduction of CO 2 (CO 2 RR) was carried out using an AUTOLAB PGSTAT302N potentiostat and a cell with 3-electrode configuration consisting of working, counter, and reference electrodes in a homemade gas diffusion cell.As recently reported by our group, the setup of a homemade cell includes a flow field, catholyte and anolyte compartment and an end plate [29].For CO 2 RR measurements, a Pt/C electrode was employed as the counter electrode, leak-free Ag/AgCl as reference electrode [29], and the prepared electrocatalyst was coated onto hydrophobic carbon paper (carbon-based gas-diffusion, Sigracet 39 BB) and utilized as the working electrode with a geometrical area of 1 cm 2 .Fumasep FAA 3-50 or Sustainion were used as an anion exchange membrane to separate the cathode and the anode, and both the cathodic and anodic chambers contained 1.5 ml of 0.5 M KHCO 3 , prepared by saturating a 0.5 M KOH solution with CO 2 (99.999%,BOC) for 30 mins to ensure that it was fully saturated with CO 2 .Impedance spectroscopy measurements were employed to determine the ohmic resistances to enable iR-corrected potentials, the Nyquist plots are shown in figure S1.Throughout the experiments, a CO 2 flow of 10 sccm was maintained through the gas diffusion layer (GDL) into the system.Prior to electrochemical testing, the electrodes surface was preconditioned by means of cyclic voltammetries in a non-faradaic region.Meanwhile, the electrochemical reactions were carried out by varying potentials between −1.1 and −1.5 V vs. Ag/AgCl for 40 mins at each potential with 50% iR-compensation to prevent overcorrected results obtained during iR compensations and avoid potentiostatic oscillations.The potential is then manually iR-corrected for another 50% post-measurement (see table S1).The electrochemically produced gases were analyzed directly using online gas chromatography (GC) (SRI Instruments, Inc., Model 8610 C) equipped with a thermal conductivity detector and flame-ionization recorded at 5, 21, and 37 min of reaction time.Liquid products were analyzed using a high-performance liquid chromatograph (Agilent HPLC 1260 series) that had a HIPLEX-H column, a variable-wavelength detector, and a refractive index detector.FE and partial current density of the product (J product ) was calculated by applying the following equation: Where α is the number of transferred electrons, Q is the volumetric flowrate (ml min −1 ), C is the concentration of gaseous products detected by GC (vol product /vol total product ), F is the Faraday constant (96 485 C mol −1 ), V m is the molar volume of gas (ml mol −1 ), I is the total current density (mA cm −2 ), n is the number of moles of a desired product, t is the reaction time (s).

Characterization
SEM images and EDX mapping were recorded with a LEO GEMINIE 1525 (Carl Zeiss) with an x-act detector (Oxford Instruments) TEM and electron diffraction was conducted on JEOL 2100Plus operated at 200 kV.XPS were performed in a Thermo Fisher K-Alpha XPS system and the spectra were analyzed with the Avantage software.Charging effects were corrected by calibrating all spectra to the carbon C 1 s peak at 284.8 eV.XRD was conducted on an x-ray diffractometer (Bruker D2 Phaser) with Cu Kα radiation (λ = 1.5418Å), where the voltage and current of the x-ray generator were set to be 30 kV and 10 mA, respectively.The diffraction patterns of Cu nanopowders were collected from 2θ = 10 • to 85 • with the increment of 0.02 • and the step time of 1.For inductively coupled plasma mass spectrometry (ICP-MS) measurements, 5 mg of TAP 900@Fe was digested in aqua regia (25 v/v% HNO 3 (70%, Certified AR, Eur.Ph., for analysis Fisher Chemical, Fisher Scientific and 75 v/v% HCl (37%, Certified AR, Eur.Ph., for analysis Fisher Chemical, Fisher Scientific)) using a MARS 6 microwave operating at 215 • C over 15 min with 1500 W power rating.The digested solution was diluted ×20 and measured against calibration standards containing concentrations of 0, 2, 50, 100, 200, and 500 Fe ppb.

Results and discussion
A porous, Fe-NC catalyst was prepared by pyrolysis at high temperature of a mixture comprising TAP and MgCl 2 .6H 2 O and subsequent metal coordination in methanol reflux (figure 1(a)).XPS analysis of TAP900@Fe confirmed the presence of Fe (0.8 wt%, table S2, figure S2) as well as different kinds of nitrogen functionalities (4.7 wt%), namely pyridinic (398.3 eV), pyrrolic (400.9 eV), graphitic (401.9 eV) and nitrogen coordinated to Fe (399.7 eV, figure 1(c)).C1s spectra displays three different chemical species corresponding to sp2 C-C bonds (284.8 eV), C-N bonds (285.4 eV) and carbon bonded to oxygen functional groups (286.3 eV, figure 1(b)).XPS also showed the presence of oxygen functional groups in the surface of TAP900@Fe as well as weakly adsorbed species (figure 1(d)) [36].The Fe-content was further analyzed by means of ICP showing 0.37 wt%, within lower error range of our previous report [31].The chemical composition and the nature of the nitrogen functionalities matches closely with previous reports which we would like to refer the reader for further in-depth characterization (such as x-ray absorption spectroscopy, Mossbauer spectroscopy, surface area analysis, etc.) [31].Furthermore, XRD and electron diffraction patterns from TEM of the commercial Cu nanoparticles employed in this work show a mixture of Cu and Cu 2 O with Cu (111) being the main exposed facet [37] (figure S3).
To create hybrid tandem electrocatalysts based on FeNC single atoms and Cu, we spray-coated TAP900@Fe and Cu nanopowder with a hydrophobic polymer (polytetrafluoroethylene, PTFE, added to increase the hydrophobicity and suppress hydrogen evolution activity) in hydrophobic carbon paper following different configurations: First TAP900@Fe and then Cu (TAP900@Fe-Cu), first Cu and then TAP900@Fe (Cu-TAP900@Fe) and a mixture of them both (TAP900@Fe + Cu) and analyzed their chemical composition and morphology by means of SEM and XPS.Through SEM we confirmed that the morphology of TAP900@Fe consists of thin carbon layers and PTFE particles as previously observed (figure S4); and that the three hybrid electrocatalysts showed partially embedded Cu nanoparticles within their structures (figure S5), additionally cross-section images (figure S6) showed that the thickness of the Fe and Cu layer was approximately 80-95 µm.To further distinguish between the Cu and PTFE particles, TAP900@Fe-Cu was prepared in the absence of PTFE to perform EDX analysis (figure 2) and therefore to reliably identify the Cu particles.The EDX analysis cross-section (figures 2(a), (b) and S6(a)-(c)) shows that copper (with a similar morphology than that observed in figure S5) is predominantly present in the uppermost 20 µm of the catalyst layer.As the concentration of Fe falls below 1 wt%, it cannot be reliably detected.However, the presence of C from the catalyst at the same time as the Fe provide evidence for the TAP900@Fe catalyst across the catalyst layer (figure S7).From the top view EDX mapping (figures 2(c), (d) and S7(d)-(f)) it can be concluded that a mixture of Cu, Fe and C are components of the uppermost catalyst layer.These results suggest that the sequential spray-coating protocol allows to control the layout of the tandem electrocatalysts and therefore to potentially modulate the CO spillover mechanism during CO2RR.However, owing to the high surface area of TAP900@Fe (>3250 m 2 g −1 ) [31], the layout and chemical composition of the electrode in the surface and the bulk is hard to elucidate by means of SEM-EDX alone.
The chemical composition in the surface of the electrodes was analyzed by means of XPS, high resolution spectra for Cu2p, N1s and Fe2p were recorded.However, the low nitrogen and Fe content within TAP900@Fe is masked by the N1s contributions from the employed Cu nanopowder (arising from the relatively low purity of the precursor, 99.5%) and by attenuation of the Cu and F Auger lines (from the GDL which is coated in PTFE) which fall in the same binding energy region as Fe (figure S8-S10) [38].Therefore, the Cu2p spectra was employed to analyze the differences in the surface chemical composition of the electrodes prepared with different configurations (figure 3).From XPS we observed that the Cu content in the surface of the electrodes is strongly affected by the order of deposition on the carbon paper; TAP900@Fe-Cu displayed nearly 35 wt% of Cu on the surface, followed by the mixture (TAP900@Fe + Cu) with 15 wt% and Cu-TAP900@Fe with 3.4 wt% (figure 3(a)).Additionally, the Cu chemical states change with the  configuration; all the hybrid electrodes show a slight shift of the 2p 3 contribution within Cu2p towards higher binding energies which we attribute to partial oxidation by the oxygen functionalities within TAP900 (figure 3(b)) [39].This suggests intimate contact between TAP900@Fe and Cu nanopowder which would support an efficient CO spillover from TAP900@Fe to Cu.
XPS results confirm the different configuration within the tandem electrodes which will strongly impact the selectivity for the CO 2 RR.We speculate than in TAP900@Fe-Cu, CO 2 will be reduced to CO in the surface of TAP900@Fe with a high efficiency (as recently reported by our group showing >95% FE and TOF  of 4.9 e −1 site −1 s −1 at −0.59 V vs RHE) and subsequent CO spillover from the surface of TAP900@Fe-Cu nanopowder, where CO reduction takes place towards the formation of C 2+ products (figure 4) [40].
Electrochemical CO 2 RR measurements were carried out in 0.5 M KHCO 3 in a gas diffusion electrode with 40 wt% Pt/C as anode and leak-free Ag/AgCl as reference electrode.All the measurements were carried out at the same potentials-the data presented have been iR-corrected and converted to the RHE scale (table S1).Prior to the electrochemical measurements, the electrodes were conditioned performing cyclic volammetry; the voltammograms show that the tandem systems display a higher capacitive current, suggesting a higher electrochemical surface area (figure S11).Interestingly, the Cu-based tandems display a much higher oxidation current than that of a bare Cu electrode (prepared with the same loading and with PTFE), probably arising from the Cu surface oxidation and suggesting a higher exposure of the Cu sites owing to the high surface area of TAP900@Fe.Bare TAP900@Fe exhibited a very high FE for the production of CO, with the highest FE of 99% observed at −0.60 V vs. RHE (figure 5(a)), in agreement with our previous results [29].Cu nanopowder generated high-added value carbon products like formate and methanol, and even some further reduced C 2+ products, including ethanol, acetate, and n-propanol under the potential range between −0.6 V and −0.8 V vs. RHE (figure S12), while the tandem catalysts demonstrated the ability to produce selectively ethanol and propanol (depending on the spray-coating sequence), at much lower overpotentials (figure 5).The production of these carbon products, particularly ethanol and propanol under such a small overpotential, may be attributed to the spillover mechanism of CO from TAP900@Fe to Cu, resulting in the formation of C 2+ products.This phenomenon mainly occurs when utilizing Cu-based tandem catalysts or when performing CO reduction in Cu-based materials [21,40,41].In this work, we observed that the carbon products depend on the spray coating sequence, which resulted in the catalyst having different electrochemical properties.Cu-based tandem catalysts are known to produce a large amount of H 2 due to the high HER activity of Cu [20,42], thus, H 2 was observed in all tandem catalysts with a relatively high %FE and partial current density (>20% and >10 mA cm −2 , respectively figures 5, S13); above all when the Cu nanoparticles were spray coated in the first place.The lowest HER (aside from bare TAP900@Fe) was observed for TAP900@Fe-Cu which arises from the high TOF co of TAP900@Fe, that leads to a full CO * coverage in the Cu nanoparticles (figure S13(b)).As expected, the partial current density of CO production (J CO , figure S13) in the bare TAP900@Fe catalyst was higher compared to all tandem catalysts.We would like to note, that in the case of the tandem catalyst, a lower overall FE was obtained owing to the liquid product crossover (formate, acetate etc) through the anion-exchange membrane to the anodic compartment of the gas diffusion electrode as previously observed for the employed setup [43][44][45].
The liquid product distribution resulting from the tandem electrocatalysts is shown in figure 6; TAP900@Fe-Cu produced ethanol in all the screened potentials with a maximum FE of 3.2% ± 0.2% at −0.41 vs RHE, which entails a substantial decrease in overpotential versus Cu nanopowder, and even resulted in the formation of propanol at −0.63 V vs RHE with a FE of 1.2% ± 0.3%.The production of such alcohols has been suggested to arise from the coupling of two adjacent CO molecules followed by proton coupled electron transfer [46], in our system this is allowed by the spillover effect of CO, enabling subsequent surface diffusion and interaction with neighboring active sites.We would also like to note that generally, the production of propanol in CO 2 RR is relatively low <5% FE due to the limited occurrence of the pathway leading to their formation [47,48].While we acknowledge that the FE to produce alcohols shown in this work is not within the highest in the state of the art (probably due to the highly hydrophobic electrode surface achieved by PTFE and the unoptimized commercial Cu nanoparticles), the overpotential required for Figure 6.Faradaic efficiency for the liquid products of the TAP900@Fe-Cu (a), Cu-TAP900@Fe (b), and TAP900@Fe + Cu (c).
the formation of propanol is one of the lowest observed in literature (table S3) and upcoming work will target the modulation of the hydrophobicity of the Cu layer as well as the optimization of Cu content which we believe will lead to high FE for C 2+ alcohols at reduced overpotentials.
In the case of Cu-TAP900@Fe, HER was predominant in the screened potential window, and the formation of ethanol was observed at lower applied potentials with a maximum FE of 3.4% ± 0.4% at −0.43 V vs RHE, whereas propanol production was detected at higher potentials (figure 6(b)), reaching a FE of 0.4% ± 0.2% at −0.57V vs RHE.We would like to note that, while in this system TAP900@Fe is spray coated on top of a Cu electrode, as observed by SEM, the deposition of TAP900@Fe is likely to affect the Cu nanoparticle distribution reaching localized segments of the electrocatalyst where CO 2 is first reduced to CO and then CO further reduced on Cu (as expected for the TAP900@Fe-Cu system).This slight difference in product selectivity probably arises due to the increased transfer of * CO and * H species at higher applied potentials, facilitating the formation of propanol as proposed by Rahaman et al [49].Interestingly, the only liquid product observed for TAP900@Fe + Cu was formate.The formation of formate involves a direct hydrogen atom transfer to CO 2 , leading to the formation of HCOO−.This process requires fewer intermediate steps and electron/proton transfers in comparison to the complex mechanisms involved in C 2+ chemicals production [50].We hypothesize that in the case of TAP900@Fe + Cu, the CO * spillover is unoptimized and owing to its low residence time, combined with the hydrophobicity of the electrode (which arises from the PTFE coating) the production of C 2+ products is hindered.
The difference in product distribution between the three systems highlights the importance of the targeted design of tandem catalysts for CO 2 RR.TAP900@Fe-Cu, with a higher Cu surface content displayed the lowest HER owing to the initial CO 2 reduction to CO in the surface of TAP900@Fe which provides a high CO coverage in the Cu (111) facets towards ethanol and propanol production.Cu-TAP900@Fe displayed nearly twice as high HER as TAP900@Fe-Cu owing to the initial CO2RR in Cu and certain C 2+ alcohol production owing to localized tandem effects.Finally, TAP900@Fe + Cu showed intermediate HER and CO production levels and just formate as liquid product.We tested the stability of TAP900@Fe-Cu by recording the XRD patterns of the electrode after 6 h of continuous electrolysis (figures S14 and 15).Given the mixture of components in the GDL (such as carbon, PTFE and KHCO 3 residues from the electrolyte) and the likely dissolution of Cu under potentiostatic conditions, the Cu XRD patterns are attenuated, but still present, while the diffraction peaks corresponding to TAP900@Fe cannot be observed owing to the amorphous character of the material.

Conclusions
In this study, we have prepared tandem catalysts comprising a FeNC material as the CO-making catalysts and commercial Cu nanopowder with exposed Cu (111) facets for C 2+ production.The electrodes are synthesized following a sequential spray-coating deposition that leads to different conformations and electrochemical activities.The tandem catalyst that consists of Cu spray coated on FeNC displays a much lower HER than that of commercial Cu nanopowder and allowed the formation of C 2+ products at a much lower overpotential, including propanol (at −0.63 V vs RHE with a FE of 1.2 ± 0.3%).This is attributed to the spillover mechanism of CO from TAP900@Fe to Cu, which promotes the formation of C 2+ products.Moving forward, the approaches currently being studied to improve the efficiency and FE C2+ of single atom-Cu tandem electrocatalysts entail (a) modulating the CO 2 flow to screen the impact of CO residence time in the surface of Cu, (b) analyzing the effect of Cu content in the surface of a CO-making catalyst and (c) modulate the hydrophobicity of the Cu layer by tuning the amount of PTFE employed in the electrode.

Figure 2 .
Figure 2. SEM cross-section image of TAP900@Fe-Cu (a) (prepared in the absence of PTFE) with EDX mapping for Cu (b).SEM top view for the same electrode (c) and EDX imaging for Cu (d).

Figure 3 .
Figure 3. Cu2p XPS spectra of the different electrodes (a) and the surface Cu content within them (b).

Figure 4 .
Figure 4. Representation of the tandem electrocatalyst with a TAP900@Fe-Cu configuration.