Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2020 Aug 3;5(31):19453–19463. doi: 10.1021/acsomega.0c01510

The Active Center of Co–N–C Electrocatalysts for the Selective Reduction of CO2 to CO Using a Nafion-H Electrolyte in the Gas Phase

Hitoshi Ogihara †,‡, Tomomi Maezuru †, Yuji Ogishima †, Yuta Inami †, Mayuko Saito †, Shoji Iguchi †, Ichiro Yamanaka †,*
PMCID: PMC7424585  PMID: 32803039

Abstract

graphic file with name ao0c01510_0011.jpg

To contribute a solution for the global warming problem, the selective electrochemical reduction of CO2 to CO was studied in the gas phase using a [CO2(g), Co–N–C cathode | Nafion-H | Pt/C anode, H2/water] system without using carbonate solutions. The Co–N–C electrocatalysts were synthesized by partial pyrolysis of precursors in inert gas, which were prepared from various N-bidentate ligands, Co(NO3)2, and Ketjenblack (KB). The most active electrocatalyst was Co–(4,4′-dimethyl-2,2′-bipyridine)/KB pyrolyzed at 673 K, denoted Co–4,4′-dmbpy/KB(673K). A high performance of CO formation (331 μmol h–1 cm–2, 217 TOF h–1) at 0.020 A cm–2 with 78% current efficiency was obtained at −0.75 V (SHE) and 273 K under strong acidic conditions of Nafion-H. Characterization studies using extended X-ray absorption fine structure (EXAFS), X-ray photoelectron spectroscopy (XPS), transmission electron microscopy–energy-dispersive X-ray (TEM-EDX), X-ray diffraction (XRD), and temperature-programmed desorption with mass spectrometry (TPD-MS) indicated the active site as Co coordinated with four N atoms bonding the surface of KB, abbreviated Co–N4–Cx structure. A model of the reduction mechanism of CO2 on the active site was proposed.

1. Introduction

Recycling CO2 as an energy carrier of the renewable energy by the conversion of CO2 to useful chemicals such as CO, methanol, alkenes, higher hydrocarbons, etc., is promising for the reduction of the CO2 emission and also for the break from the dependency on petroleum resources. Attractive methods of the CO2 reduction have been proposed using heterogeneous catalytic,1−3 photochemical,4−6 biochemical,7 and electrochemical reactions.8−12 Strong points of the electrochemical reduction of CO2 are (i) to directly apply renewable electricity such as solar and wind power, (ii) to use various proton sources such as water, alcohol, hydrogen, and so on, (iii) to easily control reaction rates by a cathode potential, and (iv) to operate the reaction under mild conditions.

Our target product is CO, which is an industrial major material into a lot of useful chemicals and is a reductant for iron making.12 A standard potential between CO2 and CO (eq 1) is slightly negative than that between H+ and H2 (eq 2). A good reductant of CO can function as a useful intermediate material in chemical processes. The rate-determining step in eq 1 has been proposed to be the one-electron reduction of CO2 to CO2·– having a more negative reduction potential (eq 3). Although a potential of the second one-electron reduction is more positive (eq 4), it is considered that the total reaction rate of eq 1 is slow:8,9

1. 1
1. 2
1. 3
1. 4

Hydrogen evolution reaction occurs easily (eq 2) to compare with the rate-determining-step (eq 3) in general systems for the CO2 reduction. A suppression of the H2 evolution is key to achieve the selective reduction of CO2 to CO. Therefore, carbonate solutions saturated CO2 at higher pH are usually used as favorable electrolyte solutions for the reduction of CO2 with suppressing the evolution of H2 (E°/V = 0.00–0.059 pH). Hori and co-workers have done many attractive works using various metal electrodes for the electroreduction of CO2 in carbonate aqueous solutions and found active metal electrodes such as Cu, Au, Ag, and Zn.13−19

If we want to realize a fast electroreduction of the CO2 conversion, a concentration of CO2 in carbonate aqueous solutions should control the reduction rate of CO2 as equal to the current density because of the low solubility of CO2 in water (34 mM).20 A gas-electrolysis cell using gas diffusion electrodes (GDEs) should be employed for a fast-electrochemical reaction.21−30 Gaseous CO2 is directly reduced on a GDE. Our research group has reported electrosynthesis in the gas phase, which has advantages to achieve a higher formation rate and selectivity of useful chemicals.31,32 When a GDE connected with Nafion-H membrane as an electrolyte is employed for the CO2 reduction, most of the active electrocatalysts such as metallic Cu and Ag nanomaterials studied in carbonate solutions cannot work for the reduction using the Nafion-H membrane without carbonate solutions.18,19 Because the acidity of the Nafion-H membrane is too strong, the evolution of H2 dominantly proceeds during the CO2 electroreduction and the Cu and Ag electrocatalysts dissolve into Nafion-H. Therefore, most of the works using the metal nanoparticle electrocatalyst deposited on GDEs employed ion-exchange membranes and carbonate solutions or KOH solutions. Very good performances for the reduction of CO2 to CO, CH4, or C2H4 have been achieved, though their cathode potentials were very negative, less than −1.0 V (SHE).21−30 Significant amounts of carbonate or KOH solutions have to circulate for the electrolysis operation, and a large amount of CO2 dissolves in the solutions. This is a serious disadvantage to put to a practical process. Therefore, new electrocatalysts working on the GDE without using carbonate or KOH solutions should be developed for the electroreduction of CO2 in the gas phase.

We have recently reported new electrocatalysts prepared from Co(NO3)2, N-bidentate ligands, and carbon powder (KB; Ketjenblack) for the reduction of CO2 to CO by a gas-electrolysis cell using Nafion-H.33,34 Strasser et al. have reported similar electrocatalysts of M–N–C (M: Ni, Fe, Mn, Cu, Co) prepared from polyaniline, metal chlorides, and KB for the reduction of CO2 to CO and CH4 using carbonate and KOH solutions.29,30 Especially, the Ni–N–C electrocatalyst on GDE is very active for the CO2 reduction using carbonate and KOH solutions; however, the electrocatalysts do not work for the reduction of CO2 but remarkably promote H2 formation in acidic conditions.29,30 In contrast, our electrocatalysts can work for the CO2 reduction under strong acid conditions of Nafion-H without using any carbonate, KOH, and salt solutions, as shown in Figure 1.33,34 This is the particular electrocatalysis to compare with other electrocatalysts.21−30

Figure 1.

Figure 1

Open in a new tab

Schematic diagram of electroreduction of CO2(g) using a [Co–N–C cathode | Nafion-117 | Pt/KB anode] electrolysis cell.

A partial pyrolysis treatment of a raw material of Co(NO3)2, N-bidentate ligands, and KB is essential for the preparation of an effective electrocatalyst denoted Co–N–C electrocatalyst. A lot of N-bidentate ligands for Co were screened in the previous work, and several effective ligands were found; especially, 4,4′-dimethyl-2,2′-bipyridine (4,4′-dmbpy) showed a highest current efficiency of 75% with a good formation rate at −0.70 V (SHE).33 Therefore, we focused on the selective electrocatalysis of the partial pyrolyzed Co(NO3)2–4,4′-dmbpy/KB materials for the reduction and reported the basic kinetics and electrocatalysis in the previous studies.33,34 However, on the basis of the formation rate of CO, effective N-bidentate ligands were 6,6′-diamino-2,2′-bipyridine (6,6′-dabpy) and 4,4′-di-tert-butyl-2,2′-bipyridine (4,4′-dtBubpy), though their current efficiency were lower than that of 4,4′-dmbpy.33

Purposes of this work are (i) to evaluate the electrocatalysis of the Co–N–C compounds prepared from 6,6′-dabpy, 4,4′-dtBubpy, other N-bidentate ligands, and the 4,4′-dmbpy ligand to clarify the most effective electrocatalyst and (ii) to reveal a structure of the active site and a reaction mechanism of the CO2 reduction over the Co–N–C catalyst by characterization studies.

2. Results and Discussion

2.1. Effects of N-Bidentate Ligands to Co

Previously, we have compared various Co–N-bidentate-ligands/KB electrocatalysts activated at 573 K and chose the 4,4′-dimethyl-2,2′-bipyridine ligand as the best one based on a CE(CO).33,34 However, we did not confirm influences of pyrolysis temperatures of other Co–ligand/KB precursors. Therefore, effects of pyrolysis temperatures of seven precursors using ligands of 1: bpy, 2: 4,4′-dmbpy, 3: 5,5′-dmbpy, 4: 6,6′-dmbpy, 5: 6,6′-dabpy, 6: 4,4′-dtBubpy, and 7: phen were studied.

When the pyrolysis temperature was 573 K, good reproducibility of the electrocatalysts for the CO2 reduction was obtained.34 The order of ligands based on the CE(CO) was 2 (80%) > 3 (70) > 6 (55) > 5 (40) > 1 (39), 4 (39) > 7 (38) at −0.70 V. To focus on the FR(CO), the order was 5 (240 μmol h–1 cm–2) > 6 (100) > 2 (60) > 3 (45) > 4 (40) > 1 (35), 7 (35). The 5 and 6 ligands were more active than the 2 ligand. Therefore, we focused on the CEs(CO) and FRs(CO).

Figure 2 shows the effects of pyrolysis temperatures of the various electrocatalyst precursors on the electroreduction of CO2 at −0.70 V(SHE). The major products were CO and H2, and the sum of CE(CO) and CE(H2) was approximately 100%. A trace amount of CH4 was detected; however, CH4 was a minor product and a CE(CH4) was less than 0.5%. Therefore, we focused on only the CO formation in this work.

Figure 2.

Figure 2

Open in a new tab

Effects of pyrolysis temperatures of Co–ligand/KB precursors on their electrocatalysis for CO2 reduction: (a) CO formation rate, (b) CO current efficiency, and (c) current density. T: 273 K, cathode potential: −0.70 V (SHE), cathode: 1wt%Co(N-ligand)/KB(18 mg)+VGCF (30 mg)+PTFE powder (3 mg), anode: Pt(50 wt %)/KB (25 mg)+VGCF (25 mg)+PTFE powder (5 mg). CO2: 1 atm (10 mL min–1). Cathode potential: controlling by using a Ag/AgCl reference electrode but the value converted basis on the SHE. N-bidentate ligands, (1) bpy: 2,2′-bipyridine, (2) 4,4′-dmbpy: 4,4′-dimethyl-2,2′-bipyridine, (3) 5,5′-dmbpy: 5,5′-dimethyl-2,2′-bipyridine, (4) 6,6′-dmbpy: 6,6′-dimethyl-2,2′-bipyridine, (5) 6,6′-dabpy: 6,6′-diamino-2,2′-bipyridine, (6) 4,4′-dtBubpy: 4,4′-di-tert-butyl-2,2′-bipyridine, (7) phen: 1,10-phenanthroline.

When the pyrolysis temperature of the precursors was raised from 573 to 673 K, the order based on the FRs(CO) changed from 5 > 6 > 2 > 3 > 4 > 1, 7 at 573 K to 2 > 3 > 6 > 4 > 5 > 1 > 7 at 673 K. On the other hand, the order based on the CEs(CO) changed from 2 > 3 > 6 > 5 > 1, 4 > 7 at 573 K to 2 > 3 > 6 > 4, 1 > 7 > 5 at 673 K. At a higher pyrolysis temperature of 773 K for the precursors, the FRs(CO) slightly decreased and the CEs(CO) drastically decreased because of steep increases in the j values due to an acceleration of a H2 formation.

Herein, the 2 ligand showed excellent function for the acceleration on both CE(CO) and FR(CO), but the 5 and 6 ligands were not under higher temperature activation at 673 K. The Co–2-ligand/KB(673K) electrocatalyst showed the maximum FR(CO) of 253 μmol h–1 cm–2 with a high CE(CO) of 79%. At a lower pyrolysis temperature at 473 K, electroreduction activities of the Co–2, 3, 6, 4, 1, and 7-ligand/KB(473K) materials disappeared, indicating no formation of CO and H2. As can be noticed, the Co–5-ligand/KB(473K) electrocatalyst showed a significant FR(CO) and CE(CO). The 5 ligand showed a different property on pyrolysis temperatures to compare with the 2 and other ligands.

The FRs(CO) at the Co–2-ligand/KB(673K) and Co–5-ligand/KB(573K) electrocatalysts were very good; therefore, we focused on the 2 (4,4′-dmbpy) and 5 (6,6′-dabpy) ligands.

Figure 3 shows the effects of cathode potentials on the CO2 reduction by the Co–2-ligand/KB(673K) and Co–5-ligand/KB(573K) electrocatalysts. The j values and FRs(CO) increased with a decrease in cathode potential on both electrocatalysts. The CEs(CO) for the Co–2-ligand/KB(673K) electrocatalyst were almost constant between −0.55 and −0.70 V, whereas that for the Co–5-ligand/KB(573K) electrocatalyst increased with a decrease in potential. Although maxima FRs(CO) for both electrocatalysts were almost the same values at −0.75 V, the CE(CO) of 78% for the Co–2-ligand/KB(673K) electrocatalyst was much higher than that of 34% for the Co–5-ligand/KB(573K) electrocatalyst. Figure S1 in the Supporting Information shows time courses of the electroreduction of CO2 by using a fresh Co–2-ligand/KB(673K) electrocatalyst at −0.75 V and 273 K. In the early stage of the reaction for 60 min, gradual decreases in j, FR(CO), and CE(CO) were observed; however, electroreduction activities were stable after 60–240 min. The turnover number of Co on the formation of CO for 240 min was 5.35 × 104 times. This indicated that the Co–2-ligand/KB(673K) electrocatalyst was stable and active for the reduction under our reaction conditions. We could conclude that the 2 ligand was more efficient than the 5 ligand.

Figure 3.

Figure 3

Open in a new tab

Effects of cathode potentials on the electroreduction of CO2 by the (2) Co–4,4′-dmbpy/KB(673K) and (5) Co–6,6′-dabpy/KB(573K) electrocatalysts: (a) FR(CO), CE(CO) and (b) j. T: 273 K, cathode potentials: controlling by using a Ag/AgCl reference electrode, potentials in the X axis converted basis on the SHE, cathode catalyst: 1.0wt%Co–4,4′-dmbpy/KB(673K), other conditions same as in Figure 1.

2.2. Activation Process of Co–4,4′-dmbpy/KB

As described above, the 2 ligand (4,4′-dmbpy) was the most effective for the Co–ligand/KB(673K) electrocatalysts. The heat treatment of the Co–4,4′-dmbpy/KB precursor is an essential process for the generation of the active site for the electroreduction of CO2, as shown in Figure 2. To clarify the thermal decomposition processes, products (m/z values: parent signal and fragment signals) from the precursors were observed. Figure 4 shows TPD-MS profiles of 1wt%Co–4,4′-dmbpy/KB. Reference TPD-MS profiles of KB, 4,4′-dmbpy, Co(NO3)2·6H2O/KB, and 4,4′-dmbpy/KB are indicated in Figure S2 of the Supporting Information. Focusing on the reference profiles, the decomposition peak of Co(NO3)2·6H2O/KB was first detected at 490 K, as shown in Figure S2c; the main products were H2O (m/z = 18), NO (30, (14)), and CO2 (44, (12, 16, 28)). Co(NO3)2·6H2O decomposed to H2O, NO, and Co3O4. At 790 K, the evolution of CO (28, (12)), CO, and CO2 are likely to evolve during the reduction of Co3O4 to Co by a carbon support. Note that any thermal decomposition products were not detected for the profile of the KB support at <1000 K, as shown in Figure S2a. In the profile of 4,4′-dmbpy/KB in Figure S2d, no Co, the thermal decomposition of 4,4′-dmbpy was observed from 600 K. In addition to H2 (m/z = 2) and H2O desorption, significant signals assigned to hydrocarbons, m/z = 14 (CH2+), 15 (CH3+), 16 (CH4+), 27 (C2H3+), 28 (C2H4+), and 29 (C2H5+), were observed. The 4,4′-dmbpy ligand decomposed at 600–700 K, desorption of CH4 (16, (15)) at 750–900 K, and desorption of C2 hydrocarbons at >850 K (28 (27)). It is noted that a pure 4,4′-dmbpy reagent showed no decomposition products at <1200 K because of the evaporation of 4,4′-dmbpy, as shown in Figure S2b. Thus, 4,4′-dmbpy strongly adsorbed on the KB support or immobilized on the surface.

Figure 4.

Figure 4

Open in a new tab

TPD-MS profiles of 1wt%Co–4,4′-dmbpy/KB(343K) samples. Heating rate: 4 K min–1 from 423 to 1223 K, carrier gas: He (20 mL min–1).

The decomposition profile of 1wt%Co–4,4′-dmbpy/KB is shown in Figure 4. The desorption signals assigned to NO, H2O, and CO2 were observed at 500 K, corresponding to the decomposition of nitrate, as shown in Figure S2c. Over 550 K, desorption signals assigned to H2, CH4, CO2, and C2 hydrocarbons were detected, similar to Figure S2d, but their decomposition behavior was different. For example, CH4 is the major product in 4,4′-dmbpy/KB at 600–700 K in Figure S2d, while C2 hydrocarbons and CO2 are dominant for 1wt%Co–4,4′-dmbpy/KB in Figure 4. These differences suggest that the decomposition processes of 4,4′-dmbpy with and without Co are not identical.

The interesting point is that the CO2 and C2 desorption temperature at 550–650 K in Figure 4 overlapped with the heat-treatment temperature required to exhibit the CO2 reduction (573–673 K; Figure 2). Therefore, it could be assumed that the decomposition of 4,4′-dmbpy coordinated to Co is the key reaction to form the active site for the CO2 reduction. The second desorption of C2 fragments from Co–4,4′-dmbpy/KB in Figure 4 was confirmed over 750 K, at which the temperature almost corresponded to the sharp decrease in CO2 electroreduction activity over 773 K. Thus, the deep decomposition of 4,4′-dmbpy coordinated to Co would result in the loss of electrocatalytic activity of Co–4,4′-dmbpy/KB for the CO2 electroreduction.

2.3. Characterization of Co–4,4′-dmbpy/KB

As described above, the electrocatalysis of Co–4,4′-dmbpy/KB depends on pyrolysis temperatures. In order to clarify the effect of heat-treatment temperatures of the electrocatalyst precursor on a structure of Co surrounding, transmission electron microscopy (TEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and extended X-ray absorption fine structure (EXAFS) measurements were performed. To obtain a significant signal of the active site, a loading of Co was increased by 3 wt %, holding the ratio of dmbpy and Co at 5. This 3wt%Co–dmbpy/KB(673K) electrocatalyst showed good performances for the CO2 reduction to CO, as confirmed in a previous report.33

2.3.1. TEM-EDX Analysis

Figure S3 in the Supporting Information shows TEM images and energy-dispersive X-ray (EDX) analyses of the 1wt%Co–4,4′-dmbpy/KB(673K), 3wt%Co–4,4′-dmbpy/KB(673K), and 3wt%Co–4,4′-dmbpy/KB(973K) electrocatalysts. For 1wt%Co–4,4′-dmbpy/KB(673K), only the round-shaped KB support was observed; there were no particles on the KB support in Figure S3a. An EDX analysis of the 1wt%Co–4,4′-dmbpy/KB(673K) material showed that a small amount of Co was uniformly present on the KB support in Figure S3b, indicating the extremely small Co species whose size was under the resolution of high-resolution TEM. Similar TEM images were obtained for the 3wt%Co–4,4′-dmbpy/KB(673K) material in Figure S3c. The morphology of Co species was changed by the heat treatment at 973 K.

Particles of ca. 8 nm were formed for the 3wt%Co–4,4′-dmbpy/KB(973K) material in Figure S3d, and its EDX analysis indicated that the particles are composed of Co. In addition, the EDX analysis of 3wt%Co–4,4′-dmbpy/KB(973K) showed that Co was also detected in the area of the KB surface where the particles were absent in Figure S3e. Therefore, we considered that Co particles and invisible Co species coexist in the 3wt%Co–4,4′-dmbpy/KB(973K) material. The morphological change of Co from 673 to 973 K was simply recognized as a result of sintering of Co species on the KB support.

In addition, SEM-EDS mappings with a low resolution are indicated in Figure S4a–c. The contrasts of Co and N were not good; however, Co and N elements were uniformly distributed on the surface of carbon.

2.3.2. XRD Analysis

Figure S5 shows XRD patterns of the KB support and 3wt%Co–4,4′-dmbpy/KB(343K), 3wt%Co–4,4′-dmbpy/KB(673K), and 3wt%Co–4,4′-dmbpy/KB(973K) materials. The KB showed broad diffraction lines assigned to the graphite structure in the spectrum (Figure S5a). For the 3wt%Co–4,4′-dmbpy/KB(343K), numerous diffraction lines at 10°–40° would be assigned to the Co(4,4′-dmbpy) complex in the spectrum (Figure S5b). The numerous diffraction lines disappear, and new diffraction lines assigned to CoO was faintly observed for the 3wt%Co–4,4′-dmbpy/KB(673K) material in the spectrum (Figure S5c). The crystal size of CoO on the KB support was estimated as a few nanometers. This was not in conflict with the corresponding TEM image in Figure S3c. In the XRD pattern of the 3wt%Co–4,4′-dmbpy/KB(973K) material, new diffraction lines assigned to metallic Co was observed in the spectrum (Figure S5d), corresponding to the TEM image in Figure S3d. On the other hand, changes in the carbon structure were observed in the XRD data. The diffraction line of C(002) at 23.8° of the KB in the spectrum (Figure S3a) was shifted to a higher angle in Figure S3c,d. Ozkan et al. reported that the formation of the C–N structure brought a shift to a higher angle in C(200).35 This proposed a partial formation of the C–N structure in the Co–4,4′-dmbpy/KB(673, 973K) materials.

Based on the above results, the outline of structural changes in the 3wt%Co–4,4′-dmbpy/KB materials during heat treatments could be illustrated as follows: a crystal compound of Co–4,4′-dmbpy on KB was pyrolyzed to invisible Co species in the TEM images and also CoO species at 673 K. A deep pyrolysis of the invisible Co species at 973 K resulted in the formation of Co0 particles; meanwhile, the invisible Co species remained on KB.

2.3.3. XPS Analysis

Figure 5i shows C1s XPS spectra of 3wt%Co–4,4′-dmbpy/KB((a) 343K, (b) 673K, and (c) 973K) and (d) KB. The spectra in Figure 5a should reflect the two components: (comp-1) a carbon structure of the KB (mainly sp2 carbon) around 286.0 eV and (comp-2) hydrocarbon moieties in 4,4′-dmbpy on the KB surface around 285.5 eV. As the heat-treatment temperature became higher, the intensity of the comp-2 on the surface gradually decreased and the comp-1 became dominant. This reflected the thermal decomposition of 4,4′-dmbpy corresponding to the results of TPD-MS in Figure 4 and XRD in Figure S5.

Figure 5.

Figure 5

Open in a new tab

(i) C1s XPS spectra, (ii) N1s XPS spectra, and (iii) Co2p3/2 XPS spectra of (a) 3wt%Co–4,4′-dmbpy/KB(343K), (b) 3wt%Co–4,4′-dmbpy /KB(673K), and (c) 3wt%Co–4,4′-dmbpy/KB(973K) and (d) KB. N1s XPS spectra in (ii), 398.9 eV: pyridinic nitrogen (red line), 400.3 eV: N–Co (blue line), 401.5 eV: graphitic nitrogen (green line), 407.2 eV: nitrate from Co(NO3)2·6H2O (orange line).

Figure 5ii shows N1s XPS spectra of the three samples.36 The N1s spectrum of the 3wt%Co–4,4′-dmbpy/KB(343K) material could be fitted by three components having the peaks at 398.9, 400.3, and 407.2 eV. The peaks at 398.9 and 407.2 eV were assigned to pyridinic nitrogen and nitrate from Co(NO3)2·6H2O,37−39 respectively. The peak at 400.3 eV would be assigned to a Co–N center in the Co(4,4′-dmbpy)3 complex; however, to our knowledge, the XPS spectra of Co–pyridine complexes have not been reported. As similar structures to a Co–N center, N1s binding energies of Ge–N coordination by pyridine and protonated pyridine were reported to be 400.85 and 401.2 eV, respectively.37,40 The previous reports suggested that the coordination of other elements to pyridine resulted in the positive shift of N1s binding energy to compare with a 398.9 eV of pyridine, which suggested the assignment of the 400.3 eV peak to Co–N bonds.

For the N1s spectrum of the 3wt%Co–4,4′-dmbpy/KB(673K) sample in Figure 5ii, the peak of NO3 disappeared owing to the thermal decomposition, as described in the TPD-MS profile. The peaks assigned to Co–N and pyridinic nitrogen species were observed. For the 3wt%Co–4,4′-dmbpy/KB(973K) sample, peaks assigned to Co–N, pyridinic nitrogen, graphitic nitrogen (401.5 eV) species, and nitrogen bonded to oxygen (403.8 eV) were observed.37,38,41,42

The decrease in the intensity of the peak assigned to Co–N with the heat-treatment temperature implied that Co–N bonds were dissociated during the heat treatment, corresponding to the results of the TPD-MS and XRD studies. The observation of the peak assigned to graphitic nitrogen indicated that nitrogen atoms were incorporated in the carbon structure of the KB support at 973 K.

Figure 5iii shows Co2p3/2 XPS spectra of 3wt%Co–4,4′-dmbpy/KB(343, 673, and 973K). The Co(4,4′-dmbpy)3 complex was on the KB surface of the 3wt%Co–4,4′-dmbpy/KB(343K) material. Fulghum et al. measured Co2p3/2 XPS spectra of Co porphyrin and assigned several peaks to the Co–N center at 779–785 eV.42 Even though the coordination environments of Co by porphyrin and 4,4′-dmbpy are not identical, the XPS spectrum around 782 eV (779–785 eV) in Figure 5i should be assigned to the Co–N center of the Co(4,4′-dmbpy)3 complex.43 The peak around 782 eV became weaker and a broad tail appeared at a lower binding energy with increasing heat-treatment temperature. This indicated the dissociation of Co–N bonds and formation of Co compounds in 3wt%Co–4,4′-dmbpy/KB(673 and 973K).

It was reported that Co2p3/2 XPS spectra contributed with CoO and Co species were observed at 780.0 and 778.1 eV, respectively.40 It seems that a CoO compound formed on 3wt%Co–4,4′-dmbpy/KB(673K) and CoO and Co compounds formed on 3wt%Co–4,4′-dmbpy/KB(973K), which corresponded to the XRD patterns at Figure S5b,c, respectively. The XRD, TEM, and XPS data indicated that Co metal particles formed by the decomposition and reduction of the Co(4,4′-dmbpy)3 complex with carbon and a part of Co metal were oxidized in air during the characterization procedures.

2.3.4. EXAFS Analysis

Figure 6a shows Co K-edge XANES spectra of Co–dmbpy/KB(298K), Co–dmbpy/KB(673K), and Co–dmbpy/KB(673K-HNO3) samples and reference samples of Co–TPP, Co–Pc, Co(NO3)2, CoO, and Co0 obtained by transmission mode. The XANES spectra of Co–dmbpy/KB(298K) changed to Co–dmbpy/KB(673K) by the heat treatment. Co–dmbpy/KB(673K) did not show pre-edge peaks around 7715 eV observed for Co–TPP and Co–Pc, which is a characteristic peak of the Co–N4 structure with D4h symmetry.44 A broad shoulder peak appears at 7710 eV, which corresponds to the Co metal structure (Co foil reference).45 This shoulder peak was reduced by washing with HNO3 aq., Co–dmbpy/KB(673K-HNO3).

Figure 6.

Figure 6

Open in a new tab

(a) Co K-edge XANES and (b) Fourier transform of k3-weighted Co K-edge EXAFS spectra of Co–dmbpy/KB(298K), Co–dmbpy/KB(673K), and Co–dmbpy/KB(673K-HNO3) electrocatalysts and reference samples of Co–TPP, Co–Pc, Co(NO3)2, CoO, and Co0 by transmission mode.

Figure 6b shows Fourier transform of k3-weighted Co K-edge EXAFS spectra of the three Co–dmbpy/KB samples and reference samples. A peak at 2.3 Å observed for the Co foil reference is assigned to the first neighboring atom of Co. The same peak is observed for Co–dmbpy/KB(673K). Peaks at 1.8 Å observed for CoO and Co(NO3)2 reference samples could be assigned to the first neighboring atom of O. Peaks at 1.7 Å observed for Co–Pc and Co–TPP reference samples could be assigned to the first neighboring atom of N. In the case of the Co–dmbpy/KB(298K) sample, a peak at 1.7 Å could be assigned to the first neighboring atom of N of the dmbpy ligand. Peaks at 1.7 and 2.2 Å were observed in the Co–dmbpy/KB(673K) sample. These were assigned to the first neighboring atom of N derived from the dmbpy ligand and the first neighboring atom of Co in the Co metal particle, respectively. When the Co–dmbpy/KB(673K) sample was washed with HNO3 aq., the peak at 2.2 Å disappeared. A peak at 1.8 Å was observed for the Co–dmbpy/KB(673K-HNO3) sample, which is assigned to the coordination of C, N, and/or O.

Though it is difficult to determine the coordination element, it should be assigned to the first neighboring atom of N derived from the dmbpy ligand considering the XPS results.

Table 1 shows coordination numbers and distances of Co to neighboring atoms calculated from curve-fitting results. The coordination number of N to Co was about 2.3 and that of Co to Co was zero in Co–dmbpy/KB(298K). An excess amount of the dmbpy ligand (ligand:Co = 5:1) was used in the precursor preparation, as described in Section 4. A dmbpy–Co species adsorbed on the surface of carbon (KB). The coordination numbers of N to Co and Co to Co were about 2.4 and 1.8, respectively, for the Co–dmbpy/KB(673K) sample. In the Co–dmbpy/KB(673K-HNO3) sample, a coordination number of N is about 3.8. This coordination number is close to 4 of Co–Pc and Co–TPP; however, the 2.09 Å distance between Co and N of the Co–dmbpy/KB(673K-HNO3) sample was longer than the 1.92 Å distance for Co–Pc and Co–TPP.44 The coordination structure of Co and N indicated as Co–N4–Cx on the Co–dmbpy/KB(673K-HNO3) would be similar to the structures of Co–Pc and Co–TPP but not the same.

Table 1. Coordination Numbers and Distances of Co to Neighboring Atoms Calculated from the Curve-Fitting Results in Figure 5b.
Co–dmbpy/KB Co– numb. R (Å) dE a DW b
not treated N 2.32 1.90 –9.89 0.046
673-treated N 2.37 1.89 –13.97 0.071
Co 1.75 2.49 –2.25 0.085
673K-HNO3-treated N 3.75 2.09 –2.09 0.100
Open in a new tab
a

Edge shift.

b

Debye–Waller factor.

2.4. H+ and CO2 Reaction on the Active Site

Figure 7 shows the effects of partial pressure of CO2, abbreviated P(CO2), on the electroreduction of CO2 over the 4,4′-dmbpy/KB(673K) electrocatalyst. The P(CO2) at cathode was controlled by mixing CO2 and Ar by mass-flow valves, and the total flow rates were fixed at 20 mL min–1. The FR(CO) increased with increasing P(CO2). The j and CE(CO2) gradually increased with increasing P(CO2) than 0.25 atm. On the other hand, the FR(H2) was almost constant between 1.00 and 0.25 atm of P(CO2), whereas the FR(H2) was very high at P(CO2) = 0 atm. The FR(H2) should steeply increase between 0 and 0.25 atm of P(CO2).

Figure 7.

Figure 7

Open in a new tab

Effects of P(CO2) on (a) formation rates of CO and H2 and (b) current density and current efficiency to CO and H2 during the electroreduction of CO2 at the Co–4,4′-dmbpy/KB(673K) electrocatalyst. T: 273 K, cathode catalyst: 1.0wt%Co–4,4′-dmbpy/KB(673K), cathode potential: −0.70 V (SHE), gas mixture of CO2 and Ar (total 10 mL min–1), other conditions same as in Figure 2.

The dependency of FR(CO), FR(H2), CE(CO), and CE(H2) on P(CO2) indicates that the adsorption process of CO2 on the cathode contributes to the rate-determining step of the electroreduction of CO2. Apparently, the CO2 adsorption seemed to suppress the formation of H2 even at a low P(CO2). Figure S4 shows the j values in streams of Ar and CO2 at the Co–4,4′-dmbpy/KB(673K) electrocatalyst as functions of cathode potentials. The j in Ar is for the formation of H2 abbreviated “j(H2) in Ar”. The “j in CO2” is the formation of CO and H2; therefore, the j in CO2 is divided into j for the CO formation, “j(CO) in CO2”, and for the H2 formation, “j(H2) in CO2”. Figure 8 shows the replotted data of Figure S6 to ln|j| versus (η), similar to the Tafel plot (log|j| vs η). Good linearities were observed for the four j values. The slopes of plots indicate ηzF/RT values in the Butler–Volmer equation (α: charge transfer coefficient, z: number of electron).

Figure 8.

Figure 8

Open in a new tab

Effects of the cathode potential on j in CO2, j(H2) in Ar and CO2, and j(CO) in CO2. T: 273 K, cathode catalyst: 1.0wt%Co–4,4′-dmbpy/KB(673K), cathode potential: −0.55 to −0.75 V (SHE), cathode gas (10 mL min–1): CO2 or Ar, other conditions same as in Figure 1.

In the usual H2 formation on the conventional metal cathode (such as Pt) at 298.15 K, the α and z values are 0.5 and 1.0, respectively, and the αnF/RT value is 19.48, which corresponds to 118 mV/dec in the Tafel plot. In the case of j(H2) in Ar, the αnF/RT value was 15.70 (147 mV/dec) at the Co–4,4′-dmbpy/KB(673K) electrocatalyst. If the rate-determining step in the H2 formation in Ar was the one-electron reduction (z = 1), the α value was 0.37. In the case of “j in CO2”, a low αzF/RT value of 6.66 was obtained. As mentioned above, the “j in CO2” was separated into the “j(H2) in CO2” and “j(CO) in CO2” and their αzF/RT values were 6.13 (376 mV/dec) and 6.85 (336 mV/dec), respectively. If the one-electron reduction (z = 1) was the rate-determining step in the H2 formation, a small α value of 0.21 was obtained. In the case of the CO2 reduction, if the one-electron reduction (z = 1) was the rate-determining step for the CO formation, the α value was 0.23. The interpretation of lower α values was discussed in the next section.

2.5. Reaction Scheme on the Co–4,4′-dmbpy/KB Cathode

Based on the results of the characterization studies, the structural change in Co–4,4′-dmbpy/KB during the heat treatment is proposed as shown in Figure 9. The Co–4,4′-dmbpy complex on KB partially pyrolyzed and reacted with the carbon surface during the heat treatments from above 473 to 673 K as described in Figures 3–6; original Co–N bonds between Co3+ and 4,4′-dmbpy convert to new Co–N bonds assigned to the Co–N4–Cx compound, which catalyze the selective electroreduction of CO2 to CO (Figure 9a). A BET surface area of 369 m2 g–1 and a total pore volume of 0.56 cm3 g–1 for the Co–4,4′-dmbpy/KB(673K) electrocatalyst decreased to 790 m2 g–1 and 0.82 cm3 g–1 for the KB support, respectively. Micropores disappeared during the activation process at 673 K; however, the Co–4,4′-dmbpy/KB(673K) electrocatalyst has an enough surface area. A small part of the Co–4,4′-dmbpy complex on KB converts to CoO and Co0 by the 673 K treatment. The presence of CoO and Co metal is negligible to consider the CO2 reduction activity because they do not have electrocatalytic activity for the CO formation. Using higher-temperature treatments over 673 K, other Co–Ny–Cx compounds and Co0 particles formed on KB. In addition, graphitic nitrogen species were produced during the thermal activation of the carbon surface and 4,4′-dmbpy ligands at the 973 K treatment. As described so far, chemical species and structures of Co compounds on the carbon surface drastically vary during the heat treatments. The Co–N4–Cx structure mentioned on Table 1 and in Figure 9a is the active site for the CO2 electroreduction to CO in our electrolysis system; however, a formula and structure of Cx indicated by hashed lines in Figure 9 have not been clarified.

Figure 9.

Figure 9

Open in a new tab

Reaction scheme for (a) conversion of Co–4,4′-dmnpy to the active site of Co–N4–Cx on the KB surface, (b) electroreduction of H+ to H2 formation on the active site, and (c) electroreduction of CO2 to CO on the active site.

Though the electrolysis conditions such as reaction phase, supporting electrolyte, pH value, etc., were very different from our conditions, it has been already reported that the Co–N4 center in Co–TPP and Co–Pc was effective for the CO2 reduction.21,23,29,39,46−50 We have also reported that Co–phthalocyanine supported on a carbon support (vapor-grown carbon fiber, Showa Denko Co.) showed a lower electrocatalytic activity for the reduction of CO2 to CO in the gas phase.50 These facts proposed that a Co–N4 structure was effective for the CO2 reduction. However, a 2.09 Å length of Co–N bonds of the Co–N4–Cx structure observed in the Co–4,4′-dmbpy/KB(673K) is longer than a 1.92 Å length of Co–TPP and Co–Pc, as described in Table 1.47 The particular structure of Co–N4–Cx would be suitable for the selective reduction of CO2 to CO.

As described in the previous section, the charge transfer coefficient, α value, of 0.37 for j(H2) in Ar was smaller than the typical value of 0.5 observed for the H2 formation on the conventional Pt electrode. If the active site for the electroreduction of H+ is the Co–N4–Cx site, then the Co–N4–Cx site is isolated on the KB surface, as described in the XAFS study. Atomic hydrogen species on the Co–N4–Cx site should migrate to the other atomic hydrogen species and coupled to H2, as indicated in Figure 9b.

Therefore, the migration process of atomic hydrogen species should affect the j value (FR(H2)). The α value of 0.23 for the “j(H2) in CO2” was about a half of that of “j(H2) in Ar”. This indicates that the H2 formation on the Co–4,4′-dmbpy/KB(673K) electrocatalyst was remarkably inhibited in CO2. We propose from this fact that the adsorption of CO2 on the electrocatalyst surface inhibits the migration of atomic hydrogen species and decelerate the H2 formation. The α value of 0.23 for the CO formation was also small. The electrochemical reduction was slow, and the chemical reaction affected the rate-determining step in the CO formation. We propose from this fact that atomic hydrogen species on the Co–N4–Cx site react with the adsorbed CO2 on the surface. This chemical reaction should be slow, and the adducts of hydrogen species and CO2 on the Co–N4–Cx site finally produce CO, as shown in Figure 9c.

As described in Section 1, Strasser’s group reported the active Ni–N–C electrocatalyst for the reduction of CO2 to CO using carbonate and KOH solutions and identified the active site of the Ni–N4–Cx structure; however, the Ni and other metal electrocatalysts did not work for the reduction of CO2 but remarkably promoted H2 formation in acidic conditions at pH < 1.29,30 Strasser’s group Ni–N4–Cx structure and our Co–N4–Cx structure are very similar; however, our Co–N4–Cx electrocatalyst was very active and stable (Figure S1) for the CO2 reduction under strong acidic conditions of Nafion-H at pH < 0. This fact suggests that the fine structure is different.

As described in Figures 2 and 3, the Co–6,6′-dabpy/KB precursor was activated at a lower temperature of 473 K and showed maxima FR(CO) and CE(CO) by the 573 K treatment. We could not reveal the activation process of the Co–6,6′-dabpy/KB precursor and a structure of active site on the Co–6,6′-dabpy/KB(573K) electrocatalyst. It is interesting that the active structures of the two electrocatalysts are the same or not. By the way, we have studied the direct synthesis H2O2 by the electroreduction of O2 at the Co–bipyridine, phenanthroline, and porphyrin derivatives supported on carbons heat-treated around 1023 K.29,51−54 Recently, the synthesis of a high concentration of pure H2O2 aqueous solutions with 18.7 wt % (5.5 M) was achieved by using an electrocatalyst prepared from Co–TPP/KB pyrolyzed at 1023 K.54 The structure of the active site was identified as Co bicoordinated with N on the carbon surface, abbreviated Co–N2–Cx structure, by characterization studies using EXAFS, XPS, etc.51−53 Although the initial Co loading of 0.05 wt % for the Co–N2–Cx electrocatalyst is far from that of 1.0 wt % for the Co–4,4′-dmbpy/KB(673K) electrocatalyst, the Co–N2–Cx electrocatalyst did not show electrocatalytic activity for the CO2 reduction;54 in contrast, the Co–4,4′-dmbpy/KB(673K) electrocatalyst, Co–N4–Cx, was not active for the H2O2 formation. The two active sites consist of the same elements (Co, N, and C), but they perform different and unique electrocatalysis, respectively. We will report details in electrocatalysis depending on y in Co–Ny–Cx compounds prepared from Co–4,4′-dmbpy/KB, Co–6,6′-dabpy/KB, and Co–TPP/KB precursors.

3. Conclusions

Electroreduction of CO2 to CO using the gas-electrolysis cell with the Nafion-H membrane electrolyte (strong acidic conditions) and electrocatalysts prepared by pyrolysis of N-bidentate ligands, Co(NO3)2, and KB was studied. Pyrolysis temperatures remarkably affected their electrocatalytic activities. The Co–4,4′-dmbpy/KB(673K) electrocatalyst showed the highest performance for the CO2 reduction to CO among the electrocatalysts tested in this work. The Co–N4–Cx structure was identified as the active site for the reduction of CO2 to CO from the characterization studies using EXAFS, XPS, TEM-EDX, XRD, and TPD-MS. The Co–N4–Cx structure resembled Co–TPP and Co–Pc, but the distance of the Co–N bonds was longer than those of Co–TPP and Co–Pc. This unique structure would be suitable for the reduction of CO2. How to synthesize the Co–N4–Cx active site with a higher concentration on the GDE is essential to realize an actual CO2 reduction process. The onset potential for the CO formation on the Co–4,4′-dmbpy/KB(673K) electrocatalyst was about −0.50 V (SHE), which was fairly positive than other studies;21−28,47−49 however, we have to improve the electrocatalysis and the onset potential near −0.11 V (eq 1) and to achieve a significant j value, for example, 100 mA cm–2, with a higher CE(CO) under strong acidic conditions to contribute a solution for the global warming problem.

4. Experimental Section

4.1. Electrocatalyst Preparation

N-bidentate ligands used in this study were 2,2′-bipyridine (1: bpy), 4,4′-dimethyl-2,2′-bipyridine (2: 4,4′-dmbpy), 5,5′-dimethyl-2,2′-bipyridine (3: 5,5′-dmbpy), 6,6′-dimethyl-2,2′-bipyridine (4: 6,6′-dmbpy), 6,6′-diamino-2,2′-bipyridine (5: 6,6′-dabpy), 4,4′-di-tert-butyl-2,2′-bipyridine (6: 4,4′-dtBubpy), and 1,10-phenanthroline (7: phen). A ligand and Co(NO3)2·6H2O with a ratio of 5 (mol/mol) were stirred in ethanol to form Co complexes. Ketjenblack EC300J powder (KB; 790 m2 g–1) was added in the solutions of Co complexes and mixed well using a magnetic stir bar. Electrocatalyst precursors were prepared by drying the ethanol solvent in the mixture. The precursor powder was placed in a flat-bottom quartz reactor and was activated by the heat treatment in a He stream for 1 h at 423–973 K for 3 h. A loading of Co was 1.00 wt %.

When we used N2 or Ar instead of He during the catalyst synthesis, very good reproducibility of catalytic activities was obtained. We monitored desorbed products such as CO, CO2, H2O, N2, N2O, CH4, and light hydrocarbons from the catalyst precursors by TCD gas chromatography with He carrier gas. Therefore, we used He during the catalyst synthesis.

4.2. Cathode and Anode Preparation

A round cathode sheet (0.1 mm-thick, 2 cm2) was prepared by the hot-press method using the prepared 1wt%Co(N-ligand)/KB electrocatalyst (18 mg), vapor-grown carbon fiber (VGCF; Showa Denko Co.; 30 mg), and PTFE powder (F-104, Daikin Co.; 3 mg). A content of Co in a sheet of the cathode was 0.18 mg (3.05 μmol). A loading of Co was 1.53 μmol cm–2. The anode was prepared from Pt(50 wt %)/KB (25 mg), VGCF (25 mg), and the PTFE powder (5 mg) in a similar way. A loading of Pt was 32.05 μmol cm–2.

4.3. Electrolysis Procedures

The membrane electrode assembly (MEA) was fabricated by hot-pressing the cathode, Nafion 117, and the anode under 59 MPa at 413 K. An electrolysis cell of diaphragm type was assembled using the MEA, glass flange parts, and PTFE parts, as indicated in Figure 1. Ar (20 mL min–1) was flowed through the cell to purge air and then pure H2 (20 mL min–1) and pure CO2 (10 mL min–1) were introduced into the anode and cathode compartments, respectively. In addition, to humidify the Nafion membrane, H2 was bubbled through deionized water at 273 K and the relative humidity was about 80%. Electrochemical reduction was performed under potentiostatic conditions using an electrochemical measurement system (Hokuto Denko Co. HZ-5000) and a Ag/AgCl reference electrode (0.197 V vs SHE) at 273 K. Potentials in this paper were converted to those based on the SHE.

When we replaced the anode reaction from H2 oxidation to water oxidation, total voltages between the cathode and anode increased but the same electroreduction of CO2 at the Co(N-ligand)/KB cathode occurred. To focus on the cathode reaction and the electrocatalysis, we employed a fast anode reaction of H2 oxidation. In addition, the anode reaction using H2 can function as an apparent reversible hydrogen electrode (RHE). We utilized the apparent RHE to evaluate the reliability of the cathode potentials.

4.4. Product Analysis

After applying each cathode potential from −0.5 to −0.75 V (SHE) for 30 min, we observed steady states of the electroreduction of CO2. A reaction gas mixture from the cathode was analyzed three times for every 10 min using an online six-way valve connected to a gas chromatograph (CO analysis: Shimadzu GC-8APT with a TCD detector, a 5 Å molecular sieves column, and He carrier gas) and using a sampling port and a gas-tight microsyringe (H2 analysis: Shimadzu GC-8APT with a TCD detector, an activated-carbon column, and Ar carrier gas). Product yields were calculated from gas chromatography data and calibration factors. Formation rates (FR) of CO and H2 were calculated from the yields and outlet flow rates. Current efficiencies (CE) to CO and H2 were evaluated from (eq 5):

4.4. 5

4.5. TPD-MS Measurements

Temperature-programmed desorption mass spectrometry measurements were performed by combining a catalyst analyzer (BELCAT-B, MicrotracBEL Corp.) and a gas analytical system equipped with quadrupole mass spectrometer (M-200GA-DM, Anelva Corp.). The sample powder was placed in a quartz tube and attached to the catalyst analyzer. The sample was heated to 423 K at 10 K min–1 and held for 1 h, and subsequently the temperature was raised to 1223 K at 4 K min–1. The measurement was conducted in a He gas stream of 20 mL min–1. A part of the outlet gas was introduced into the gas analytical system by using a capillary tube to obtain the TPD-MS profile.

4.6. TEM and SEM Observation

Transmission electron microscopy images were measured using a JEM-2010F (JEOL) microscope equipped with an energy-dispersive X-ray spectrometry instrument (Genesis). Scanning electron microscopy images were measured using a JCM-6000Plus (JEOL).

4.7. XRD Analysis

X-ray diffraction patterns were measured using a MiniFlex 600/TISS (Rigaku) diffractometer using Cu Kα X-ray (1.5418 Å).

4.8. XPS Measurements

X-ray photoelectron spectroscopy measurements were carried out using a JPS-9010MC (JEOL) equipped with a monochromatic Al Kα source operated at 25 kV and 10 mA. A C1s binding energy of 284.6 eV was used for the charge collection. Background spectra were subtracted by the Shirley method. N1s spectra were curve-fitted using Lorentzian–Gaussian combination peaks.

4.9. XAFS Measurement

XAFS spectra of Co K-edge were measured by transmission mode at the beamline BL9C of Photon Factory (PF) of the Institute for Materials Structures Science, High Energy Accelerator Research Organization (KEK-IMSS), Tsukuba, Japan. The X-ray beam was monochromatized using a Si(111) double-crystal monochromator. The X-ray energy was calibrated using Co0 foil (E° = 7709 eV). XAFS samples were prepared by mixing the electrocatalyst and BN powders and pressing into a pellet (10 mm in diameter). The X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectra were analyzed using the REX2000 software (Rigaku). The backscattering amplitudes and the phase shifts were calculated using the FEFF program by setting S02 to 1.0 without SCF calculations.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c01510.

  • (Figure S1) Time courses of the reduction, (Figure S2) additional TPD-Mass data, (Figure S3) TEM-EDX images of Co–dmbpy/KB materials, (Figure S4) SEM-EDS mapping, (Figure S5) XRD patterns of Co–dmbpy/KB materials, and (Figure S6), j-potential data for CO and H2 formations (PDF)

The authors declare no competing financial interest.

Supplementary Material

ao0c01510_si_001.pdf (756.5KB, pdf)

References

  1. Kondratenko E. V.; Mul G.; Baltrusaitis J.; Larrazábal G. O.; Pérez-Ramírez J. Status and perspectives of CO2 conversion into fuels and chemicals by catalytic, photocatalytic and electrocatalytic processes. Energy Environ. Sci. 2013, 6, 3112–3135. 10.1039/c3ee41272e. [DOI] [Google Scholar]
  2. Behrens M. Heterogeneous Catalysis of CO2 Conversion to Methanol on Copper Surfaces. Angew. Chem., Int. Ed. 2014, 53, 12022–12024. 10.1002/anie.201409282. [DOI] [PubMed] [Google Scholar]
  3. Herron J. A.; Kim J.; Upadhye A. A.; Huber G. W.; Maravelias C. T. A general framework for the assessment of solar fuel technologies. Energy Environ. Sci. 2015, 8, 126–157. 10.1039/C4EE01958J. [DOI] [Google Scholar]
  4. Morris A. J.; Meyer G. J.; Fujita E. Molecular Approaches to the Photocatalytic Reduction of Carbon Dioxide for Solar Fuels. Acc. Chem. Res. 2009, 42, 1983–1994. 10.1021/ar9001679. [DOI] [PubMed] [Google Scholar]
  5. Dhakshinamoorthy A.; Navalon S.; Corma A.; Garcia H. Photocatalytic CO2 reduction by TiO2 and related titanium containing solids. Energy Environ. Sci. 2012, 5, 9217–9233. 10.1039/c2ee21948d. [DOI] [Google Scholar]
  6. Mao J.; Li K.; Peng T. Recent advances in the photocatalytic CO2 reduction over semiconductors. Catal. Sci. Technol. 2013, 3, 2481–2498. 10.1039/c3cy00345k. [DOI] [Google Scholar]
  7. Shi J.; Jiang Y.; Jiang Z.; Wang X.; Wang X.; Zhang S.; Han P.; Yang C. Enzymatic conversion of carbon dioxide. Chem. Soc. Rev. 2015, 44, 5981–6000. 10.1039/C5CS00182J. [DOI] [PubMed] [Google Scholar]
  8. Benson E. E.; Kubiak C. P.; Sathrum A. J.; Smieja J. M. Electrocatalytic and homogeneous approaches to conversion of CO2 to liquid fuels. Chem. Soc. Rev. 2009, 38, 89–99. 10.1039/B804323J. [DOI] [PubMed] [Google Scholar]
  9. Whipple D. T.; Kenis P. J. A. Prospects of CO2 Utilization via Direct Heterogeneous Electrochemical Reduction. J. Phys. Chem. Lett. 2010, 1, 3451–3458. 10.1021/jz1012627. [DOI] [Google Scholar]
  10. Martín A. J.; Larrazábal G. O.; Pérez-Ramírez J. Towards sustainable fuels and chemicals through the electrochemical reduction of CO2: lessons from water electrolysis. Green Chem. 2015, 17, 5114–5130. 10.1039/C5GC01893E. [DOI] [Google Scholar]
  11. Lu Q.; Rosen J.; Jiao F. Nanostructured Metallic Electrocatalysts for Carbon Dioxide Reduction. ChemCatChem 2015, 7, 38–47. 10.1002/cctc.201402669. [DOI] [Google Scholar]
  12. Kato Y. Hydrogen Utilization for Carbon Recycling Iron Making System. ISIJ Int. 2012, 52, 1433–1438. 10.2355/isijinternational.52.1433. [DOI] [Google Scholar]
  13. Hori Y.; Wakebe H.; Tsukamoto T.; Koga O. Electrocatalytic process of CO selectivity in electrochemical reduction of CO2 at metal electrodes in aqueous media. Electrochim. Acta 1994, 39, 1833. 10.1016/0013-4686(94)85172-7. [DOI] [Google Scholar]
  14. Chen Y.; Li C. W.; Kanan M. W. Aqueous CO2 Reduction at Very Low Overpotential on Oxide-Derived Au Nanoparticles. J. Am. Chem. Soc. 2012, 134, 19969–19972. 10.1021/ja309317u. [DOI] [PubMed] [Google Scholar]
  15. Zhu W.; Michalsky R.; Metin Ö.; Lv H.; Guo S.; Wright C. J.; Sun X.; Peterson A. A.; Sun S. Monodisperse Au Nanoparticles for Selective Electrocatalytic Reduction of CO2 to CO. J. Am. Chem. Soc. 2013, 135, 16833–16836. 10.1021/ja409445p. [DOI] [PubMed] [Google Scholar]
  16. Hori Y.; Ito H.; Okano K.; Nagasu K.; Sato S. Silver-coated ion exchange membrane electrode applied to electrochemical reduction of carbon dioxide. Electrochim. Acta 2003, 48, 2651–2657. 10.1016/S0013-4686(03)00311-6. [DOI] [Google Scholar]
  17. Lu Q.; Rosen J.; Zhou Y.; Hutchings G. S.; Kimmel Y. C.; Chen J. G.; Jiao F. A selective and efficient electrocatalyst for carbon dioxide reduction. Nat. Commun 2014, 5, 3242–3242. 10.1038/ncomms4242. [DOI] [PubMed] [Google Scholar]
  18. Rosen J.; Hutchings G. S.; Lu Q.; Rivera S.; Zhou Y.; Vlachos D. G.; Jiao F. Mechanistic Insights into the Electrochemical Reduction of CO2 to CO on Nanostructured Ag Surfaces. ACS Catal. 2015, 5, 4293–4299. 10.1021/acscatal.5b00840. [DOI] [Google Scholar]
  19. Hsieh Y.-C.; Senanayake S. D.; Zhang Y.; Xu W.; Polyansky D. E. Effect of Chloride Anions on the Synthesis and Enhanced Catalytic Activity of Silver Nanocoral Electrodes for CO2 Electroreduction. ACS Catal. 2015, 5, 5349–5356. 10.1021/acscatal.5b01235. [DOI] [Google Scholar]
  20. Aschenbrenner O.; Styring P. Comparative study of solvent properties for carbon dioxide absorption. Energy Environ. Sci. 2010, 3, 1106–1113. 10.1039/c002915g. [DOI] [Google Scholar]
  21. Maeda M.; Kitaguchi Y.; Ikeda S.; Ito K. Reduction of carbon dioxide on partially-immersed Au plate electrode and Au-SPE electrode. J. Electroanal. Chem. 1987, 238, 247–258. 10.1016/0022-0728(87)85178-1. [DOI] [Google Scholar]
  22. Mahmood M. N.; Masheder D.; Harty C. J. Use of gas-diffusion electrodes for high-rate electrochemical reduction of carbon dioxide. I. Reduction at lead, indium- and tin-impregnated electrodes. J. Appl. Electrochem. 1987, 17, 1159–1170. 10.1007/BF01023599. [DOI] [Google Scholar]
  23. Furuya N.; Matsui K. Electroreduction of carbon dioxide on gas-diffusion electrodes modified by metal phthalocyanines. J. Electroanal. Chem. 1989, 271, 181–191. 10.1016/0022-0728(89)80074-9. [DOI] [Google Scholar]
  24. Cook R. L.; MacDuff R. C.; Sammells A. F. High Rate Gas Phase CO2 Reduction to Ethylene and Methane Using Gas Diffusion Electrodes. J. Electrochem. Soc. 1990, 137, 607–608. 10.1149/1.2086515. [DOI] [Google Scholar]
  25. Delacourt C.; Ridgway P. L.; Kerr J. B.; Newman J. Design of an Electrochemical Cell Making Syngas (CO+H2) from CO2 and H2O Reduction at Room Temperature. J. Electrochem. Soc. 2008, 155, B42–B49. 10.1149/1.2801871. [DOI] [Google Scholar]
  26. Dufek E. J.; Lister T. E.; Stone S. G.; McIlwain M. E. Operation of a Pressurized System for Continuous Reduction of CO2. J. Electrochem. Soc. 2012, 159, F514–F517. 10.1149/2.011209jes. [DOI] [Google Scholar]
  27. Kang P.; Zhang S.; Meyer T. J.; Brookhart M. Rapid selective electrocatalytic reduction of carbon dioxide to formate by an iridium pincer catalyst immobilized on carbon nanotube electrodes. Angew. Chem., Int. Ed. 2014, 53, 8709–8713. 10.1002/anie.201310722. [DOI] [PubMed] [Google Scholar]
  28. Liu Z.; Yang H.; Kutz R.; Masel R. I. CO2 Electrolysis to CO and O2 at High Selectivity, Stability and Efficiency Using Sustainion Membranes. J. Electrochem. Soc. 2018, 165, J3371–J3377. 10.1149/2.0501815jes. [DOI] [Google Scholar]
  29. Móller T.; Ju W.; Bagger A.; Wang X.; Luo F.; Thanh T. N.; Varela A. S.; Rossmeisl J.; Strasser P. Efficient CO2 to CO electrolysis on solid Ni-N-C catalysts at industrial current density. Energy Environ. Sci. 2019, 12, 640–647. 10.1039/C8EE02662A. [DOI] [Google Scholar]
  30. Varela A. S.; Ju W.; Bagger A.; Franco P.; Rossmeisl J.; Strasser P. Electrochemical Reduction of CO2 on Metal-Nitrogen-Doped Carbon Catalysts. ACS Catal. 2019, 9, 7270–7284. 10.1021/acscatal.9b01405. [DOI] [Google Scholar]
  31. Yamanaka I. Direct and Safe Synthesis of Hydrogen Peroxide from Oxygen and Hydrogen Using Fuel Cell Reactors. J. Jpn. Pet. Inst. 2014, 57, 237–250. 10.1627/jpi.57.237. [DOI] [Google Scholar]
  32. Otsuka K.; Yamanaka I. Electrochemical cells as reactors for selective oxygenation of hydrocarbons at low temperature. Catal. Today 1998, 41, 311–325. 10.1016/S0920-5861(98)00022-4. [DOI] [Google Scholar]
  33. Ogihara H.; Maezuru T.; Ogishima Y.; Yamanaka I. Electrochemical Reduction of CO2 to CO by a Co-N-C Electrocatalyst and PEM Reactor at Ambient Conditions. ChemistrySelect 2016, 1, 5533–5537. 10.1002/slct.201601082. [DOI] [Google Scholar]
  34. Ogihara H.; Maezuru T.; Ogishima Y.; Yamanaka I. Electrocatalytic Activity of Co-4,4’dimethyl-2,2’-bipyridine Supported on Ketjenblack for Reduction of CO2 to CO Using PEM Reactor. Electrocatalysis 2018, 9, 220–225. 10.1007/s12678-017-0419-1. [DOI] [Google Scholar]
  35. Matter P. H.; Zhang L.; Ozkan U. S. The role of nanostructure in nitrogen-containing carbon catalysts for the oxygen reduction reaction. J. Catal. 2006, 239, 83–96. 10.1016/j.jcat.2006.01.022. [DOI] [Google Scholar]
  36. Rehr J. J.; Albers R. C.; Zabinsky S. I. High-order multiple-scattering calculations of x-ray-absorption fine structure. Phys. Rev. Lett. 1992, 69, 3397–3400. 10.1103/PhysRevLett.69.3397. [DOI] [PubMed] [Google Scholar]
  37. Pels J. R.; Kapteijn F.; Moulijn J. A.; Zhu Q.; Thomas K. M. Evolution of nitrogen functionalities in carbonaceous materials during pyrolysis. Carbon 1995, 33, 1641–1653. 10.1016/0008-6223(95)00154-6. [DOI] [Google Scholar]
  38. Casanovas J.; Ricart J. M.; Rubio J.; Illas F.; Jiménez-Mateos J. M. Origin of the Large N 1s Binding Energy in X-ray Photoelectron Spectra of Calcined Carbonaceous Materials. J. Am. Chem. Soc. 1996, 118, 8071–8076. 10.1021/ja960338m. [DOI] [Google Scholar]
  39. Hendrickson D. X.; Hollander J.; Jolly W. L. Nitrogen ls electron binding energies. Correlations with molecular orbital calculated nitrogen charges. Inorg. Chem. 1969, 8, 2642–2647. 10.1021/ic50082a020. [DOI] [Google Scholar]
  40. Bae S.-S.; Kim S.; Kim J. W. Langmuir 2009, 25, 275–279. 10.1021/la8021722. [DOI] [PubMed] [Google Scholar]
  41. Niwa H.; Kobayashi M.; Horiba K.; Harada Y.; Oshima M.; Terakura K.; Ikeda T.; Koshigoe Y.; Ozaki J.-i.; Miyata S.; Ueda S.; Yamashita Y.; Yoshikawa H.; Kobayashi K. X-ray photoemission spectroscopy analysis of N-containing carbon-based cathode catalysts for polymer electrolyte fuel cells. J. Power Sources 2011, 196, 1006–1011. 10.1016/j.jpowsour.2010.08.054. [DOI] [Google Scholar]
  42. Artyushkova K.; Levendosky S.; Atanassov P.; Fulghum J. XPS Structural Studies of Nano-composite Non-platinum Electrocatalysts for Polymer Electrolyte Fuel Cells. Top. Catal. 2007, 46, 263–275. 10.1007/s11244-007-9002-y. [DOI] [Google Scholar]
  43. Biesinger M. C.; Payne B. P.; Grosvenor A. P.; Lau L. W. M.; Gerson A. R.; Smart R. S. C. Resolving surface chemical states in XPS analysis of first row transition metals, oxides and hydroxides: Cr, Mn, Fe, Co and Ni. Appl. Surf. Sci. 2011, 257, 2717–2730. 10.1016/j.apsusc.2010.10.051. [DOI] [Google Scholar]
  44. Alves M. C. M.; Dodelet J. P.; Guay D.; Ladouceur M.; Tourillon G. Origin of the electrocatalytic properties for oxygen reduction of some heat-treated polyacrylonitrile and phthalocyanine cobalt compounds adsorbed on carbon black as probed by electrochemistry and x-ray absorption spectroscopy. J. Phys. Chem. 1992, 96, 10898–10905. 10.1021/j100205a054. [DOI] [Google Scholar]
  45. Zhou J.; Yang Y.; Li C.; Zhang S.; Chen Y.; Shi S.; Wei M. Synthesis of Co–Sn intermetallic nanocatalysts toward selective hydrogenation of citral. J. Mater. Chem. A 2016, 4, 12825–12832. 10.1039/C6TA04542A. [DOI] [Google Scholar]
  46. Vallejos-Burgos F.; Utsumi S.; Hattori Y.; García X.; Gordon A. L.; Kanoh H.; Kaneko K.; Radovic L. R. Pyrolyzed phthalocyanines as surrogate carbon catalysts: Initial insights into oxygen-transfer mechanisms. Fuel 2012, 99, 106–117. 10.1016/j.fuel.2012.03.055. [DOI] [Google Scholar]
  47. Manbeck G. F.; Fujita E. A review of iron and cobalt porphyrins, phthalocyanines and related complexes for electrochemical and photochemical reduction of carbon dioxide. J. Porphyrins Phthalocyanines 2015, 19, 45–64. 10.1142/S1088424615300013. [DOI] [Google Scholar]
  48. Atoguchi T.; Aramata A.; Kazusaka A.; Enyo M. Electrocatalytic activity of CoII TPP-pyridine complex modified carbon electrode for CO2 reduction. J. Electroanal. Chem. 1991, 318, 309–320. 10.1016/0022-0728(91)85312-D. [DOI] [Google Scholar]
  49. Kramer W. W.; McCrory C. C. L. Polymer coordination promotes selective CO2 reduction by cobalt phthalocyanine. Chem. Sci. 2016, 7, 2506–2515. 10.1039/C5SC04015A. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Yamanaka I.; Tabata K.; Mino W.; Furusawa T. Electroreduction of Carbon Dioxide to Carbon Monoxide by Co-pthalocyanine Electrocatalyst under Ambient Conditions. ISIJ Int. 2015, 55, 399–403. 10.2355/isijinternational.55.399. [DOI] [Google Scholar]
  51. Yamanaka I.; Ichihashi R.; Iwasaki T.; Nishimura N.; Murayama T.; Ueda W.; Takenaka S. Electrocatalysis of heat-treated cobalt-porphyrin/carbon for hydrogen peroxide formation. Electrochim. Acta 2013, 108, 321–329. 10.1016/j.electacta.2013.06.072. [DOI] [Google Scholar]
  52. Yamanaka I.; Tazawa S.; Murayama T.; Iwasaki T.; Takenaka S. Catalytic synthesis of neutral hydrogen peroxide at a CoN2Cx cathode of a polymer electrolyte membrane fuel cell (PEMFC). ChemSusChem 2010, 3, 59–62. 10.1002/cssc.200900246. [DOI] [PubMed] [Google Scholar]
  53. Yamanaka I.; Tazawa S.; Murayama T.; Ichihashi R.; Hanaizumi N. Catalytic Synthesis of Neutral H2O2 Solutions from O2 and H2 by a Fuel Cell Reaction. ChemSusChem 2008, 1, 988–992. 10.1002/cssc.200800176. [DOI] [PubMed] [Google Scholar]
  54. Iwasaki T.; Masuda Y.; Ogihara H.; Yamanaka I. Direct Synthesis of Pure H2O2 Aqueous Solution by CoTPP/Ketjen-Black Electrocatalyst and the Fuel Cell Reactor. Electrocatalysis 2018, 9, 236–242. 10.1007/s12678-017-0444-0. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao0c01510_si_001.pdf (756.5KB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES