Selective production of methanol by the electrochemical reduction of CO₂ on boron-doped diamond electrodes in aqueous ammonia solution

Abstract

The electrochemical reduction of CO₂ was investigated in an aqueous ammonia solution using boron-doped diamond electrodes. Methanol was mainly produced by reduction at a potential of −1.3 V (vs. Ag/AgCl) with a faradaic efficiency as high as 24.3%. Also, even in an aqueous ammonium bicarbonate solution (pH 7.9) without CO₂ bubbling, methanol was produced, while no methanol production was observed at higher (10.6) and lower pH (3.38). These observations suggest that the selectivity for methanol production in aqueous ammonia solutions is due to the electrochemical reduction of bicarbonate ions which are formed by the reaction between ammonia and CO₂. Moreover, we present the important role of ammonia as an electrolyte for the selective production of methanol by electrochemical reduction of CO₂.

---

Introduction

CO₂ reduction is an important next-generation technology and has been widely studied by researchers, since it emphasizes the value of processes by producing chemicals or fuel. Although the high thermodynamic stability of CO₂ makes its reduction difficult in general, efficient and suitable processes must be achieved. One useful strategy is to develop suitable catalysts, for example, the Ru(II)–Re(I) supramolecular photocatalyst,¹ Re(CO)₃(bpy)Cl complexes,² and the Ag(II) porphyrin complex.³ On the other hand, several studies have also been reported on direct electrochemical reduction using metal electrodes,⁴ electrodes with some modifications,⁵ and carbon-based electrodes.^6–9 However, a high overpotential is needed for direct electron transfer to the CO₂ molecule (eqn (1)).

CO₂ + e⁻ → CO₂˙⁻     E₀ = −1.9 V (vs. NHE) (1)

Thus, hydrogen evolution, which also occurs together with CO₂ reduction, is a competing reaction, resulting in a high production of hydrogen gas.^10,11

Conductive boron-doped diamond (BDD) is a material of great interest due to its superior electrochemical properties, such as wide potential window, low background current, chemical inertness, and mechanical durability, and it has been used for many electrochemical applications.^12–14 The wide potential window and conductive behavior of BDD electrodes are suitable for supporting CO₂ reduction in aqueous solutions and minimizing hydrogen evolution. In fact, we reported on the electrochemical reduction of CO₂ using BDD electrodes in methanol, seawater, and NaCl solutions, producing formaldehyde and formic acid with high faradaic efficiency.¹⁵

On the other hand, carbon dioxide capture and storage (CCS) technology is one of the promising technologies for reducing carbon emissions. Amine solutions, such as monoethanolamine (MEA) and diethanolamine (DEA), are used for CCS technology, since they are chemically strong CO₂ absorbers. In addition, ammonia (NH₃) solution is also known as a strong absorber and has high loading capacity for CO₂ scrubbing systems (3 times higher than MEA solutions).^16–18

Along these lines, in this present work, the electrochemical reduction of CO₂ on a BDD electrode was performed in weakly alkaline aqueous NH₃ solution, resulting in the selective production of methanol.

Experimental

Chemicals

NH₄HCO₃ was purchased from Sigma Aldrich, and other reagents were purchased from Wako Pure Chemical Industries. All reagents were used without any further purification. Ultrapure water was obtained from a Symply-Lab water system (Direct-Q UV3, Millipore).

Preparation of the BDD electrodes

BDD films were deposited onto Si (111) wafers using a microwave plasma-assisted chemical vapor deposition system (Model AX 5400, CORNES Technology Corp.). Details of the preparation are described elsewhere.¹⁴ A mixture of trimethoxyborane and acetone was used as the carbon and boron sources to achieve a B/C ratio of 1%.

Electrochemical measurements

The electrochemical measurements were conducted in two-compartment cell separated by a Nafion membrane with 100 mL of solution in each cell (Fig. S1†). BDD, Pt mesh, and Ag/AgCl were used as a working electrode, a counter electrode, and a reference electrode, respectively. BDD with an area of 4.9 cm² and the Ag/AgCl electrode were immersed on the catholyte side (1 M NH₃), and the Pt mesh electrode was on the anolyte side (0.1 M NaCl). The BDD and Pt electrodes were pretreated by ultrasonication in ultrapure water. Ultrasonication in acetone and ultrapure water was also performed after each reduction process to clean the electrodes. Before reduction started, N₂ gas was purged for 30 minutes to remove oxygen gas in the solution, followed by CO₂ gas bubbling for 2 hours with a flow rate of 200 sccm. Cyclic voltammograms were taken before and after gas bubbling with a scan rate of 100 mV s⁻¹. Electrochemical reduction of CO₂ was performed by chronoamperometry for 2 hours at potentials ranging from −1.2 V to −1.5 V (vs. Ag/AgCl) at the room temperature and the atmospheric pressure. To study the mechanism, the reduction of an aqueous ammonium bicarbonate (NH₄HCO₃) solution was performed without CO₂ bubbling. All electrochemical measurements were recorded using a potentiostat (Autolab PGSTAT204, Metrohm Autolab B.V.).

Products analysis

The liquid products were analyzed using a gas chromatography-mass spectrometry (GCMS-QP2010 Ultra, Shimadzu Corp.) by the headspace method using ethanol as the internal standard. Meanwhile, the gas products were analyzed by gas chromatography with a flame ionization detector and a thermal conductivity detector (GC-2014, Shimadzu Corp.).

Results and discussion

Fig. 1 shows cyclic voltammograms (CVs) on a BDD electrode in 1 M NH₃ aqueous solution. After nitrogen gas purging, no reduction peak was observed for potentials ranging from 0 V to −1.8 V (vs. Ag/AgCl) (Fig. 1a). After CO₂ gas was bubbled into the solution at a flow rate of 200 sccm for 2 hours, still no specific reduction peak was observed. However, a cathodic current was increased from around −1.0 V (vs. Ag/AgCl) (Fig. 1b).

Fig. 1 CVs on a BDD electrode in 1 M NH₃ aqueous solution after nitrogen gas purging ((a) solid line) and after CO₂ bubbling for 2 hours ((b) dashed line) with a scan rate of 100 mV s⁻¹.

After CO₂ saturation (2 hours bubbling) in 1 M NH₃ aqueous solution, the electrochemical reduction was performed for 2 hours at various potentials at room temperature and atmospheric pressure. The products after electrolysis at −1.3 V (vs. Ag/AgCl) were methanol (0.25 mg L⁻¹, faradaic efficiency: 24.3%), carbon monoxide (0.002 mg L⁻¹, faradaic efficiency: 0.05%), methane (0.0006 mg L⁻¹, faradaic efficiency: 0.13%), and hydrogen (0.04 mg L⁻¹, faradaic efficiency: 19.7%). Here, the dependence of the faradaic efficiency on the applied potential is summarized in Table 1. When the applied potential was −1.3 V (vs. Ag/AgCl), the efficiency of the methanol production was the highest. At the same time, the efficiencies for the production of CO and CH₄ were quite low. Thus, methanol can be selectively produced by CO₂ reduction on a BDD electrode in an aqueous NH₃ solution. On the other hand, when the applied potential was more negative, the faradaic efficiency of the hydrogen evolution increased and that of the methanol production diminished.

Table 1 The faradaic efficiencies of the products for 2 hours reduction of CO₂ on a BDD electrode in 1 M NH₃ aqueous solution at the potentials from −1.2 V to −1.5 V (vs. Ag/AgCl)

Potential (V vs. Ag/AgCl)	Faradaic efficiency (%)
	CH3OH	CO	CH4	H2
−1.2	2.61	0	0.24	0
−1.3	24.3	0.05	0.13	19.7
−1.4	15.1	0.12	0.03	25.8
−1.5	2.02	0.20	0	57.2

---

The pH was monitored during CO₂ bubbling in the 1 M NH₃ aqueous solution. The initial pH was 11.7. During CO₂ bubbling for 2 hours, the pH decreased to 7.7 (Fig. 2). In aqueous solutions, NH₃ and CO₂ react as follows (eqn (2)–(4)):¹⁶

NH₃ + CO₂ + H₂O ⇄ NH₄⁺ + HCO₃⁻   K_eq.(293) = 1.02 × 10³ (2)

NH₃ + HCO₃⁻ ⇄ NH₂CO₂⁻ + H₂O      K_eq.(293) = 3.61 (3)

NH₃ + HCO₃⁻ ⇄ CO₃²⁻ + NH₄⁺    K_eq.(293) = 1.04 × 10⁻¹ (4)

Fig. 2 pH of the aqueous ammonia solution during CO₂ gas bubbling. The measurement was conducted while stirring the solution at ∼25 °C. After 2 hours bubbling, pH decreased to 7.7.

On the other hand, according to the distribution of carbonaceous species in the aqueous solution, the pH determines which species are dominant in the solution. It is known that dissolved CO₂ (aq) is the dominant species at pH < 5, while at pH levels from 7.5 to 9, HCO₃⁻ (bicarbonate ion) is dominant, and CO₃²⁻ (carbonate ion) is dominant above pH 12. The reaction is generally denoted as follows (eqn (5) and (6)):¹⁹

CO₂ (aq) + H₂O (l) ⇄ H⁺ (aq) + HCO₃⁻ (aq)      pK₁ = 6.35 (5)

H⁺ (aq) + HCO₃⁻ (aq) ⇄ 2H⁺ (aq) + CO₃²⁻ (aq)   pK₂ = 10.33 (6)

Since the reaction of aqueous NH₃ with CO₂ rapidly produces bicarbonate (pH 7.6–8.0) after CO₂ saturation, it is assumed that bicarbonate ions are reduced to methanol in our system.

In order to confirm that bicarbonate ions are reducible species, the electrochemical reduction of a 0.1 M NH₄HCO₃ aqueous solution (pH 7.9) was performed without CO₂ bubbling. As the result, methanol was mainly produced in common with the reduction in an aqueous NH₃ solution saturated with CO₂ (Fig. 3).

Fig. 3 Faradaic efficiencies of methanol production by the 2 hours reduction of 0.1 M NH₄HCO₃ aqueous solution (pH 7.9) on a BDD electrode at the potentials from −1.2 V to −1.5 V (vs. Ag/AgCl). The faradaic efficiencies of the products for 2 hours reduction of CO₂ on a BDD electrode in 1 M NH₃ aqueous solution at the potentials from −1.2 V to −1.5 V (vs. Ag/AgCl).

In addition, we investigated into the influence of pH by adding HCl for low pH (<4) and NaOH for high pH (>10) to the NH₄HCO₃ solution. As the results, methanol was not detected in either condition at the same reduction potential of −1.3 V (vs. Ag/AgCl). These results suggest that bicarbonate ions are reducible species, not CO₂ and CO₃²⁻. That is, we propose the mechanism of methanol production in our system as follows (eqn (7)):

HCO₃⁻ + 5H₂O + 6e⁻ → CH₃OH + 7OH⁻ (7)

On the other hand, we considered the effect of NH₃ on methanol production by the electrochemical reduction of bicarbonate ions. The electrochemical reduction was performed in 0.1 M KOH solution saturated with CO₂ (pH 7.0) and in 0.1 M NaOH solution saturated with CO₂ (pH 7.2) at a potential of −1.3 V (vs. Ag/AgCl) for 2 hours on a BDD electrode, resulting in no methanol production. It can be explained by the buffering effect of the solution. During the electrochemical reduction, the hydroxide ions (OH⁻) were produced by the reduction of bicarbonate ions and also by the hydrogen evolution. Therefore, the local pH will become slightly higher. However, in the aqueous NH₃ solution, OH⁻ reacts with ammonium ions to form ammonia and water by the reaction (eqn (8)):

NH₄⁺ + OH⁻ ⇄ NH₃ + H₂O (8)

On the other hand, in the KOH and NaOH solution, the bicarbonate ions will react with OH⁻ to form carbonate ions by the following reaction (eqn (9)):

HCO₃⁻ + OH⁻ ⇄ CO₃²⁻ + H₂O (9)

As mentioned previously, the reducible species are bicarbonate ions, and carbonate ions could not be reduced to methanol. Therefore, the presence of NH₃ is important for methanol production. In addition, ammonia has the advantage of the higher loading capacity of CO₂ than other aqueous solutions (Tables S1 and S2†). However, a clear and complete mechanism will need further study.

For comparison, the electrochemical reduction of CO₂ on glassy carbon electrode was conducted at a potential of −1.3 V (vs. Ag/AgCl) for 2 hours. However, no methanol was detected, and most of the product was H₂ reaching an amount of 0.26 mg L⁻¹ (on the BDD electrode, it was 0.04 mg L⁻¹). Therefore, methanol can be selectively produced by CO₂ reduction on a BDD electrode, as long as the hydrogen evolution can be suppressed. This kind of continuous electrochemical reduction at quite high potential may cause surface corrosion of the glassy carbon electrode (emphasizing roughness, changing pores volume, etc.).^19,20 On the other hand, BDD has high durability. In order to show evidence for the durability of BDD, the surface morphology of a BDD electrode was examined by SEM after using the BDD electrode for more than 30 hours in a reduction process. The SEM images revealed no difference as compared to the as-grown BDD (Fig. 4). This is consistent with the results in our previous report.¹⁵

Fig. 4 SEM images of the BDD surface (a) before and (b) after more than 30 hours reduction.

Conclusions

We found that methanol can be selectively produced by the electrochemical reduction of CO₂ in an aqueous ammonia solution on BDD electrodes at the potential of −1.3 V (vs. Ag/AgCl) with the faradaic efficiency as high as 24.3%. The methanol production is most likely due to the electrochemical reduction of bicarbonate ions which are formed by the reaction between ammonia and CO₂. In addition, BDD electrodes show the high durability and the suppression of hydrogen evolution for the electrochemical reduction of CO₂, compared with a glassy carbon electrode. This study shows the possibility for the practical applications on reducing industrial stored-CO₂ (HCO₃⁻), which usually use amine-based solutions for CO₂ scrubbing systems.

Acknowledgements

Prastika Krisma Jiwanti acknowledges the graduate scholarship from Indonesia Endowment Fund for Education (LPDP), Republic of Indonesia.

Notes and references

1. A. Nakada, K. Koike, K. Maeda and O. Ishitani, Green Chem., 2016, 18, 139.
2. J. M. Smieja and C. P. Kubiak, Inorg. Chem., 2010, 49, 9283.
3. J. Y. Becker, B. Vainas, R. Eger and L. Kaufman, J. Chem. Soc., Chem. Commun., 1985, 1471.
4. S. Kaneco, K. Iiba, K. Ohta, T. Mizuno and A. Saji, Electrochim. Acta, 1998, 44, 573.
5. M. Ma, K. Djanashvili and W. A. Smith, Phys. Chem. Chem. Phys., 2015, 17, 20861.
6. B. R. Eggins, E. M. Brown, E. A. McNeill and J. Grimshaw, Tetrahedron Lett., 1988, 29, 945.
7. P. A. Christensen, A. Hamnett, A. V. G. Muir and N. A. Freeman, J. Electroanal. Chem. Interfacial Electrochem., 1990, 288, 197.
8. K. Hara, A. Kudo and T. Sakata, J. Electroanal. Chem., 1997, 421, 1.
9. N. Yang, S. R. Waldvogel and X. Jiang, ACS Appl. Mater. Interfaces, 2015 DOI:10.1021/acsami.5b09825.
10. T. Yamamoto, D. A. Tryk, K. Hashimoto, A. Fujishima and M. Okawa, J. Electrochem. Soc., 2000, 147, 3393.
11. R. M. Hernández, J. Márquez, O. P. Márquez, M. Choy, C. Ovalles, J. J. Garcia and B. Scharifker, J. Electrochem. Soc., 1999, 146, 4131.
12. W. T. Wahyuni, T. A. Ivandini, P. K. Jiwanti, E. Saepudin, J. Gunlazuardi and Y. Einaga, Electrochemistry, 2015, 83, 357.
13. Y. Einaga, J. Appl. Electrochem., 2010, 40, 1807.
14. T. Yano, D. A. Tryk, K. Hashimoto and A. Fujishima, J. Electrochem. Soc., 1998, 145, 1870.
15. K. Nakata, T. Ozaki, C. Terashima, A. Fujishima and Y. Einaga, Angew. Chem., Int. Ed., 2014, 53, 871.
16. F. Mani, M. Peruzzini and P. Stoppioni, Green Chem., 2006, 8, 995.
17. S. Y. Park, K. B. Yi, C. H. Ko, J.-H. Park, J.-N. Kim and W. H. Hong, Energy Fuels, 2010, 24, 3704.
18. H. Bai and A. C. Yeh, Ind. Eng. Chem. Res., 1997, 36, 2490.
19. H. Zhong, K. Fujii, Y. Nakano and F. Jin, J. Phys. Chem. C, 2015, 119, 55.
20. G. M. Swain, J. Electrochem. Soc., 1994, 141, 3382.

---

Footnote

† Electronic supplementary information (ESI) available: The apparatus for electrochemical reduction, pH measurements, CO₂ concentration, and SEM images. See DOI: 10.1039/c6ra20466j

---