Copper-nickel-modified Boron-doped Diamond Electrode for CO2 Electrochemical Reduction Application: A Preliminary Study

CO2 electrochemical reduction (CO2ER) activity is known to be influenced by electrode materials. In this study, we report the fabrication of a copper-nickel-modified boron-doped diamond (CuNi-BDD) electrode using wet chemical seeding and electrodeposition. Annealing was performed to improve the stability of the modified electrode during electrolysis. Characterization of the modified BDD electrodes shows successful deposition without damage to the surface of the BDD support material. CO2ER was conducted with the CuNi-BDD electrode, which produces various important products including methanol, formic acid, CO, and CH4. Additionally, a different applied potential affected the product distribution. CO2ER was also conducted on the surfaces of Cu-BDD and Ni-BDD electrodes for comparison.


Introduction
In the last few decades, the concentration of CO 2 gas in the atmosphere has increased tremendously. This cheap and highly abundant gas, which is a known greenhouse gas, can be converted to useful and valuable chemicals or converted back to fuel [1][2][3]. Many attempts to reduce CO 2 have been reported, including electrochemistry [4] and photoelectrochemistry [5], which have been widely developed in recent years. However, CO 2 ER requires a high overpotential, which is a drawback of many reported methods. The applied potential remained much higher than the standard reduction potential of CO 2. Therefore, an important approach to overcoming this problem is developing a suitable catalytic system. For this purpose, a suitable electrode material is required.
Previous studies on CO 2 ER have been reported mostly on copper electrodes, resulting in various product distributions, including hydrocarbons and oxygenated species [6][7][8]. Other studies reported that a copper-modified nickel electrode showed high catalytic activity with hydrocarbon formation [9]. Additionally, computational experiments on a copper-nickel alloy with a ratio of 3:1 for CO 2 reduction showed an increase in the CO 2 reduc-tion electrocatalytic activity to form methane due to overpotential suppression [10].
Meanwhile, recent studies have focused on CO 2 ER on the surface of a BDD electrode [11,12]. The BDD electrode is known to have a wide potential window that is considerably useful for suppressing the production of H 2 gas as a CO 2 ER competitor. Additionally, the BDD electrode has high chemical and mechanical stability, making it suitable for practical application. CO 2 ER on a bare BDD electrode in aqueous alkali metal solution has been known to produce HCOOH with more than 90% faradaic efficiency. However, the high overpotential (>−2 V) needed for the system remains a drawback. Therefore, modification of the BDD electrode surface is suggested to enhance the catalytic activity.
Metal modification on the BDD electrode is a simple technique for increasing the catalytic effect of the electrode. The use of a metal-modified BDD electrode for CO 2 ER has been reported in several publications [13][14][15]. Such systems successfully decreased the overpotential or produced carbon compounds (C 2 /C 3 ) with a longer chain, which were barely observed on the BDD electrode. Therefore, CO 2 ER on other metals deposited on the BDD electrode is worth further investiga-tion. In this study, the BDD electrode surface was mod fied with Cu and Ni particles. Cu is a well catalyst for CO 2 ER, whereas Ni is an active metal for hydrogen production. A strategy for combining two metals with high activity but different behavior is d sired to provide an overall high activity for the CO reaction to produce useful chemicals. The detailed fa rication of the modified electrodes and a preliminary report of their application to CO 2 ER is presented.

Materials and Methods
Chemicals. CuSO 4 .5H 2 O (>99.5%), Ni(NO NaCl, and Na 2 SO 4 were purchased from Wako Pure Chemical Industries. NaBH 4 was purchased from Sigma Aldrich. All reagents were used without further purific tion. Ultrapure water was obtained from a Simply water system (Direct-Q UV3, Millipore).
Electrode Fabrication. BDD Films were deposited on Si (111) wafers using microwave plasma ical vapor deposition (Model AX 5400, Cornes Tec nology Corp.) [16]. Modification of the BDD electrode surface with metal particles was performed by wet chemical seeding [17] and followed by electrochemical method. Wet chemical seeding was performed by dro ping 100 μL of 1 M NaBH 4 in 0.1 M NaOH solution on the BDD surface and continuing with a 400 μL metal precursor (1 mM Ni(NO 3 ) 2 or CuSO 4 BDD or Cu-BDD. To prepare CuNi-BDD, we obtained a metal precursor with 3:1 v/v mixture of 1 mM Ni(NO 3 ) 2 and 1 mM CuSO 4 . This electrode was dried at room temperature and pressure for 24 h, rinsed with water, and dried under N 2 gas. This process was repea

Copper-nickel-modified Boron-doped Diamond Electrode
December tion. In this study, the BDD electrode surface was modified with Cu and Ni particles. Cu is a well-known metal ER, whereas Ni is an active metal for ction. A strategy for combining two metals with high activity but different behavior is desired to provide an overall high activity for the CO 2 ER reaction to produce useful chemicals. The detailed fabrication of the modified electrodes and a preliminary ER is presented.
O (>99.5%), Ni(NO 3 ) 2 , NaOH, were purchased from Wako Pure was purchased from Sigma Aldrich. All reagents were used without further purification. Ultrapure water was obtained from a Simply-Lab Q UV3, Millipore).
BDD Films were deposited on ve plasma-assisted chemical vapor deposition (Model AX 5400, Cornes Tech-. Modification of the BDD electrode surface with metal particles was performed by wet and followed by electrochemical Wet chemical seeding was performed by dropin 0.1 M NaOH solution on the BDD surface and continuing with a 400 μL metal 4 ) to prepare Ni-BDD, we obtained a metal precursor with 3:1 v/v mixture of 1 mM trode was dried at room temperature and pressure for 24 h, rinsed with gas. This process was repeat-ed three times and followed by electrodeposition. Electrodeposition was performed in each 1 mM metal precursor at a potential of −1 by rapid thermal annealing (RTA) at 700 in N 2 atmosphere. The electrode was repeatedly activa ed by cyclic voltammetry thereafter and characterized using scanning electron microscopy X-ray spectroscopy (SEM-EDS), Raman spectroscopy, and X-ray photoelectron spectroscopy (XPS).

CO 2 ER Method and Product Analysis.
conducted in two cells separated by a Nafion me brane. The catholyte and anolyte were 0.1 M NaCl and 0.1 M Na 2 SO 4 , respectively. The trode, a Pt spiral, and Ag/AgCl electrodes were used as working (WE), counter (CE), and reference (RE) ele trodes, respectively. All potentials were measured against Ag/AgCl unless otherwise stated. was also equipped with a gas trap to carry gas products from the cell ( Figure 1). N 2 gas was purged through the catholyte to remove dissolved O gas bubbling at a flow rate of 100 sccm. was performed for 1 h. Subsequently, the collected gas products were analyzed by gas chromatography with a flame ionization detector and thermal conductivity d tector (GC-2014, Shimadzu Corp.). The liquid products were analyzed using gas chromatography trometry (GCMS-QP2010 Ultra, Shimadzu Corp.) using auto-injection headspace method (5 mL sample in 20 mL volume vial) with the selected ion monitoring mode and employing high-performance liquid chromato raphy with an electroconductivity detector (Prominence, Shimadzu Corp.). ed three times and followed by electrodeposition. Electrodeposition was performed in each 1 mM metal −1.2 V for 15 min, followed by rapid thermal annealing (RTA) at 700°C for 5 min atmosphere. The electrode was repeatedly activated by cyclic voltammetry thereafter and characterized using scanning electron microscopy-energy dispersive EDS), Raman spectroscopy, ray photoelectron spectroscopy (XPS).
ER Method and Product Analysis. CO 2 ER was conducted in two cells separated by a Nafion membrane. The catholyte and anolyte were 0.1 M NaCl and , respectively. The modified BDD electrode, a Pt spiral, and Ag/AgCl electrodes were used as working (WE), counter (CE), and reference (RE) electrodes, respectively. All potentials were measured against Ag/AgCl unless otherwise stated. This system s trap to carry gas products gas was purged through the catholyte to remove dissolved O 2 gas, followed by CO 2 gas bubbling at a flow rate of 100 sccm. Electrolysis was performed for 1 h. Subsequently, the collected gas were analyzed by gas chromatography with a flame ionization detector and thermal conductivity de-2014, Shimadzu Corp.). The liquid products were analyzed using gas chromatography-mass spec-QP2010 Ultra, Shimadzu Corp.) using njection headspace method (5 mL sample in 20 mL volume vial) with the selected ion monitoring mode performance liquid chromatography with an electroconductivity detector (Prominence,

Results and Discussion
Preparation and Characterization of Modified BDD Electrodes. First, wet chemical seeding was applied to modify the surface of the BDD electrode to stabilize the metal particles because the common electrodeposition method might lead to instability and detachment of me al particles. In this study, NaBH 4 was used as a reducing agent. NaBH 4 (1 M) in 0.1 M NaOH solution was dropped on the surface of the BDD electrode followe by the metal precursor. Thus, reduction occurred on the surface of the electrode, with strong adsorption to the electrode surface. Electrochemical deposition was then performed on the electrode modified by wet seeding, aiming to increase the deposited metals. However, up to this stage, the fabricated electrode could still be m chanically unstable. Therefore, RTA was conducted at 700°C for 5 min to stabilize the particles on the BDD surface. However, RTA passivates the modified ele trode, thereby requiring an additional electrochemical polishing treatment. In this work, several cycles of c clic voltammetry were conducted as an electrochemical polishing technique.
An SEM image of the CuNi-BDD electrode shows that the Cu and Ni particles are successfully de BDD electrode surface (Figure 2), while EDS indicates that the contents of Cu and Ni particles are 0.11% and 0.14%, respectively (C 99.5% and O 0.59%). XPS was

Preparation and Characterization of Modified BDD
First, wet chemical seeding was applied to modify the surface of the BDD electrode to stabilize the particles because the common electrodeposition method might lead to instability and detachment of metwas used as a reducing (1 M) in 0.1 M NaOH solution was dropped on the surface of the BDD electrode followed by the metal precursor. Thus, reduction occurred on the surface of the electrode, with strong adsorption to the electrode surface. Electrochemical deposition was then performed on the electrode modified by wet seeding, tals. However, up to this stage, the fabricated electrode could still be mechanically unstable. Therefore, RTA was conducted at°C for 5 min to stabilize the particles on the BDD surface. However, RTA passivates the modified elecg an additional electrochemical polishing treatment. In this work, several cycles of cyclic voltammetry were conducted as an electrochemical BDD electrode shows that the Cu and Ni particles are successfully deposited on the BDD electrode surface (Figure 2), while EDS indicates that the contents of Cu and Ni particles are 0.11% and 0.14%, respectively (C 99.5% and O 0.59%). XPS was also conducted (Figure 3). XPS spectra of the modified BDD electrode show element BDD surface. The peak at 933.9 eV represents Cu 2p while that at 856 eV represents Ni 2p ence of metal particles on the modified BDD electrode surface was confirmed.

BDD Electrode: Wide Scan (A), and Narrow Scan of Cu 2p
December 2019  Vol. 23  No. 4 also conducted ( Figure 3). XPS spectra of the modified BDD electrode show elemental analysis of the modified BDD surface. The peak at 933.9 eV represents Cu 2p 3/2 , while that at 856 eV represents Ni 2p 3/2 . Thus, the presence of metal particles on the modified BDD electrode of CuNi-BDD Electrode

Cu 2p 3/2 (B) and Ni 2p 3/2 (C)
Additionally, Raman spectroscopy was performed to examine the stability of the BDD electrode after all the deposition treatments ( Figure 4). The high temperature of RTA and several deposition steps was previously thought to damage the sp 3 carbon bonding of BDD. However, the modified BDD electrode retained its cha acteristic properties, where sp 3 carbon bonding at 1332 cm −1 and the B-B bond peak at 1220 cm lyzed, and no sp 2 carbon bonding at approximately 1500 cm −1 was detected [18]. Therefore, the fabrication method used in this study did not damage the BDD ele trode as a support for Cu and Ni deposition, confirming the high compatibility of this material plication.

Copper-nickel-modified Boron-doped Diamond Electrode
December Additionally, Raman spectroscopy was performed to of the BDD electrode after all the deposition treatments (Figure 4). The high temperature of RTA and several deposition steps was previously carbon bonding of BDD. However, the modified BDD electrode retained its charcarbon bonding at 1332 B bond peak at 1220 cm −1 were anacarbon bonding at approximately 1500 . Therefore, the fabrication method used in this study did not damage the BDD electrode as a support for Cu and Ni deposition, confirming the high compatibility of this material for practical ap- 2.16 CO 2 ER application. CO 2 ER was performed for 1 h at three different potentials: −1.2, −1.5, and −1.7 V. The total current density was evaluated for each electrode at each potential. Comparison with an electrode fabricated without RTA is also presented ( Table 1) show that BDD electrodes modified by RTA show a relatively lower total current density than those without RTA treatment. This condition could be explained by imperfect activation through cyclic voltammetry, which results in incomplete removal that is known to exist after annealing this condition, the stability of the modified BDD ele trode could be improved. Thus, enhancement of activation treatment is recommended for future work to overcome this issue.
CO 2 ER on the Ni-BDD electrode shows the best pe formance at a potential of −1.2 V, which may be due to lower H 2 evolution providing a pathway for CO form simple C 1 molecules, unlike Cu known to be active for CO CO 2 ER products at all the potentials applied in this work. However, the higher potential required to produce ethanol in this study (>−1.5 V), compared with the pr vious report [14], might be due to the modified BDD electrode not being fully activated or a different detailed system being used in this study.
However, as shown in Figure 5A, the CO on the CuNi-BDD electrode include methano thane. Additionally, Figure 5B shows the trend of the partial current density rather than the faradaic efficiency because the total current density of CO siderably for each electrode. The trend of the partial current density is similar to that for the product content. The CO 2 ER on the CuNi-BDD electrode, which co bines the behavior of two different metals with its own, shows good performance at known that CO 2 ER on Ni metal produces a large amount of H 2 gas [19] with a very low CO cy. Ni is also a well-known metal catalyst that can strongly adsorb CO, which is known to be a CO intermediate. However, Cu is well various products via CO 2 ER, including hydrocarbons and oxygenated species [20]. Cu also has a lower CO adsorption strength than Ni metal. Based on these r sults, modification of the BDD surface with bimetallic Ni and Cu, as a combination of metals with different behavior in CO 2 ER, is expected performance of CO 2 ER [21]. However, the performance of CuNi-BDD in CO 2 ER is between those of Cu and Ni-BDD. The presence of Cu particles may d crease the production of hydrogen, which is the main product on Ni metal. Nevertheless, the behavior and product trends of CO 2 ER on CuNi cantly different from those of the monometal electrode. Thus, adjusting the metal composition can ER was performed for 1 h at −1.2, −1.5, and −1.7 V. The total current density was evaluated for each electrode at each potential. Comparison with an electrode fabricated without RTA is also presented ( Table 1). The results show that BDD electrodes modified by RTA show a relatively lower total current density than those without RTA treatment. This condition could be explained by imperfect activation through cyclic voltammetry, which results in incomplete removal of the passivation layer that is known to exist after annealing [17]. Regardless of this condition, the stability of the modified BDD electrode could be improved. Thus, enhancement of the activation treatment is recommended for future work to BDD electrode shows the best per-−1.2 V, which may be due to evolution providing a pathway for CO 2 ER to ules, unlike Cu-BDD, which is known to be active for CO 2 ER, thereby resulting in ER products at all the potentials applied in this work. However, the higher potential required to produce −1.5 V), compared with the pre-, might be due to the modified BDD electrode not being fully activated or a different detailed system being used in this study.
However, as shown in Figure 5A, the CO 2 ER products BDD electrode include methanol and methane. Additionally, Figure 5B shows the trend of the partial current density rather than the faradaic efficiency because the total current density of CO 2 ER differs considerably for each electrode. The trend of the partial r to that for the product content. BDD electrode, which combines the behavior of two different metals with its own, shows good performance at −1.5 V and −1.7 V. It is ER on Ni metal produces a large with a very low CO 2 ER efficienknown metal catalyst that can strongly adsorb CO, which is known to be a CO 2 ER ate. However, Cu is well-known for producing ER, including hydrocarbons . Cu also has a lower CO adsorption strength than Ni metal. Based on these results, modification of the BDD surface with bimetallic Ni and Cu, as a combination of metals with different ER, is expected to improve the catalytic . However, the performance ER is between those of Cu-BDD BDD. The presence of Cu particles may decrease the production of hydrogen, which is the main product on Ni metal. Nevertheless, the behavior and ER on CuNi-BDD are not significantly different from those of the monometal-modified electrode. Thus, adjusting the metal composition can optimize the catalytic effects, including the reaction rate and even the selectivity toward CO 2 ER.
Regardless of the limitations of these results, an attempt to modify the surface of a BDD electrode with metal particles, resulting in the possible generation of products other than HCOOH, has been shown in this report. For reference, previous work on CO 2 ER on the surface of a bare BDD electrode in aqueous alkali metal solution produced mainly HCOOH [22]. Thus, methanol and methane production on the CuNi-BDD electrode is an important finding and worth further study to increase the amount of CO 2 ER products, selectivity, and efficiency.

Conclusion
In summary, a CuNi-BDD electrode was successfully fabricated. The modified BDD electrode shows activity toward CO 2 ER. Products other than HCOOH, the main CO 2 ER product that used a bare BDD electrode in aqueous alkali metal solution, could be produced, including methanol and methane. However, improvement is required to achieve optimum activation of the annealed-modified electrode and increase the amount of CO 2 ER products and efficiency. Finally, research on CO 2 ER on the metal-modified BDD electrode is ongoing and worth further exploration as a method of overcoming the high overpotential of CO 2 ER on a bare BDD electrode.