Enhancement of the Catalytic Effect on the Electrochemical Conversion of CO₂ to Formic Acid Using MXene (Ti₃C₂T_x)-Modified Boron-Doped Diamond Electrode

Abstract

The rising concentration of carbon dioxide (CO₂) as one of the greenhouse gases in the atmosphere is a major source of worry. Electrochemical reduction of CO₂ is one of many ways to convert CO₂ gas into usable compounds. An electrochemical technique was applied in this study to reduce CO₂ using a boron-doped diamond (BDD) working electrode modified with MXene (Ti₃C₂T_x) material to improve electrode performance. MXene concentrations of 0.5 mg/mL (MXene-BDD 0.5), 1.0 mg/mL (MXene-BDD 1.0), and 2.0 mg/mL (MXene-BDD 2.0) were drop-casted onto the BDD surface. MXene was effectively deposited on top of the BDD surface, with Ti weight loads of 0.12%, 4.06%, and 7.14% on MXene-BDD 0.5, MXene-BDD 1.0, and MXene-BDD 2.0, respectively. The modified working electrode was employed for CO₂ electroreduction with optimal CO₂ gas aeration. The existence of the MXene substance in BDD reduced the electroreduction overpotential of CO₂. For the final result, we found that the MXene-BDD 2.0 electrode effectively generated the most formic acid product with a maximum reduction potential as low as −1.3 V (vs. Ag/AgCl).

1. Introduction

The causes of extreme usage of fossil fuel in transportation and industry has been well reported, due to high CO₂ gas emissions and the shortage of fossil fuel [1,2]. According to recent research, the use of fossil fuel energy contributes to 96.5% of CO₂ emissions. Heavy CO₂ emissions led to global warming [3]. Since the industrial revolution, human activity has increased by 30%, and it is expected to continue growing year by year [4]. Indonesia generated 617.51 million tons of CO₂ gas emissions in 2019 [5]. As a result, an alternative energy source that possibly keeps CO₂ levels in the atmosphere stable is necessary. There have been attempts to overcome this problem through carbon storage; the other method is CO₂ conversion, a process that creates chemical stocks or further converts it back to fuels. As a result, these substances may be put to good use and may improve its value [6]. As for CO₂ conversion, researchers have been attracted to electrochemical techniques because of their mild working conditions, adjustable reaction conditions and rates, reusable catalysts and electrolytes, simple equipment, diverse product selectivity, and higher conversion efficiency [7,8]. The CO₂ molecule may be reduced through electrochemical reduction into CO₂^•−, which can then be used to generate a number of different compounds [9,10,11].

Meanwhile, the boron-doped diamond (BDD) electrode is a carbon-based electrode [12], which has excellent properties, such as high physical and chemical stability, wide potential window, low background current, and biocompatibility [13,14,15,16]. This wide potential window suppresses the hydrogen evolution reaction that is widely known to be the competitor for CO₂ electrochemical reactions, especially in aqueous solutions. The Einaga group has been working on CO₂ reduction for years using the polycrystalline BDD electrode, and formic acid (HCOOH) is known to be the main product on the bare BDD condition, with its potential around −2.2 V [17,18]. This high reduction potential can be suppressed by modifying the BDD surface with metal nanoparticles or higher catalytic materials. As an example, the modification of the BDD surface with nanoparticles such as Cu or Pd can suppress the reduction potential, as well as the production of higher carbon-containing compound products.

On the other hand, Ti₃C₂T_x serves as one of the members of MXene, a class of 2D morphological carbide metal transition materials with nanoscale thicknesses (nanosheets) [19]. Tx in the Ti₃C₂T_x structure stands for the surface terminal functional groups, i.e., –OH, –O–, –F, and/or –Cl. This material can be modified by adding specific functional groups. This material’s further important attributes include excellent conductivity, mechanical strength, and flexibility [20]. MXene is studied computationally and prospected for CO₂ reduction reaction [21]. However, experimental studies are required to realize the potential of MXene. In our previous study, we found that modifying Ti₃C₂T_x on a bare BDD electrode might increase the catalytic effect and potentially reduce the CO₂ overpotential reduction required on a bare BDD, which can then be utilized for an actual CO₂ reduction application [22]. In this report, the electrochemical reduction product of CO₂ on the surface of a Ti₃C₂T_x (MXene)-modified BDD electrode with mainly HCOOH will be investigated. The CO₂ reduction to HCOOH on the surface of the MXene-modified BDD electrode is predicted to be greater than that on the bare BDD electrode.

2. Materials and Methods

2.1. Materials

The materials used in this research were isopropanol, KOH (≥85%), KCl (99.0–100.5%), HCl (37%), HNO₃ (≥90%), H₂SO₄ (95–98%), HCOOH (≥96%), and a Nafion membrane from Sigma Aldrich. A microwave plasma-assisted chemical vapor deposition (MP-CVD, Model AX-5400, CORNES Technology Corp., Tokyo, Japan) was used to produce 1% (B/C) boron-doped diamond on the surface of a silicon wafer (111) for 6 h. All chemicals were used without additional purification.

2.2. Preparation and Characterization of the MXene-BDD Electrode

MXene was prepared, based on our previous work, using LiF–HCl treatment [23]. The as-prepared MXene was then characterized with scanning electron microscopy (SEM, Hitachi SU3500, Japan), X-ray diffraction (XRD, Bruker D8 Advance, Germany), and Raman spectroscopy (HORIBA HR Evolution Raman, Japan). To prepare the modified MXene-BDD electrode, first, the MXene solution was prepared at concentrations of 0.5 mg/mL (MXene-BDD 0.5), 1.0 mg/mL (MXene-BDD 1.0), and 2.0 mg/mL (MXene-BDD 2.0) in ultrapure water. Further, 20 µL of each solution was drop-casted onto the surface of the BDD electrode and dried at room temperature. After that, the modified electrodes were characterized with SEM and were employed for CO₂ electrochemical reduction.

2.3. CO₂ Electrochemical Reduction Process

The CO₂ electrochemical reduction processes were performed in a two-compartment, closed electrochemical cell, with the cathode compartment carrying 0.5 M KCl electrolyte and the anode compartment carrying 0.5 M KOH electrolyte. The electrolyte used in this work was based on our previous work [24]. The use of the KOH aqueous solution might suppress the overpotential in the anolyte side, as reported in a previous work [25]. The Nafion membrane, an ion exchange membrane, separated the two compartments. This cell used a three-electrode system, with Pt as the counter electrode, Ag/AgCl as the reference electrode, and the modified and unmodified BDD electrodes with a geometric area of 0.754 cm² as the working electrodes. Prior to each reduction process, N₂ and CO₂ gas purging was carried out for 15 min, and the CO₂ electrochemical reduction was performed for 1 h. All electrochemical data were recorded using Emstat3+ blue PalmSens. High performance liquid chromatography (HPLC, Agilent 1260, Japan) was used to evaluate the product of the CO₂ reduction.

3. Results

3.1. MXene-BDD Electrode Preparation and Characterization

SEM was used to study the surface morphology of the as-prepared MXene. It can be seen that the MXene sheets had a layered structure, showing the nature of two-dimensional nanomaterials (Figure 1a,b). The XRD result, depicted in Figure 1c, showed the main peak of Ti₃C₂T_x MXene at 2θ of around 6°, corresponding to the (002) plane. Moreover, the Raman spectra of Ti₃C₂T_x MXene demonstrated several important regions that contained information about its local structure. The Raman shift at 150 cm⁻¹ to 250 cm⁻¹ corresponded to the flake region, a combination of Ti, C, and T_x surface group (–F, –O–, and –OH) vibrations, while the bands at 225 cm⁻¹ to 500 cm⁻¹ were solely attributed to the T_x vibrations. In addition, the carbon region, located at 550 cm⁻¹ to 800 cm⁻¹, was ascribed to the in-plane and out-of-plane vibrations of the carbon layer [26]. Two prominent bands were observed at 1375 cm⁻¹ and 1575 cm⁻¹, assigned to the D band (sp² carbon) and G band (perturbed sp² carbon), respectively [23].

The BDD (B/C 1%) was fabricated on a silicon wafer using MP-CVD technology, with carbon and boron precursors from methane and B(OCH₃)₃. Prior to modification, BDD was pre-treated to clear the BDD electrode of impurities of elements or other compounds on its surface. This pretreatment was carried out electrochemically by performing a CV for 40 cycles in 0.1 M H₂SO₄ solution, with a potential range of −2.5 V to +2.5 V (vs. Ag/AgCl) and with a scan rate of 1 V/s. The modified BDD electrode was created by depositing MXene (Ti₃C₂T_x) on the BDD surface using the drop-cast technique. MXene concentrations ranged from 0.5 mg/mL, to 1.0 mg/mL, and to 2.0 mg/mL; as much as 20 µL of each variation was deposited on the BDD surface and left to dry at room temperature.

Cyclic voltammetry (CV) was performed in a one-compartment cell with three electrodes: Ag/AgCl electrodes as reference electrodes, Pt spiral electrode as a counter electrode, and BDD electrodes modified with MXene compounds as working electrodes (MXene-BDD). CV was performed in 0.1 M H₂SO₄ solution at a scan rate of 100 mV/s within a potential range of −1.0 V to +1.5 V (vs. Ag/AgCl). As shown in Figure 2, the CV results of the modified BDD electrode were compared to those of the unmodified BDD electrode (bare BDD). The cathodic current started to decrease to a more negative current value at a potential of −0.4 V (vs. Ag/AgCl) when the concentration of the MXene compound deposited on the BDD surface increased, indicating the presence of the MXene compound. Moreover, the oxidation peak at a potential of around −0.5 V was observed. The linearity test of the oxidation peak at a potential of −0.5 V showed the value of R² = 0.9925, which is considered a good linearity value. Finally, these reversible redox natures (at a potential of −1.0 V to −0.1 V) belonged to the protonation/deprotonation of the MXene surface by the changes in the Ti redox state [27].

Figure 3 shows the SEM images of the BDD surface before and after MXene was deposited. From this SEM image, we can see that the amount of MXene on the surface of BDD improved as the concentration of MXene increased. In the MXene-BDD 0.5, some of the BDD grain was still exposed, while on MXene-BDD 2.0, MXene was covering the entire surface of BDD. Further analysis was performed using EDX to identify the composition of the components distributed throughout the surface of the modified BDD.

Table 1. Elemental analysis of various concentrations of MXene modified on the surface of the BDD electrode by EDX characterization.

Electrode	% Weight Load
	C	O	Ti
Bare BDD	99.77	0.23	0
MXene-BDD 0.5	94.40	5.47	0.12
MXene-BDD 1.0	74.84	17.10	4.06
MXene-BDD 2.0	74.97	17.89	7.14

shows the composition of the elements present on the MXene-BDD surface. This table also confirmed the presence of MXene on the surface of BDD, where the element of Ti increased as the concentration of MXene increased.

3.2. Electrochemical Reduction of CO₂

3.2.1. Determining the Optimal CO₂ Bubbling Time

The CO₂ gas was aerated and dissolved into the 0.5 M KCl electrolyte. CO₂ gas contained in a solution largely consisting of water can experience carbonic acid equilibrium, notably in Equations (1)–(3), where carbonic acid can be degraded to create H⁺ ions and HCO₃⁻ species, resulting in a pH decrease. Because of this pH decrease, the optimal CO₂ aeration period can be determined using the pH parameters of the electrolyte solution.

CO₂ (aq) + H₂O (l) ⇄ H₂CO₃ (aq)  pKa = 2.77

(1)

H₂CO₃ (aq) ⇄ H+ (aq) + HCO₃⁻ (aq)  pKa = 6.35

(2)

HCO₃⁻ (aq) ⇄ H+ (aq) + CO₃²⁻ (aq)  pKa = 10.33

(3)

The pH parameter indicates which carbon species are more abundant in the solution, based on the distribution of carbon species in aqueous solutions. Dissolved CO₂ is the most common species at a pH less than 5, HCO₃⁻ (bicarbonate ion) is the most abundant species at a pH of 7.5 to 9, and CO₃²⁻ (carbonate ion) is the most abundant species at a pH greater than 12. As a result, the electrolyte solution employed in this study had the most dissolved CO₂ species, leading to a reduction in dissolved CO₂ [27].

Figure 4A shows the results of pH measurements, which showed that the 0.5 M KCl electrolyte had an initial pH of 6.6. The KCl electrolyte was then aerated for 10 min with N₂ gas to remove dissolved oxygen gas from the solution. The solution’s pH increased to 6.9 as a result of the N₂ gas aeration. The solution was then aerated for 5 min with CO₂ gas, which resulted in a pH decrease to 3.9, due to an increase in the amount of H⁺ ions, causing the pH of the solution to decrease. Meanwhile, after 10 and 15 min of aeration, the pH decreased, but the solution’s pH remained constant at 3.7 as the aeration time increased. This was due to the fact that the CO₂ gas that was converted into dissolved CO₂ was already saturated when it was aerated into the KCl electrolyte solution, and therefore, the pH of the electrolyte solution would remain stable at 3.7. According to the equilibrium reactions from Equations (1)–(3), the higher the concentration of CO₂ dissolved in the solution, the higher the concentration of H+ generated, and therefore, the pH value decreased. In addition, an electrochemical investigation was conducted with linear sweep voltammetry (LSV) to identify the optimal CO₂ aeration time by monitoring the peak reduction current of the dissolved CO₂ species. LSV was performed at a scan rate of 100 mV/s over a potential range of 0 V to −1.7 V (vs. Ag/AgCl). The LSV in Figure 4B showed no peaks after 10 min of N₂ gas aeration. Because the purpose of N₂ gas aeration is to remove dissolved oxygen, this indicates that the electrolyte was cleared of dissolved oxygen. After aerating CO₂ gas for 10 and 15 min, it showed a reduction current peak at roughly −1.5 V (vs. Ag/AgCl), suggesting the existence of the dissolved CO₂ species in the electrolyte. There was no reduction in the present CO₂ peak after 5 min of aeration. This was possible, since the amount of dissolved CO₂ was still low and could not cover the surface of the working electrode. The largest shift in CO₂ reduction current occurred after 15 min of aeration, compared to 5 min of aeration, and was only slightly different from 10 min of aeration. This indicated that 15 min of CO₂ aeration generated the largest quantity of CO₂ dissolved in the 0.5 M KCl electrolyte. Following that, 10 min of N₂ aeration and 15 min of CO₂ aeration were employed to begin CO₂ electroreduction applications.

3.2.2. Preliminary Study of the Electrochemical Reduction of CO₂

The first study of CO₂ electroreduction through LSV was carried out by comparing the reduction activity of each variant of the working electrode, including bare BDD and modified BDD electrodes. The CO₂ reduction activity experiment was conducted using the LSV across a potential range of 0 V to −1.7 V, with a scan rate of 100 mV/s after N₂ and CO₂ gas aerations.

The reduction activities of the bare BDD and the modified BDD electrodes with varied MXene concentrations are shown in Figure 5. There was a reduction signal at roughly −1.5 V for the bare BDD electrode, showing that there was reduction activity of dissolved CO₂ to its reduced products, which could not be seen at the LSV once nitrogen gas was added. Furthermore, the presence of CO₂ gas caused a shift in hydrogen evolution activity into a more negative potential, since CO₂ reduction activity released reduced CO₂ products, which covered the surface of the BDD, lowering hydrogen evolution activity [28]. There was a slight reduction signal of CO₂ at a potential of about −1.4 V for MXene-BDD 0.5, a reduction signal from CO₂ at a potential of around −1.2 V for MXene-BDD 1.0, and a reduction signal for CO₂ gas at a potential of around −1.0 V for MXene-BDD 2.0. Based on these results, all MXene-BDD electrodes at different concentrations showed that as the deposited MXene concentration increased, the CO₂ reduction activity shifted to a positive potential; it was suggested that the overpotential that occurred in the CO₂ electroreduction was lower. Furthermore, the intensity of the signals rose with increasing concentrations of deposited MXene, indicating an increase in the quantity of CO₂ reduction products formed at the surface of the BDD electrode.

3.3. CO₂ Electrochemical Reduction Products

All products of CO₂ electrochemical reduction in the cathode side were collected and analyzed using an HPLC instrument. In this report, we focused on the HCOOH products, given that HCOOH is a common byproduct of CO₂ electroreduction on the BDD-based electrode. To explain the mechanism of CO₂ reduction on the MXene-BDD electrode, knowing the active reduced species in the solution is important. The pH of the electrolyte solution after 10 min of CO₂ aeration was 3.7. The pH parameter indicates which carbon species are more abundant in the solution, based on the distribution of carbon species in aqueous solutions [29]. Dissolved CO₂ was the most common species at a pH less than 5, HCO₃⁻ (bicarbonate ion) was the most abundant species at a pH of 7.5 to 9, and CO₃²⁻ (carbonate ion) was the most abundant species at a pH greater than 12. As a result, the electrolyte solution employed in this study had the most dissolved CO₂ species, leading to a reduction in dissolved CO₂. The HCOOH can be produced using three methods. The first pathway, the CO₂^•− anion radical, is created initially by giving one electron to CO₂, where the oxygen in the CO₂^•− (i) anion radical is bound to the electrode surface. In this case, protonation will occur on the carbon atom to generate the HCOO (ii) intermediate, followed by the second electron transfer and the protonation step to produce the product HCOOH. The second method is not similar to the first, according to theoretical calculations. After the production of HCOO• (ii) intermediates, •OCHO (iii) intermediates can be generated by giving up one electron. Following that, HCOOH is produced by the protonation of •OCHO (iii). The third pathway is when the carbon in CO₂^•− attaches to the surface of the electrode (iv); the CO₂^•− intermediate may be reduced by the protonation of its own oxygen atom, resulting in the creation of •COOH (v). This intermediate can be converted to HCOOH or can lose H₂O to generate CO [28]. The reaction’s equation is the following reaction: CO₂ + 2H⁺ + 2e⁻ ⇄ HCOOH [30].

3.3.1. The Effect of MXene Concentration Variation

Using each of the bare BDD and MXene-BDD electrodes at varied concentrations, the quantification of CO₂ electroreduction products at a potential of −1.7 V (vs. Ag/AgCl) was compared to the data given in Figure 6. As the variety of MXene deposited on the surface of the BDD electrode increased, so did the quantity of HCOOH products produced by the MXene-BDD electrode. The MXene-BDD 2.0 electrode produced the greatest HCOOH concentration, with a HCOOH concentration of 11.41 ppm. This could be explained by looking at the SEM EDS data, which revealed that the higher the concentration of MXene deposited on the electrode’s surface, the higher the mass percentage of MXene particles deposited. In addition, it showed that MXene catalyzed the production of HCOOH from the electrochemical reduction of CO₂, where MXene had a high surface area and high electrical conductivity. Besides that, the EDX data (Table 1) showed a higher amount of element O, which was from the –O and –OH functional groups on the MXene surface. This promoted higher electronegativity on the surface of BDD with a higher MXene concentration, which led CO₂ to easily come closer to the electrode surface and become further reduced.

3.3.2. The Effect of Reduction Potentials Variation

CO₂ electrochemical reduction was performed for 1 h using the chronoamperometric technique under predetermined optimal conditions. The applied potential for electrochemical CO₂ reduction was varied, specifically at −1.3 V, −1.5 V, and −1.7 V (vs. Ag/AgCl) for each working electrode of bare BDD and MXene-BDD. The product results and the faradaic efficiency for CO₂ reduction are shown in Figure 7. It is suggested that the use of MXene-BDD 2.0 is able to produce large amounts of HCOOH. This may due to the catalytic effect of MXene on the CO₂ reduction process, as explained in the previous section. The HCOOH production decreased as more negative potential reduction was applied. This loss in product yield at an increasingly negative potential might be caused by a huge shift in current density and the high production of H₂ gas, which caused the release of MXene deposited on the BDD surface. As shown in the Figure 7, the bare BDD electrode generated the most HCOOH at a reduction potential of −1.7 V (vs. Ag/AgCl), the MXene-BDD 0.5 and MXene-BDD 1.0 electrodes produced the most at a potential of −1.5 V (vs. Ag/AgCl), and MXene-BDD electrode 2.0 produced the most at a reduction potential of −1.3 V (vs. Ag/AgCl). This data trend is supported by the previous report, in which the high yield of HCOOH was produced at a more negative potential (~2.2 V) [17], and thus, the modified MXene could suppress this higher potential up to a potential of −1.3 V. Moreover, the faradaic efficiency evaluation of BDD and all modified MXene-BDD electrodes showed a similar trend. MXene-BDD 2.0 could produce the highest efficiency of HCOOH of up to ~97% at a potential of −1.3 V. This suggests that MXene could suppress the energy used in the CO₂ electrochemical reduction process.

3.4. MXene Stability Test against CO₂ Electroreduction Application

The stability of the MXene compound deposited on the BDD surface needed to be evaluated on the modified BDD electrode utilized for CO₂ electroreduction applications. The electrochemical approach using the CV method was performed for the characterization, as it is the fastest and most effective way to see the differences just after the CO₂ reduction process finishes. The stability determination of the deposited MXene on the BDD surface was compared before and after being used for CO₂ electrochemical reduction. The CV result of the MXene-BDD 2.0 electrode is shown in Figure 8, revealing the relatively instable MXene on the BDD surface after the reduction process at a potential of −1.7 V for 1 h. This was confirmed by the CV characterization results, which revealed the shift of the potential window of the electrode to a more negative potential. Furthermore, there was an oxidation peak of MXene before reduction at a potential of roughly −0.8 V (vs. Ag/AgCl) and a current of 142 µA, while the peak of MXene fell to 24 µA after reduction. This revealed that there was a 79.9% decrease in MXene on the BDD surface, indicating that the MXene compound was less stable on the BDD electrode surface. To overcome this problem, the MXene was re-deposited for each application. The future challenge is to optimize the unstable modification of MXene on the surface of BDD, which might be overcome by some chemical modifications, rather than let it be physically deposited.

4. Conclusions

MXene (Ti₃C₂T_x) was successfully deposited on the surface of BDD electrodes to be utilized for CO₂ electrochemical reduction applications. The CO₂ reduction result was HCOOH with the highest yield obtained on the MXene-BDD 2.0 electrode at a reduction potential of −1.3 V (vs. Ag/AgCl). The HCOOH formed had a concentration of 28.9 ppm. This result suggests that modifying MXene (Ti₃C₂T_x) on a bare BDD electrode might increase the catalytic effect and potentially solve the high potential reduction requirement for CO₂ reduction on a bare BDD. The further challenge is to improve the stability of the modified MXene on the BDD surface. Additionally, improving the yield and selectivity of the MXene-BDD is currently attempted by modifying it with metal or alloy nanoparticles in the MXene nanolayers, which will be reported on in the near future. By improving the stability and obtaining a higher yield of HCOOH at a low overpotential using MXene modified on a chemically and physically stable BDD electrode, we could realize CO₂ utilization technology by producing higher-valued products with low cost and clean technology.