Systematic Analysis of Electrochemical CO2 Reduction with Various

Mar 22, 2016 - With the Combi-system, our analysis centered on the CO2 supply to the electrode, which affects the rate of the CO2 reduction and the pr...
0 downloads 0 Views 2MB Size
Research Article pubs.acs.org/acscombsci

Systematic Analysis of Electrochemical CO2 Reduction with Various Reaction Parameters using Combinatorial Reactors Hiroshi Hashiba,* Satoshi Yotsuhashi, Masahiro Deguchi, and Yuka Yamada Advanced Research Division, Panasonic Corporation, So̅raku-gun, Kyoto 619-0237, Japan S Supporting Information *

ABSTRACT: Applying combinatorial technology to electrochemical CO2 reduction offers a broad range of possibilities for optimizing the reaction conditions. In this work, the CO2 pressure, stirring speed, and reaction temperature were varied to investigate the effect on the rate of CO2 supply to copper electrode and the associated effects on reaction products, including CH4. Experiments were performed in a 0.5 M KCl solution using a combinatorial screening reactor system consisting of eight identical, automatically controlled reactors. Increasing the CO2 pressure and stirring speed, or decreasing the temperature, steadily suppressed H2 production and increased the production of other reaction products including CH4 across a broad range of current densities. Our analysis shows that the CO2 pressure, stirring speed, and reaction temperature independently contributed to the limiting rate of CO2 supply to the electrode (Jlim). At a constant temperature, the limiting current density of CH4 increased proportionally with Jlim, illustrating that the production rate of CH4 was proportional to CO2 supply. Varying the CO2 pressure and stirring speed hardly affected the maximum Faradaic efficiency of CH4 production. However, changes to the reaction temperature showed a significant contribution to CH4 selectivity. This study highlights the importance of quantitative analysis of CO2 supply in clarifying the role of various reaction parameters and understanding more comprehensively the selectivity and reaction rate of electrochemical CO2 reduction. KEYWORDS: combinatorial screening, electrochemistry, CO2 reduction, copper, methane production



INTRODUCTION The electrochemical reduction of carbon dioxide (CO2) is an important technology option for the conversion of electricity to chemical energy sources, offering the potential to create carbonneutral fuels when combined with renewable energy sources such as solar or wind power. This subject has been attracting interest from a broad range of researchers for decades. A number of transition metals have been found to reduce CO2 to carbon monoxide (CO) and various other organic materials.1,2 It is interesting that methane (CH4) can be produced electrochemically from CO2 despite requiring eight electrons. As the major constituent of natural gas, CH4 satisfies many energy demands in our daily lives, and the efficient production of CH4 from CO2 could help to realize a sustainable society that is less dependent on fossil fuels. Among catalysts tested for the electrochemical CO 2 reduction reaction, copper (Cu) is the only single-metal catalyst that produces CH4 as a major reaction product.1 However, the production of CH4 is not efficient, leading to problems in achieving rapid and selective CH4 production with a low overpotential.3−5 Indeed, CO2 reduction on Cu cannot be fully controlled because of its complicated reaction mechanism, which strongly depends on reaction parameters such as voltage, current density, and the electrolyte. Several theoretical and experimental studies have targeted better understanding of these mechanisms.3−11 However, an understanding of how to link these insights into the reaction © XXXX American Chemical Society

mechanism and the practical control of CO2 reduction for the efficient production of CH4 remains incomplete. While detailed analysis of the CO2 reduction mechanisms is the most common research approach, practical control of the reaction requires an alternative that analyzes the chemical reaction systematically across a wide parameter range. If such an evaluation is quantitative and precise, it can reveal the essence of the complicated reaction and provide new scientific and practical insights toward more efficient CH4 production. Combinatorial technology is a powerful tool for the systematic screening of chemical reactions with multiple parameters. Originating from organic material synthesis,12 the research field using combinatorial technology has expanded into electrochemistry and now includes the analysis of catalysts for fuel cells13 and water-splitting.14,15 However, to the best of our knowledge, there are few examples of published research that focus on CO2 reduction with combinatorial technology,16 and none include the screening of reaction parameters. Such a dearth of previous studies may arise from the difficulty of precisely evaluating the various reaction products when operating at high throughputs, which is crucial for determining the properties of the reaction parameters as well as the catalysts for CO2 reduction. Received: February 10, 2016

A

DOI: 10.1021/acscombsci.6b00021 ACS Comb. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Combinatorial Science

The variations in product distributions with current density for experiments [1−3] are shown in Figure 1a−c. Here, the Faradaic efficiencies of hydrogen (H2), CH4, and other products, including CO, ethylene (C 2 H 4 ), formate (HCOO−), alcohols, and aldehydes, were plotted against current density, elucidating the effect of stirring speed under a CO2 pressure of 1.3 atm and temperature of 25 °C. Without stirring, the Faradaic efficiency of H2 production reached 80% once the current density increased above 20 mA cm−2 (Figure 1a). Conversely, the Faradaic efficiencies of CH4 and other products decreased with increasing current density. However, at a stirring speed of 250 rpm, the trend changed so that increasing current density decreased the Faradaic efficiency of H2 and that of CH4 peaked at a current density of approximately 40 mA cm−2 (Figure 1b). The peak Faradaic efficiency of CH4 was around 50% when the stirring speed was increased to 500 rpm, as shown in Figure 1c. The results indicate that changes to the stirring speed significantly changed the product selectivity, especially the ratio between H2 and products derived from CO2. The effect of changing the CO2 pressure at a stirring speed of 500 rpm and temperature of 25 °C was investigated by considering the three pressure conditions in panels a, b, and c of Figure 2, representing experiments [3, 6, and 9], respectively. Under a CO2 pressure of 4 atm (Figure 2b), the Faradaic efficiency of H2 was suppressed and remained under 40% even when the current density was increased to 200 mA cm−2. Simultaneously, CH4 production dominated the reaction across the broad current density range from 60 to 200 mA cm−2. That trend was further emphasized under 9 atm, as shown in Figure 2c, suggesting near-complete suppression of H2 production and enhancement of CH4 production across an even broader current density range. Figure 3a−c shows the temperature dependence of product distributions obtained under the various temperature conditions of experiments [3, 10, and 12]. The stirring speed and the CO2 pressure held constant at 500 rpm and 1.3 atm, respectively. The peak Faradaic efficiency of CH4 increased from 34% to 71% as the temperature was decreased 40 to 10 °C while, simultaneously, the Faradaic efficiency of H2 decreased. That the increase of Faradaic efficiency of CH4 was abrupt, which was not observed in Figures 1 and 2, suggested the temperature’s effect on the selectivity of CH4 differed from that of the CO2 pressure and stirring speed. Building on the results shown above, we now discuss the effect of CO2 pressure, stirring speed, and temperature on the CO2 supply to the electrode. In a general chemical reaction, the limiting current density (LCD), ilim, corresponds to the limiting rate of mass transport of a reactant, Jlim, as described by the following equation:24

To clarify the effect on the CO2 reduction reaction of various reaction parameters in a rapid and quantitative manner, we introduced a combinatorial screening reactor system (Combisystem). This enabled us to control reaction parameters precisely, including the temperature and the pressure of CO2 inside the reactor. With eight reactors and programmed protocols that included stages spanning from gas introduction to product analysis, an analysis of the parameter-dependence of the reaction products could be readily obtained faster than for conventional one-by-one experiments. With the Combi-system, our analysis centered on the CO2 supply to the electrode, which affects the rate of the CO2 reduction and the product selectivity. In this work, we systematically analyzed the effect of varying the CO2 pressure, stirring speed, and reaction temperature especially on CO2 supply and CH4 production because these parameters have been reported to relate to the CO2 supply and the chemistry of CO2 reduction on a Cu electrode.2,17−23 Although they have been reported, to our knowledge, the effects of these parameters have not been investigated systematically, nor have their roles in the reaction been clarified in a quantitative manner. Such a failure to quantitatively analyze the CO2 supply has been identified as a leading reason for the variation in product distribution obtained by different authors.3 Employing the Combi-system provided a large amount of data for determining the limiting rate of CO2 supply and permitted a broader perspective when considering the effect of these reaction parameters on CO2 supply and CH4 production.



RESULTS AND DISCUSSION Experimental conditions for various conditions of CO2 pressure, stirring speed and temperature are shown in Tables 1 and 2. For each product, the effect of current density on the Table 1. Experimental Matrix Detailing Values of CO2 Pressure and Stirring Speed Analyzed in This Worka CO2 pressure (atm) stirring speed (rpm)

1.3

4

9

0 250 500

[1] [4] [7]

[2] [5] [8]

[3] [6] [9]

A temperature of 25 °C was used in all of the experiments detailed in this table.

a

Table 2. Experimental Matrix of Values of CO2 Pressure and Temperature Analyzed in This Worka CO2 pressure (atm) temperature (°C)

1.3

4

10 25 40

[10] [7] [12]

[11] [8] [13]

ilim =

zF J n lim

(1)

where z and n represent the number of electrons and reactants necessary for the reaction, respectively, and F is the Faradaic constant. On the basis of eq 1, one can directly discuss the extent of mass transport using an experimentally obtained LCD. However, in the cases studied in this work, the Cucatalyzed reaction simultaneously produced several reaction products from CO2, each with a different number of electrons and molecules of CO2 needed for the reaction. Thus, the total LCD of the CO2 reduction did not directly correspond to the limiting rate of mass transport of CO2 (CO2 supply). To

a

A stirring speed of 500 rpm was used for all experiments described in this table.

Faradaic efficiency was obtained. Summary results are displayed and discussed in the section below, and a full suite of the experimental results is presented in Figures S1 and S2 and Table S1 in the Supporting Information. In addition, the production rate of each product is presented in Table S2. B

DOI: 10.1021/acscombsci.6b00021 ACS Comb. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Combinatorial Science

Figure 1. Effect of current density on the Faradaic efficiency of H2 (light blue circles), CH4 (red triangles), and other products (gray squares) at (a) 0 rpm, (b) 250 rpm, and (c) 500 rpm. In these experiments, the CO2 pressure and temperature were 1.3 atm and 25 °C, respectively.

Figure 2. Effect of current density on the Faradaic efficiency of H2 (light blue circles), CH4 (red triangles), and other products (gray squares) at (a) 1.3 atm, (b) 4 atm, and (c) 9 atm. In these experiments, the stirring speed and temperature were 500 rpm and 25 °C, respectively.

Figure 3. Effect of current density on the Faradaic efficiency of H2 (light blue circles), CH4 (red triangles), and other products (gray squares) at (a) 40 °C, (b) 25 °C, and (c) 10 °C. In these experiments, the stirring speed and CO2 pressure were 500 rpm and 1.3 atm, respectively.

Figure 4. (a) Effect of CO2 pressure on Jlim at 0 rpm (red squares), 250 rpm (light blue circles), and 500 rpm (green triangles) at 25 °C. The light red square and light green triangle represent the lower limit of Jlim, which were obtained from experiments [1] and [9], respectively. (b) Effect of CO2 pressure on Jlim at 10 °C (pink squares), 25 °C (green triangles), and 40 °C (orange circles) at a stirring speed of 500 rpm.

experiments [1] and [9], Jlim could not be obtained because the rate of CO2 supply, J (see Supporting Information), failed to show a peak within the measured current density region. We therefore plotted the maximum J within the measured current density region in experiments [1] and [9] for the lower limit of

contend with this issue, we obtained Jlim by modifying eq 1 according to the description in the literature.8 The process is described in full in the Supporting Information. Figure 4a shows the variation of Jlim with changing the CO2 pressure at each stirring speed and a temperature of 25 °C. In C

DOI: 10.1021/acscombsci.6b00021 ACS Comb. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Combinatorial Science Jlim. Figure 4a clearly shows Jlim was proportional to the CO2 pressure, in agreement with previously reported work which discusses the effect of CO2 pressure on the LCD of formic acid with an In electrode.20 Moreover, while all trends were found to be linear, the gradients varied depending on the stirring speed. This suggested that the CO2 pressure and stirring speed independently affected the rate of CO2 supply to the electrode. Figure 4b shows Jlim plotted against the CO2 pressure at the various reaction temperatures investigated and a stirring speed of 500 rpm. As the temperature was decreased from 40 to 10 °C, the gradient representing the relationship between Jlim and CO2 pressure increased slightly. Taken together, these results suggested that Jlim was proportional to the CO2 pressure and independently affected by the stirring speed and the reaction temperature. From the theory of mass transport, Jlim can also be described as24

Figure 6. Variation in the Faradaic efficiency of CH4 with current density for experiments [10] (pink triangles), [11] (red diamonds), [12] (light blue circles), and [13] (blue squares). The red and blue dotted lines represent the maximum CH4 selectivity at 10 and 40 °C, respectively.

that the MaxFE of CH4 was unaffected by CO2 pressure at 10 and 40 °C (with similar results observed at 25 °C). This means that, of the investigated parameters, only temperature was found to impact the selectivity.21−23 Figures 5 and 6 also illustrate that optimization can be approached in two ways. From a chemical engineering point of view, increasing Jlim by controlling the CO2 pressure and stirring speed would bring about a proportional increase in the LCD of CH4 if the temperature was held constant. This is especially important for systems that require high reaction rates, such as, for example, the case of H2 production from the electrolysis of water,25 and CH4 product yields. Alternatively, analysis from a chemistry perspective suggests that temperature could be used to tune the reaction mechanism and affect the selectivity of reaction products, including CH4. Figures 5 and 6 also indicate that the MaxFE of CH4 cannot be increased by adjusting the CO2 pressure or stirring speed. Therefore, increasing the Faradaic efficiency of CH4 above 71% might require the use of other catalysts or electrolytes, accompanied by the optimization of the temperature. Here, the Combi-system offers significant potential for further investigations that aim to optimize the reaction environment to create a more selective CH4 production process. Another area that deserves further attention in future developments toward an efficient CH4 production process involves further analyzing the effect of the CO2 supply on the overpotential of CH4. In this study, we have demonstrated the usefulness of Combi-system in analyzing various parameters in the CO2 reduction reaction through its rapid experimentation and precise evaluation of reaction products. The CO2 supply to the Cu electrode was quantitatively analyzed by calculating Jlim for each experimental condition. The range of results obtained permitted evaluation of the effects of CO2 pressure, stirring speed, and temperature on CO2 supply and CH4 selectivity. Importantly for practical system design, when temperature was held constant, the CO2 pressure and stirring speed were together found to be proportional to the rate of CH4 production, but had little effect on CH4 selectivity. On the other hand, while varying the temperature made a small contribution to CO2 supply, it significantly affected the product selectivity of CH4. Viewing the result as a whole highlights the individual and unified roles that these reaction parameters play in the electrochemical reduction of CO2. Moreover, the novel use of Combi-system in this work opens the possibility of investigating other reaction parameters, such as catalysts and electrolytes, from the viewpoint of CO2 supply and product

DC Jlim = (2) δ where D is the diffusion coefficient of CO2, C is the CO2 concentration in the solution, and δ is the thickness of Nernst’s diffusion layer. Equation 2 indicates that the proportional dependence of Jlim on CO2 pressure can be attributed to the linear relationship between C and the CO2 pressure that is described by Henry’s law. Conversely, the stirring speed is related to δ: a high stirring rate diminishes the thickness of the diffusion layer. Here, using eq 2 and Jlim, the estimated value of δ under the stirred experimental conditions at 25 °C was approximately 0.01−0.02 cm, as shown in Table S3. This is similar to the value observed when using a conventional rotating disk electrode at 75 rpm (0.01 cm).6,8 C and D are also affected by temperature: a temperature decrease causes the CO2 solubility to increase while suppressing the thermal diffusion of CO2. This trade-off may contribute to the relatively small changes in the gradient of the curves shown in Figure 4b. Next we discuss the relationship between the analyzed parameters and CH4 production. Figure 5 shows the effect of

Figure 5. Relationship between Jlim and LCD (red squares) and MaxFE (light blue circles) of CH4.

CO2 pressure and stirring speed on CH4 production by plotting measurements of LCD and the maximum value of the Faradaic efficiency (MaxFE) of CH4 against Jlim for experiments [2−8]. The LCD values for CH4 showed a near-proportional relationship with Jlim, while MaxFE was relatively unaffected (remaining just below 60%). This suggested that the CH4 selectivity was largely almost independent of the CO2 supply (as determined by the CO2 pressure and the stirring speed). Figure 6 shows the variation in the Faradaic efficiency of CH4 with current density for experiments [10−13] and highlights D

DOI: 10.1021/acscombsci.6b00021 ACS Comb. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Combinatorial Science

used instead of KCl as the anolyte to prevent the production of chlorine (Cl2), which is known to occur via the oxidation of the Cl− anion contained in KCl. Further details relating to the reactor and the cathode electrode are presented in the Supporting Information (see Figure S7a−e). As detailed in Tables 1 and 2, 13 experiments were conducted which covered a range of CO2 pressures (1.3, 4, and 9 atm), stirring speeds (0, 250, and 500 rpm), and temperatures (10, 25, and 40 °C). Galvanostatic measurements were performed up to 100 °C in each reactor. This then allowed the variation in the current density corresponding to each individual experiment to be measured. Figure S3 shows that variation between reactors was relatively small. Figures S4 and S5 show that no drastic changes of product selectivity and reaction voltage were observed throughout the time-dependence experiment. Experimental Procedure. Prior to an experiment, the temperature inside the reactors was set to the experimental value using the chiller. Each reactor was bubbled with Ar and then with CO2, each for 60 min and at a flow rate of 125 sccm. The reactors were then pressurized with CO2 to the experimental values. Electrochemical measurements were performed to ascertain the current density while also controlling the stirring speed. After the measurements, gas samples were transferred to the GC instrument and analyzed sequentially while liquid samples were manually extracted from the reactors and then automatically analyzed by HPLC and HSGC. From the results of gas and liquid analyses, we calculated the Faradaic efficiency by dividing the charge ascribed to each product by the total charge.

selectivity. This could greatly aid further understanding of the role of each reaction parameter and the reaction mechanisms to find the optimum reaction conditions for efficient production of CH4.



EXPERIMENTAL PROCEDURES Configuration and Function of the Combi-system. A photograph and a schematic diagram of the Combi-system are shown in panels a and b of Figure 7, respectively. The system

Figure 7. (a) Photograph and (b) schematic diagram of the Combisystem experimental setup.

contained eight identical electrochemical cells, which permitted eight experiments to be completed in parallel. Each cell was contained within a steel reactor that allowed control of the gas pressure. A chiller controlled the temperature of the reactors. A range of other parameters could be controlled independently, including the stirring speed, reaction voltage, current density, and electrolyte. Argon (Ar) and CO2 were introduced to the system from connected cylinders. The system was also connected to a 7890A (Agilent, CA, United States) gas chromatography (GC) instrument, which quantitatively analyzed the reaction products (TCD for H2 and FID for CO, CH4, and C2H4). Electrochemical measurements were simultaneously taken with BT2000 (Arbin, TX, United States) multichannel potentiostats. Because the system components were software controlled (Automation Studio, Freeslate, CA, United States), experimental conditions could be designed in advance. This then permitted the gas introduction, electrochemistry, and analysis of gas products to be completed automatically. For the analysis of liquid products, a Prominence (Shimadzu, Japan) high-performance liquid chromatography (HPLC) system was used for HCOO− detection and a GC-17A (Shimadzu, Japan) with a TurboMatrix40 (PerkinElmer, MA, United States) headspace system (HS-GC) was used for the detection of aldehydes and alcohols. Experimental Conditions. We used strip-shaped Cu plates (Nilaco, Japan, 99.99%) with an active surface area of 1 cm−2 as the cathode electrode for the CO2 reduction reaction. The surface was chemically polished for 30 s using a mixture of nitric acid and phosphoric acid (S-710, Sasaki Chemical, Japan) and rinsed with deionized water. Platinum wire (BAS, Japan) was used as the anode electrode, and saturated Ag/AgCl (Corr instruments, TX, United States) was used as the reference electrode. Each cell was divided into cathode and anode compartments with Nafion 424 (Aldrich, MO, United States). The catholyte and anolyte were 0.5 M KCl (Wako, Japan, pH 3.8 after CO2 bubbling) and 3.0 M KHCO3 (Wako, Japan, pH 8.0 after CO2 bubbling), respectively, except for the experiments performed at 10 °C in which 2.0 M KHCO3 (pH 7.8 after CO2 bubbling) was used as the anolyte to avoid the eduction of the solute at that temperature. Here, KHCO3 was



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acscombsci.6b00021. Full suite of experimental data, demonstration of reproducibility and time-dependence of the experiment, determination of Jlim and δ, and the geometry of the reactor (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors thank Dr. Prasad Oruganti, Mr. Colin Masui, and Dr. Daniel Giaquinta of Freeslate Inc., United States, for codeveloping the Combi-system.



REFERENCES

(1) Hori, Y.; Kikuchi, K.; Suzuki, S. Production of CO and CH4 in electrochemical reduction of CO2 at metal electrodes in aqueous hydrogencarbonate solution. Chem. Lett. 1985, 1695−1698. (2) Azuma, M.; Hashimoto, K.; Hiramoto, M.; Watanabe, M.; Sakata, T. Electrochemical reduction of carbon dioxide on various metal electrodes in low-temperature aqueous KHCO3 Media. J. Electrochem. Soc. 1990, 137, 1772−1778. (3) Gattrell, M.; Gupta, N.; Co, A. A Review of the Aqueous Electrochemical Reduction of CO2 to Hydrocarbons at Copper. J. Electroanal. Chem. 2006, 594, 1−19. E

DOI: 10.1021/acscombsci.6b00021 ACS Comb. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Combinatorial Science (4) Hori, Y. Electrochemical CO2 Reduction of Metal Electrodes. In Modern Aspects of Electrochemistry; Vayenas, C. G., White, R. E., Gamboa-Aldeco, M. E., Eds.; Springer: New York, 2008; pp 89−189. (5) Kuhl, K. P.; Cave, E. R.; Abram, D. N.; Jaramillo, T. F. New Insights into the Electrochemical Reduction of Carbon Dioxide on Metallic Copper Surfaces. Energy Environ. Sci. 2012, 5, 7050−7059. (6) Hori, Y.; Murata, A.; Takahashi, R. Formation of Hydrocarbons in the Electrochemical Reduction of Carbon Dioxide at a Copper Electrode in Aqueous Solution. J. Chem. Soc., Faraday Trans. 1 1989, 85, 2309−2326. (7) Hori, Y.; Takahashi, I.; Koga, O.; Hoshi, N. Selective Formation of C2 Compounds from Electrochemical Reduction of CO2 at a Series of Copper Single Crystal Electrodes. J. Phys. Chem. B 2002, 106, 15− 17. (8) Gupta, N.; Gattrell, M.; MacDougall, B. Calculation for the Cathode Surface Concentrations in the Electrochemical Reduction of CO2 in KHCO3 solutions. J. Appl. Electrochem. 2006, 36, 161−172. (9) Durand, W. J.; Peterson, A. A.; Studt, F.; Abild-Pedersen, F.; Norskov, J. K. Structure Effects on the Energetics of the Electrochemical Reduction of CO2 by Copper Surfaces. Surf. Sci. 2011, 605, 1354−1359. (10) Schouten, K. J. P.; Kwon, Y.; van der Ham, C. J. M.; Qin, Z.; Koper, M. T. M. A New Mechanism for the Selectivity to C1 and C2 Species in the Electrochemical Reduction of Carbon Dioxide on Copper Electrodes. Chem. Sci. 2011, 2, 1902−1909. (11) Singh, M. R.; Clark, E. L.; Bell, A. T. Effects of Electrolyte, Catalyst, and Membrane Composition and Operating Conditions on the Performance of Solar-driven Electrochemical Reduction of Carbon Dioxide. Phys. Chem. Chem. Phys. 2015, 17, 18924−18936. (12) Geysen, H. M.; Meloen, R. H.; Barteling, S. J. Use of Peptide Synthesis to Probe Viral Antigens for Epitopes to a Resolution of a Single Amino Acid. Proc. Natl. Acad. Sci. U. S. A. 1984, 81, 3998−4002. (13) Reddington, E.; Sapienza, A.; Gurau, B.; Viswanathan, R.; Sarangapani, S.; Smotkin, E. S.; Mallouk, T. E. Combinatorial Electrochemistry: A Highly Parallel, Optical Screening Method for Discovery of Better Electrocatalysts. Science 1998, 280, 1735−1737. (14) Jaramillo, T. F.; Ivanovskaya, A.; McFarland, E. W. HighThroughput Screening System for Catalytic Hydrogen-Producing Materials. J. Comb. Chem. 2002, 4, 17−22. (15) Xiang, C.; Suram, S. K.; Haber, J. A.; Guevarra, D. W.; Soedarmadji, E.; Jin, J.; Gregoire, J. M. High-Throughput Bubble Screening Method for Combinatorial Discovery of Electrocatalysts for Water Splitting. ACS Comb. Sci. 2014, 16, 47−52. (16) Dang, T.; Ramsaran, R.; Roy, S.; Froehlich, J.; Wang, J.; Kubiak, C. P. Design of a High Throughput 25-Well Parallel Electrolyzer for the Accelerated Discovery of CO2 Reduction Catalysts via a Combinatorial Approach. Electroanalysis 2011, 23, 2335−2342. (17) Kyriacou, G. Z.; Anagnostopoulos, A. K. Influence of CO2 Partial Pressure and the Supporting Electrolyte Cation on the Product Distribution in CO2 Electroreduction. J. Appl. Electrochem. 1993, 23, 483−486. (18) Hara, K.; Tsuneto, A.; Kudo, A.; Sakata, T. Electrochemical Reduction of CO2 on a Cu Electrode under High Pressure. J. Electrochem. Soc. 1994, 141, 2097−2103. (19) Hara, K.; Kudo, A.; Sakata, T. Electrochemical Reduction of Carbon Dioxide under High Pressure on Various Electrodes in an Aqueous Electrolyte. J. Electroanal. Chem. 1995, 391, 141−147. (20) Todoroki, M.; Hara, K.; Kudo, A.; Sakata, T. Electrochemical Reduction of High Pressure CO2 at Pb, Hg and In Electrodes in an Aqueous KHCO3 Solution. J. Electroanal. Chem. 1995, 394, 199−203. (21) Hori, Y.; Kikuchi, K.; Murata, A.; Suzuki, S. Production of Methane and Ethylene in Electrochemical Reduction of Carbon Dioxide at Copper Electrode in Aqueous Hydrogencarbonate Solution. Chem. Lett. 1986, 897−898. (22) Kim, J. J.; Summers, D. P.; Frese, K. W. Reduction of CO2 and CO to Methane on Cu Foil Electrodes. J. Electroanal. Chem. Interfacial Electrochem. 1988, 245, 223−244. (23) Kaneco, S.; Hiei, N.; Xing, Y.; Katsumata, H.; Ohnishi, H.; Suzuki, T.; Ohta, K. Electrochemical Conversion of Carbon Dioxide to

Methane in Aqueous NaHCO3 Solution at less than 273 K. Electrochim. Acta 2002, 48, 51−55. (24) Atkins, P. W.; Paula, J. D. Atkins’s Physical Chemistry, 7th ed.; Oxford Univ. Press: Oxford, 2002; pp 1031−1033. (25) Carmo, M.; Fritz, D. L.; Mergel, J.; Stolten, D. A Comprehensive Review on PEM Water Electrolysis. Int. J. Hydrogen Energy 2013, 38, 4901−4934.

F

DOI: 10.1021/acscombsci.6b00021 ACS Comb. Sci. XXXX, XXX, XXX−XXX