Study of CO2 reduction at Pt-Ru electrocatalyst in polymer electrolyte cell by differential electrochemical mass spectrometry and liquid chromatography

Accepted Manuscript Study of CO2 reduction at Pt-Ru electrocatalyst in polymer electrolyte cell by differential electrochemical mass spectrometry and liquid chromatography Siyuan Jia, Shofu Matsuda, Shigehisa Tamura, Sayoko Shironita, Minoru Umeda PII:

S0013-4686(17)32721-4

DOI:

10.1016/j.electacta.2017.12.153

Reference:

EA 30941

To appear in:

Electrochimica Acta

Received Date: 26 September 2017 Revised Date:

9 December 2017

Accepted Date: 22 December 2017

Please cite this article as: S. Jia, S. Matsuda, S. Tamura, S. Shironita, M. Umeda, Study of CO2 reduction at Pt-Ru electrocatalyst in polymer electrolyte cell by differential electrochemical mass spectrometry and liquid chromatography, Electrochimica Acta (2018), doi: 10.1016/ j.electacta.2017.12.153. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Study of CO2 reduction at Pt-Ru electrocatalyst in polymer electrolyte cell by

differential

electrochemical

mass

spectrometry

liquid

RI PT

chromatography

and

SC

Siyuan Jia a, Shofu Matsuda a, Shigehisa Tamura a, Sayoko Shironita a, Minoru

a

M AN U

Umeda a,b,*

Department of Materials Science and Technology, Graduate School of Engineering,

Japan

TE D

Nagaoka University of Technology, 1603-1 Kamitomioka, Nagaoka, Niigata 940-2188,

JST, ACT-C, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan

*

Corresponding author:

AC C

EP

b

Department of Materials Science and Technology, Graduate School of Engineering, Nagaoka University of Technology, 1603-1 Kamitomioka, Nagaoka, Niigata 940-2188, Japan Tel/Fax: +81-258-47-9323. Email address: [email protected] (M. Umeda).

1

ACCEPTED MANUSCRIPT

Abstract We describe the CO2 reduction characteristics at the theoretical potential of the CO2

RI PT

reduction by the Pt-Ru electrocatalyst assessed by differential electrochemical mass spectrometry (DEMS) and high-performance liquid chromatography (HPLC) in a

SC

polymer electrolyte single cell. When using the Pt-Ru black catalyst, a small amount of

M AN U

the reduced CO2 was possibly adsorbed on the electrode surface compared to the Pt-Ru/C catalyst. Based on the DEMS analysis of the CO2 electroreduction at the Pt-Ru black catalyst, the methane was not or slightly detected and the generation of carboxylic acids was suggested in the potential range where H2 evolution does not occur. On the

TE D

other hand, the HPLC analysis demonstrated for the first time that the formic acid, acetic acid, and lactic acid were obtained by the CO2 reduction at the Pt-Ru black

EP

electrocatalyst. Based on these results, one interesting production mechanism in which

AC C

three or four kinds of compounds (formic acid, acetic acid, lactic acid, with/without methane) are produced depending on how to desorb the intermediate adsorbents from the electrocatalyst surface could be considered.

Keywords: CO2 electroreduction; polymer electrolyte cell; Pt-Ru electrocatalyst; DEMS; HPLC

2

ACCEPTED MANUSCRIPT

1. Introduction Various approaches of CO2 conversion into other compounds have been developed

RI PT

for mitigating greenhouse CO2 gas emissions, such as electrochemical, photochemical, and biological processes [1-6]. Of these approaches, the CO2 electroreduction has

SC

attracted intensive research because CO2 is reduced with a high efficiency and high

M AN U

selectivity [1-3,7-11]. When using the Cu single crystal electrode in a KHCO3 aqueous solution, the reduction current efficiency, except for hydrogen evolution, was as high as ~90% [7]. Moreover, methane, ethylene, formic acid, and acetic acid were obtained as the products of the CO2 electroreduction [7]. In the cases of using Au and Ag

TE D

electrocatalysts, CO2 was electrochemically reduced to CO with an approximate 90% selectivity [1,12]. However, these processes of the CO2 electroreduction require a very

EP

high overpotential.

AC C

When considering a practical application of the CO2 electroreduction, it is ideal that CO2 is reduced at a potential close to the theoretical CO2 reduction potential (approximately between ± 0.2 V vs. SHE [13]). Remarkably, we previously demonstrated that the CO2 reduction occurred at 0.06–0.25 V vs. DHE by employing a polymer electrolyte single cell equipped with a Nafion-based membrane electrode assembly (MEA) containing a Pt/C electrocatalyst [14]. Pt-based catalysts have been

3

ACCEPTED MANUSCRIPT

receiving a great deal of attention in recent research studies due to their excellent electrocatalytic activities [15,16]. It should be noted that the polymer electrolyte single

RI PT

cell, which is commonly used as the polymer electrolyte fuel cell, also has the advantage of possible power generation as a H2-CO2 fuel cell [17]. In the case of the

SC

CO2 reduction using the Pt/C electrocatalyst, almost all of the reduced CO2 ((CO2)r) was

M AN U

adsorbed on the electrode surface, but a slight amount of methanol can be detected outside the polymer electrolyte single cell [14]. On the other hand, when using the polymer electrolyte single cell equipped with the MEA containing the Pt-Ru/C electrocatalyst, the (CO2)r was easily desorbed from the electrode surface, which

TE D

resulted in the increased detected amount of methanol [18]. However, the potential-dependent product detection and the organic acid detection were difficult

EP

because the product evaluation was investigated by gas chromatography (GC) in these

AC C

previous reports.

Differential electrochemical mass spectrometry (DEMS) is an on-line analytical

method in which a mass spectrometer is directly connected to an electrolysis cell. The electrode reaction can be analyzed in real time with a high sensitivity. The DEMS was applied to the product evaluation of the CO2 electroreduction in an aqueous solution [19]. In our previous studies, the (CO2)r adsorbed on the electrode surface was

4

ACCEPTED MANUSCRIPT

suggested to be CO [20] and the methane production from the CO2 electroreduction was confirmed [21] by the DEMS analysis. However, the quantitative evaluation of the

RI PT

organic acids produced by CO2 electroreduction is still difficult, not only by GC, but also by DEMS. For example, formic acid and acetic acid are known to be thermally

SC

decomposed during the analysis. On the other hand, high-performance liquid

M AN U

chromatography (HPLC) utilizing ion-exclusion chromatography is an analytical method of separating and detecting organic acids by the difference in their acid dissociation constants. There is an advantage of being able to evaluate the formic acid and acetic acid with a high sensitivity [22,23].

TE D

In the present study, in order to evaluate the product generated from the CO2 electroreduction in the range of the theoretical reduction potentials, we examined the

EP

CO2 electroreduction characteristics using Pt-Ru electrocatalysts incorporated in the

AC C

MEA in the Nafion-based polymer electrolyte single cell. The production of the (CO2)r, especially the hydrogen, methane, and carboxylic acid productions, were analyzed during an electrode potential sweep by the direct introduction of the exhaust gas from the Pt-Ru electrode to the DEMS. Furthermore, the exhaust gas from the Pt-Ru electrode was collected as a solution by cooling with ice water, then introduced into the HPLC in order to analyze the organic-acid production during the CO2 electroreduction.

5

ACCEPTED MANUSCRIPT

Based on these analyses, the CO2 reduction reaction mechanism on the surface of the

AC C

EP

TE D

M AN U

SC

RI PT

Pt-Ru electrocatalyst was subsequently considered.

6

ACCEPTED MANUSCRIPT

2. Experimental 2.1. Preparation of a single cell containing a membrane electrode assembly

RI PT

In this report, a membrane electrode assembly (MEA) with a Pt-Ru electrocatalyst was prepared using the procedure we previously outlined [14,18,20,24-26]. As the

SC

working electrode, the Pt-Ru/C (product name: TEC66E50) and Pt-Ru black

M AN U

(TEC90220) powders with the atomic ratio of Pt to Ru of 1:1 were employed. As the counter electrode, the Pt/C powder (TEC10E50E) was used. All the different types of powders were purchased from Tanaka Kikinzoku Kogyo Co., Ltd. (Tokyo, Japan). The Nafion 117 membrane (6 × 6 cm, 0.180 mm thick, Dupont, Wilmington, DE, USA) was

solution,

and

a

TE D

used as the polymer electrolyte and pretreated with Milli-Q water, a 3 wt% H2O2 0.5

mol

dm-3

H2SO4

solution

before

use.

First,

the

EP

electrocatalyst-powder dispersed solution, which contains each electrocatalyst with

AC C

Nafion at the volume ratio of 1:1 and the aqueous solution mixed with 2-propanol, methanol, and Milli-Q water at the weight ratio of 1:2:1, was prepared and sprayed onto 3 × 3 cm water-repellent carbon paper (TGP-H060H, Toray Industries Inc., Tokyo, Japan) for preparation of the working electrode. The loading amount of metal was 1.0 mg cm-2. After preparing the counter electrode by spraying the Pt/C onto other carbon paper, both electrodes were hot-pressed on both sides of the Nafion 117 membrane at

7

ACCEPTED MANUSCRIPT

4.5 kN and 140°C for 10 min. A single cell was finally fabricated by incorporating the prepared MEA and the dynamic hydrogen electrode (DHE) as the reference electrode in

RI PT

the polymer electrolyte cell (Miclab, Kanagawa, Japan). The DHE employed in this study is a stable reference electrode to be operated when using a proton-exchange

M AN U

SC

membrane [27].

2.2. Electrochemical measurement and product evaluation of the CO2 electroreduction The fabricated single cell containing the Pt-Ru black or Pt-Ru/C (1:1) electrocatalyst was connected to a polymer electrolyte fuel cell power generation unit (FCG-20S, ACE,

TE D

Inc.). Humidified H2 at the volume flow of 50 cm3 min-1, humidified CO2 at the volume flow of 50 cm3 min-1, and humidified H2 at the volume flow of 10 cm3 min-1 were

EP

supplied to the counter electrode, the working electrode, and the reference electrode,

AC C

respectively. Under this condition, the cyclic voltammetry was measured at the sweep rate of 0.01 V s-1, in the potential range between 0.05 and 0.70 V vs. DHE, and at the cell temperatures of 40, 60, and 80°C. The instruments used in this study were an HA-310 potentiostat/galvanostat (Hokuto Denko, Tokyo, Japan), an HB-104 function generator (Hokuto Denko), and an HE-151 electrometer (Hokuto Denko). For the single cell containing the Pt-Ru black-based MEA, the potential sweep in

8

ACCEPTED MANUSCRIPT

the potential range between 0.00 and 0.70 V vs. DHE and the 5-min potential hold at 0.03 V vs. DHE following the potential step from 0.35 V to 0.03 V vs. DHE were also

RI PT

operated at the cell temperature of 70°C. It should be noted that the exhaust gas from the working electrode was directly introduced into the JEOL JMS-Q1050GC mass

SC

spectrometer (Tokyo, Japan). The ionization voltage for the DEMS was 23 eV. For a

M AN U

background measurement, the humidified Ar was supplied instead of the humidified CO2.

To analyze the organic acid production, a 90-min potential hold at 70°C was performed at 0.08 V vs. CE (at which the current density was 1.5 mA cm-2) for the

TE D

single cell containing the Pt-Ru black electrocatalyst. During this measurement, the exhaust gas from the working electrode was collected as a solution by cooling with ice

EP

water. The collected solution was analyzed by the Shimadzu high-performance liquid

AC C

chromatograph Prominence (Kyoto, Japan). The detector and the column used in this study were a CDD-10AVP conductivity detector (Shimadzu) and a Shim-pack SCR-102H (Shimadzu), respectively. Regarding the DEMS and HPLC calibrations, we checked those instrumental responses using standard samples with each series of quantitative experiments. For the DEMS response, we made the calibration curve (three plots) of the intensity of m/z 15

9

ACCEPTED MANUSCRIPT

using the 94.1 ppm methane standard gas (Takachiho Chemical Industrial Co., Ltd., Tokyo, Japan). For the HPLC response, we made the calibration curves (five plots for

RI PT

each curve) of formic acid, acetic acid, and lactic acid using these standard aqueous solutions. The standard formic acid, acetic acid, and lactic acid were purchased from

SC

Wako Pure Chemical Industries, Ltd. (Osaka, Japan), Nakalai Tesque, Inc. (Kyoto,

AC C

EP

TE D

M AN U

Japan), and Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan), respectively.

10

ACCEPTED MANUSCRIPT

3. Results and discussion

60℃

40℃

80℃ Onset potential / V vs. DHE

0.5

0 -10

Pt-Ru/C(1:1)

5 0 -5 -10 -15 0

Pt-Ru black 0.2 0.4 0.6 Potential / V vs. DHE

0.4 0.3 0.2

M AN U

Current density / mA cm-2

10

SC

(b)

(a)

RI PT

3.1. Cyclic voltammograms of Pt-Ru electrocatalyst

0.1

0 40

0.8

Pt-Ru/C(1:1) Pt-Ru black(1:1) 60

80

Cell Temperature / ℃

TE D

Fig. 1. Cyclic voltammograms of Pt-Ru electrocatalysts at each temperature under a CO2-saturated atmosphere (a) and temperature dependence of (CO2)r reoxidation onset

EP

potential for the Pt-Ru electrocatalysts (b).

AC C

It should be noted that the cyclic voltammogram observed in the Ar atmosphere is

similar to that of Pt-Ru under an acidic condition [28,29] as shown in Fig. 3b. Fig. 1a shows the cyclic voltammograms of the Pt-Ru black (1:1) and Pt-Ru/C (1:1) catalysts incorporated in a polymer electrolyte single cell under the CO2 atmosphere. In the CO2 atmosphere, there are the following three characteristic peaks in each voltammogram: (i) the reduction current including the H-adsorption [28,29] and CO2 reduction in the 11

ACCEPTED MANUSCRIPT

potential range around 0.05–0.20 V vs. DHE, (ii) the oxidation peak of the H-desorption in the potential range around 0.05–0.20 V vs. DHE [28,29], and (iii) the peak of the

RI PT

(CO2)r reoxidation in the potential range around 0.20–0.60 V vs. DHE, which are consistent with our previous results using the Pt-Ru/C [18,20,21]. Fig. 1b shows the

SC

temperature dependence of the (CO2)r reoxidation onset potential. The initial potential at

M AN U

which the (CO2)r reoxidation occurs was defined as the (CO2)r reoxidation onset potential. For both electrocatalysts, the onset potentials shift in the negative direction with the increased temperature, which is suggestive of improving the reactivity of the CO2/(CO2)r [30]. On the other hand, the potential width at which the onset potential

TE D

shifted is almost constant regardless of the presence or absence of the supported carbon. Hence, the onset potentials are almost the same for both electrocatalysts at the

EP

temperatures of 40, 60, and 80°C.

AC C

The ratio of the coulombic charges derived from the H-adsorption reaction or the adsorbed (CO2)r reaction to the total reduction reaction (rH or rC, respectively) were then calculated as follows: rH = QH / Qtotal × 100

(1)

rC = QC / Qtotal × 100

(2)

where QH is the coulombic charge of the H-desorption oxidation current at 0.05–0.20 V

12

ACCEPTED MANUSCRIPT

vs. DHE in Fig. 1a, QC is the coulombic charge of the (CO2)r reoxidation current at 0.20–0.60 V vs. DHE in Fig. 1a, and Qtotal is the coulombic charge of the reduction

RI PT

current at 0.05–0.20 V vs. DHE in Fig. 1a. The dependence of the calculated rH and rC on the cell temperature is shown in Fig. 2. In this experimental system, the following

SC

equation was established:

(3)

M AN U

rtotal = rH + rC + rdes

where rdes is defined as the ratio of the coulombic charges derived from the CO2 reduction reaction associated with the desorbed (CO2)r. Hence, the rdes should be estimated from the comparison of rtotal, rH, and rC. As shown in Fig. 2, the rC increases

TE D

and rH decreases with the increasing cell temperature for both electrocatalysts, which indicates that the CO2 reduction reaction preferentially occurs compared to the

EP

underpotential deposition of H when the cell temperature is high. In fact, the rC

AC C

significantly increases and rH slightly decreases for Pt-Ru/C (1:1), whereas the rC moderately increases and rH significantly decreases for Pt-Ru black (1:1). These results indicated that the rdes of the Pt-Ru black is higher than that of Pt-Ru/C from Eq. (3), which is suggestive of a small amount of (CO2)r adsorption on the electrode surface of the Pt-Ru black electrocatalyst. Therefore, we employed the Pt-Ru black electrocatalyst in the subsequent experiments because it can easily release the product to the outside of

13

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

the cell.

AC C

EP

TE D

Fig. 2. Cell-temperature dependence of rH and rC for each of the Pt-Ru electrocatalysts.

14

ACCEPTED MANUSCRIPT

3.2. DEMS analysis

(b)

0

-60 CO2 atmosphere

-80 -100

-5

0

RI PT

-40

5

-20 -40

0

-60

Ar atmosphere

-80 -100

SC

-20

Current density / mA cm

-2

Current density / mA cm

0

20

Intensity / arb. unit

1500 m/z 15

1000

80000 0 m/z 45

1500 0

m/z 15

1000

2000

m/z 45

100

40000 0 0

-5

m/z 2

M AN U

Intensity / arb. unit

m/z 2

0

Current density / mA cm-2

5

Current density / mA cm-2

20

-2

(a)

0

0.2

0.4

0.6

0.8

TE D

Potential / V vs. DHE

0

0.2 0.4 0.6 Potential / V vs. DHE

0.8

Fig. 3. Cyclic voltammograms and differential electrochemical mass spectra (m/z 2, 15,

EP

and 45) of Pt-Ru black electrocatalysts at 70°C under humidified CO2 (a) and Ar (b)

AC C

atmospheres. Dashed curves are the enlarged cyclic voltammograms between the current density from -5 to 5 mA cm-2.

Figs. 3a and 3b show a cyclic voltammogram at the Pt-Ru black with the mass intensities of m/z 2, 15, and 45 simultaneously measured under a humidified CO2 atmosphere and a humidified Ar atmosphere, respectively. As noted in the potential

15

ACCEPTED MANUSCRIPT

range of 0.15–0.00 V vs. DHE in Fig. 3, the intensity of m/z 2, which means the degree of H2 evolution, increases with the increase in the reduction current in Fig. 3b, while the

RI PT

intensity of m/z 2 in Fig. 3a is flat in the same potential range. This suggests that H2 evolution occurs in the Ar atmosphere but does not occur in the CO2 atmosphere in this

SC

potential range, as observed in our previous study (using the Pt-Ru/C electrocatalyst)

M AN U

[21]. It should be noted that the increase in the intensity of m/z 15 (a fragment of CH4) is negligibly small even when the reduction current is initially observed in Fig. 3a. Therefore, CH4 is not or slightly produced by the CO2 electroreduction at the Pt-Ru black catalyst although we previously observed the production of CH4 at the Pt-Ru/C

TE D

catalyst [21]. For the m/z 45 (a fragment of carboxylic acids), the intensity at 0.15–0.00 V in Fig. 3a increases with the increase in the reduction current, but that in Fig. 3b does

EP

not increase, which is suggestive of the generation of carboxylic acids by the CO2

AC C

electroreduction.

Fig. 4 shows the chronoamperogram and the differential electrochemical mass

spectra (m/z 2, 15, and 45) simultaneously measured at the Pt-Ru black electrocatalyst at 70°C under a CO2 atmosphere during the 5-min potential hold at 0.03 V vs. DHE after the potential step from 0.35 V to 0.03 V vs. DHE. A reduction current and a negligibly small increase in the m/z 15 are confirmed only immediately following the potential step,

16

ACCEPTED MANUSCRIPT

whereas a signal of m/z 2 remains almost constant. These results also indicate that the CH4 production from the CO2 reduction does not or slightly occur immediately after the

RI PT

potential step, then being inhibited by the electrocatalyst-surface poisoning. The CH4 generation by the CO2 reduction reaction is considered to be represented by the

SC

following equation [31]:

M AN U

CO2 + 8H+ + 8e- → CH4 + 2H2O.

(4)

Based on the result of Fig. 4 and Eq. (4), the coulombic efficiency of the CH4 production is calculated to be 5.0 × 10-2%. The increase in the intensity of m/z 45 is observed during the 5-min potential hold at 0.03 V vs. DHE, which supports the

AC C

EP

TE D

production of carboxylic acids.

17

10 0

-20 -30 2400 -40

m/z 2

1200 2000 0

m/z 15

M AN U

Intensity / arb. unit

RI PT

-10

SC

Current density / mA cm-2

ACCEPTED MANUSCRIPT

1000

60000 0

m/z 45

30000 0

100 200 Time / s

TE D

0

300

EP

Fig. 4. The chronoamperogram and differential electrochemical mass spectra (m/z 2, 15, and 45) at the Pt-Ru black electrocatalyst under a CO2 atmosphere during the 5-min

AC C

potential hold at 0.03 V vs. DHE following the potential step from 0.35 V to 0.03 V vs. DHE. The cell temperature was 70°C.

18

ACCEPTED MANUSCRIPT

3.3. HPLC analysis Since a qualitative and quantitative evaluation of the different carboxylic acids

RI PT

generated from the CO2 reduction at the Pt-Ru black electrocatalyst is difficult by the DEMS analysis, we then performed the high-performance liquid chromatography

SC

(HPLC) analysis. An exhaust solution was collected from the working electrode during the 90-min potential hold at 0.08 V vs. CE (at which the current density was 1.5 mA

M AN U

cm-2), then analyzed by HPLC. The result is shown in Fig. 5. The peaks whose retention times were ~1, ~14, 22.0, 23.8, and 25.6 min are observed in the chromatogram of the collected sample (Pt-Ru black). The peaks at the times around ~1

TE D

and ~14 min matched the peaks observed in the sample of Milli-Q water (see Fig. A.1 in Appendix). Notably, Fig. 5 demonstrates that lactic acid, formic acid, and acetic acid

EP

are obtained as the products of the CO2 reduction at the Pt-Ru black electrocatalyst

AC C

since the peaks at the times of 22.0, 23.8, and 25.6 min matched with the main peaks of the standard samples for them. As a control, an electroless sample under a CO2 atmosphere was also analyzed. No peaks are observed at any retention times except for the peaks at ~1 and ~14 min as shown in Fig. A.1 in the Appendix). As shown in Table 1, the coulombic efficiencies for the production of formic acid, acetic acid, and lactic acid were calculated to be 4.5 × 10-4%, 8.6 × 10-3%, and 9.1 ×

19

ACCEPTED MANUSCRIPT

10-2%, respectively, using the results of Fig. 5 and the following reaction equations: (5)

2CO2 + 8H+ + 8e- → CH3COOH + 2H2O

(6)

3CO2 + 12H+ + 12e- → CH3CH(OH)COOH + 3H2O.

RI PT

CO2 + 2H+ + 2e- → HCOOH

(7)

SC

Compared to the coulombic efficiencies at the Pt-Ru/C (Pt:Ru=8:2 at. ratio)

M AN U

electrocatalyst (see Table. A.1 in the Appendix), it is noted that Pt-Ru black more efficiently generates all of the formic acid, acetic acid, and lactic acid. To the best of our knowledge, the production process of not only formic acid and acetic acid, but also lactic acid during the CO2 electroreduction with the Pt-Ru catalyst at around its

TE D

theoretical potential was found for the first time in the present study, although the major production process of the CO2 electroreduction at the Pt-Ru catalyst remains unclear at

EP

present. One possible major process is the CO adsorption [32-34] since the DEMS

AC C

result of the CO2 electrode reaction by stripping voltammetry at the Pt-Ru/C electrocatalyst demonstrated the CO adsorption [20]. It could also be considered that other carboxylic acids having higher molecular weight than those detected in this study is generated since the intensity of m/z 45 significantly increases with the increase in the reduction current in Figs. 3a and 4. In the future, improving the production efficiency will be required because the coulombic efficiencies reported in this paper are low when

20

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

compared to the magnitude of rC.

21

Intensity / arb. scale

ACCEPTED MANUSCRIPT

RI PT

Pt-Ru black

Lactic acid

SC

Formic acid Acetic acid 5

10 15 20 Retention time / min

25

M AN U

0

30

Fig. 5. The liquid chromatograms of a sample collected during the 90-min CO2

TE D

reduction at 0.08 V vs. CE with Pt-Ru black electrocatalyst at 70°C and standard

EP

samples of lactic acid, formic acid, and acetic acid.

AC C

Table 1. Coulombic efficiencies of the products from the CO2 reduction at the Pt-Ru black electrocatalyst. Product

Coulombic efficiency / %

Formic acid

4.5 × 10-4

Acetic acid

8.6 × 10-3

Lactic acid

9.1 × 10-2

22

ACCEPTED MANUSCRIPT

3.4. Consideration of the production mechanism Because we demonstrated that three or four kinds of products, which were

RI PT

determined to be formic acid, acetic acid, lactic acid, with/without methane from the DEMS and HPLC analyses, were obtained outside the cell containing the Pt-Ru black

SC

electrocatalyst by the CO2 electroreduction, we focused our attention on their

M AN U

production mechanism in this section. Considering the presence of many CO2-reduction adsorbents on the electrocatalyst surface as seen in Fig. 2, we propose one potential production mechanism for the four kinds of compounds as shown in Fig. 6. During the CO2 reduction process, three kinds of reaction intermediates, which should be -CH3,

TE D

-COOH, and -CH2OH, are initially adsorbed on the electrocatalyst surface. When the reaction intermediates desorb from the surface by themselves, methane, formic acid, and

EP

methanol are suggested to be generated from the -CH3, -COOH, and -CH2OH,

AC C

respectively. On the other hand, in the case of the intermediate coupling desorption, acetic acid, ethanol, and lactic acid are suggested to be produced from the coupling of -CH3:-COOH, -CH3:-CH2OH, and -CH3:-COOH:-CH2OH, respectively. It should be noted that methanol and ethanol were not detected in this study. This might be because all of the -CH2OH was used for the production of the lactic acid. Overall, it is suggested that methane, formic acid, acetic acid, and lactic acid are produced depending on how

23

ACCEPTED MANUSCRIPT

the intermediate adsorbents are desorbed.

COOH

CH3 COOH

Lactic acid

CH2OH COOH CH3

SC

CH3

Acetic acid

RI PT

Methane

Formic acid

M AN U

Catalyst surface

Fig. 6. Schematic image of the production mechanism of methane, formic acid, acetic

AC C

EP

TE D

acid, and lactic acid due to the CO2 reduction on the surface of the Pt-Ru electrocatalyst.

24

ACCEPTED MANUSCRIPT

4. Conclusions In this study, we prepared a polymer electrolyte single cell equipped with a

RI PT

membrane electrode assembly containing the Pt-Ru electrocatalyst, then its CO2 electroreduction characteristics were analyzed by differential electrochemical mass

SC

spectrometry (DEMS) and high-performance liquid chromatography (HPLC). The

M AN U

DEMS and HPLC analyses elucidated for the first time the production process that three or four kinds of compounds classified as formic acid, acetic acid, lactic acid, with/without methane are generated by the CO2 electroreduction using the Pt-Ru black catalyst at around the theoretical reduction potential. Furthermore, we considered that

TE D

the production mechanism could be the coupling of the intermediates adsorbed on the electrocatalyst surface. These results will assist in the design of the CO2 conversion into

EP

useful compounds, although further investigation, which improves the production

AC C

efficiency, will be necessary.

25

ACCEPTED MANUSCRIPT

References [1] Y. Hori, Electrochemical CO2 reduction on metal electrodes, Mod. Aspect.

RI PT

Electrochem. 42 (2008) 89.

[2] Y. Chen, C.W. Li, M.W. Kanan, Aqueous CO2 reduction at very low overpotential on

SC

oxide-derived Au nanoparticles, J. Am. Chem. Soc. 134 (2012) 19969.

M AN U

[3] J. Qiao, Y. Liu, F. Hong, J. Zhang, A review of catalysts for the electroreduction of carbon dioxide to produce low-carbon fuels, Chem. Soc. Rev. 43 (2014) 631. [4] B. Kumar, J.M. Smieja, C.P. Kubiak, Photoreduction of CO2 on p-type silicon using Re(bipy-But)(CO)3Cl: photovoltages exceeding 600 mV for the selective reduction of

TE D

CO2 to CO, J. Phys. Chem. C 114 (2010) 14220. [5] M. Anjos, B.D. Fernandes, A.A. Vicente, J.A. Teixeira, G. Dragone, Optimization of

EP

CO2 bio-mitigation by Chlorella vulgaris, Bioresour. Technol. 139 (2013) 149. Biological

Solar Energy,

Imperial College

AC C

[6] M.D. Archer, J.E. Barber, Press, London, 2004.

[7] M. Gattrell, N. Gupta, A. Co, Electrochemical reduction of CO2 to hydrocarbons to store renewable electrical energy and upgrade biogas, Energy Convers. Manage., 48 (2007) 1255. [8] I. Takahashi, O. Koga, N. Hoshi, Y. Hori, Electrochemical reduction of CO2 at

26

ACCEPTED MANUSCRIPT

copper Cu(S)-[n(111) × (111)] and Cu(S)-[n(110) × (100)] electrodes, J. Electroanal. Chem. 533 (2002) 135.

RI PT

[9] Y. Hori, I. Takahashi, O. Koga, N. Hoshi, Selective formation of C2 compounds from electrochemical reduction of CO2 at a series of copper single crystal electrodes, J.

SC

Phys. Chem. B 106 (2002) 15.

M AN U

[10] T. Yamamoto, D.A. Tryk, A. Fujishima, H. Ohta, Production of syngas plus oxygen from CO2 in a gas-diffusion electrode-based electrolytic cell, Electrochim. Acta, 47 (2002) 3327.

[11] Y. Hori, H. Wakebe, T. Tsukamoto, O. Koga, Electrocatalytic process of CO

TE D

selectivity in electrochemical reduction of CO2 at metal electrodes in aqueous media, Electrochim Acta, 39 (1994) 1833.

EP

[12] Q. Lu, J. Rosen, Y. Zhou, G. S. Hutchings, Y.C. Kimmel, J.G. Chen, F. Jiao, A

AC C

selective and efficient electrocatalyst for carbon dioxide reduction, Nat. Commun. 5 (2014) 3242.

[13] A.J. Bard, R. Parsons, J. Jordan, Standard potentials in aqueous solution, Marcel Dekker, Inc., NY, 1985. [14] S. Shironita, K. Karasuda, M. Sato, M. Umeda, Feasibility investigation of methanol generation by CO2 reduction using Pt/C-based membrane electrode assembly

27

ACCEPTED MANUSCRIPT

for a reversible fuel cell, J. Power Sources 228 (2013) 68. [15] X. Huang, Z. Zhao, L. Cao, Y. Chen, E. Zhu, Z. Lin, M. Li, A. Yan, A. Zettl, Y.M.

RI PT

Wang, X. Duan, T. Mueller, Y. Huang, High-performance transition metal–doped Pt3Ni octahedra for oxygen reduction reaction, Science 348 (2015) 1230.

SC

[16] Q. Zhao, G. Zhang, G. Xu, Y. Li, B. Liu, X. Gong, D. Zheng, J. Zhang, Q. Wang,

M AN U

Synthesis of highly active and dual-functional electrocatalysts for methanol oxidation and oxygen reduction reactions, Appl. Surf. Sci. 389 (2016) 181. [17] M. Umeda, M. Sato, T. Maruta, S. Shironita, Is power generation possible by

114 (2013) 174908.

TE D

feeding carbon dioxide as reducing agent to polymer electrolyte fuel cell, J. Appl. Phys.

[18] S. Shironita, K. Karasuda, K. Sato, M. Umeda, Methanol generation by CO2

EP

reduction at a Pt-Ru/C electrocatalyst using a membrane electrode assembly, J. Power

AC C

Sources 240 (2013) 404.

[19] S. Wasmus, E. Cattaneo and W. Vielstich, Reduction of carbon dioxide to methane and ethene–an on-line MS study with rotating electrodes, Electrochim. Acta 35 (1990) 771. [20] S. Shironita, K. Sato, K. Yoshitake, M. Umeda, Pt-Ru/C anode performance of polymer electrolyte fuel cell under carbon dioxide atmosphere, Electrochim. Acta 206

28

ACCEPTED MANUSCRIPT

(2016) 254. [21] S. Shironita, K. Sato, M. Umeda, Mass spectrometry study of CO2 electroreduction

RI PT

at membrane electrode assembly incorporating Pt-Ru/C, Electrocatalysis, DOI: 10.1007/s12678-017-0413-7.

SC

[22] R. Kortlever, C. Balemans, Y. Kwon, M.T.M. Koper, Electrochemical CO2

M AN U

reduction to formic acid on a Pd-based formic acid oxidation catalyst, Catal. Today 244 (2015) 58.

[23] Y. Kwon, M.T.M. Koper, Combining Voltammetry with HPLC: Application to Electro-Oxidation of Glycerol, Anal. Chem. 82 (2010) 5420.

TE D

[24] M. Inoue, T. Iwasaki, K. Sayama, M. Umeda, Effect of conditioning method on direct methanol fuel cell performance, J. Power Sources 195 (2010) 5986.

EP

[25] M. Umeda, K. Sayama, M. Inoue, Temperature and methanol concentration

AC C

dependences of direct methanol fuel cell performance measured by single cell having reference electrode, J. Renew. Sustain. Energy 3 (2011) 043107. [26] M. Inoue, T. Iwasaki, M. Umeda, Electrochemical impedance spectroscopy of direct methanol fuel cell having Ag/Ag2SO4 reference electrode, Electrochemistry 79 (2011) 329. [27] A. Küver, I. Vogel, W. Vielstich, Distinct performance evaluation of a direct

29

ACCEPTED MANUSCRIPT

methanol SPE fuel cell. A new method using a dynamic hydrogen reference electrode, J. Power Sources 52 (1994) 77.

RI PT

[28] H.A. Gasteiger, N. Markovic, P.N. Ross Jr., E.J. Cairns, Methanol electrooxidation on well-characterized platinum-ruthenium bulk alloys, J. Phys. Chem. 97 (1993) 12020.

SC

[29] M. Umeda, H. Ojima, M. Mohamedi, I. Uchida, Methanol electrooxidation at Pt–

M AN U

Ru–W sputter deposited on Au substrate, J. Power Sources 136 (2004) 10. [30] E. Laviron, The use of linear potential sweep voltammetry and of a.c. voltammetry for the study of the surface electrochemical reaction of strongly adsorbed systems and of redox modified electrodes, J. Electroanal. Chem. 100 (1979) 263.

TE D

[31] A.J. Bard, R. Parsons, J. Jordan, Standard Potentials in Aqueous Solution, Marcel Dekker, Inc., New York, 1985.

AC C

(1967) 1213.

EP

[32] M.W. Breiter, On the nature of reduced carbon dioxide, Electrochim. Acta 12

[33] S. Taguchi, A. Aramata, M. Enyo, Reduced CO2 on polycrystalline Pd and Pt electrodes in neutral solution: electrochemical and in situ Fourier transform IR studies, J. Electroanal. Chem. 372 (1994) 161. [34] M.L. Marcos, J.G. Velasco, F. Hahn, B. Beden, C. Lamy, A.J. Arvia, In situ FTIRS study of ‘reduced’ CO2 on columnar-structured platinum electrodes in different acid

30

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

media, J. Electroanal. Chem. 436 (1997) 161.

31

ACCEPTED MANUSCRIPT

Highlights ・CO2 was reduced at around its theoretical reduction potential.

RI PT

・The reduced CO2 was easily desorbed from the Pt-Ru black electrocatalyst. ・Methane was not or slightly detected by the DEMS analysis as a product.

SC

・The HPLC analysis showed the production of formic acid, acetic acid, and lactic acid.

AC C

EP

TE D

M AN U

・The formation mechanism of the CO2-electroreduced products was proposed.

1