Isotopic effect on electrochemical CO₂ reduction activity and selectivity in H₂O- and D₂O-based electrolytes over palladium

Abstract

The isotopic effect on the electrochemical CO₂ reduction reaction (CO₂RR) is investigated in this study. A higher CO₂RR selectivity over its competing hydrogen evolution reaction was observed in D₂O-based electrolytes compared with the H₂O-based counterparts, which can be attributed to the lower [D⁺] concentration than [H⁺].

---

The increasing demand for effectively mitigating the growing atmospheric CO₂ level promotes the electrochemical CO₂ reduction reaction (CO₂RR) in aqueous electrolytes as one of the viable options.^1–3 This is because the CO₂RR, coupled with renewable energy resources, can produce value-added chemicals as a resource for downstream thermochemical reactions, thus potentially enabling the net reduction of CO₂.³ Extensive studies have shown that the conversion and selectivity (i.e., total current density and faradaic efficiency, respectively) of catalysts can be controlled by alloy formation,^4–6 phase transition,^7,8 additives,⁹ surface functional groups,^10–13 and different salts^14,15 in the electrolyte. However, rarely has the isotopic effect of the aqueous media been explored.

The current study investigates the effect of deuterium oxide (D₂O) on the CO₂RR over carbon-supported Pd (Pd/C, see Fig. S1 for X-ray diffraction pattern in the ESI†) with four different 0.1 M alkali-ion bicarbonate electrolytes (0.1 M AHCO₃ or 0.1 M ADCO₃, A = Na⁺ or K⁺, see the ESI† for the experimental details). Interestingly, deuterium-based electrolytes exhibited markedly lowered D₂ evolution reaction (DER) activity compared with the H₂ evolution reaction (HER) activity in H₂O-based electrolytes, which is ascribed to lower deuteron concentration ([D⁺]) than proton concentration ([H⁺]) in the corresponding electrolytes. The different [D⁺] and [H⁺] concentrations originated from the different dissociation constants (pK) of D₂O and H₂O.¹⁶ In contrast to the suppressed DER activity, the CO₂RR activity was enhanced in D₂O-based electrolytes, implying that the isotopic effect can allow for different product distribution in the synthesis gas (i.e., CO/D₂ and CO/H₂ ratios).

It is well-known that H⁺ is readily absorbed into the Pd lattice to form Pd hydride (PdH) at potentials for the CO₂RR.^6–8 This phase transformation thereby enables co-production of CO and H₂; otherwise, metallic Pd would be poisoned by CO, resulting in a negligible yield of CO production. In order to investigate if Pd forms Pd deuteride (PdD) like PdH formation, in situ X-ray absorption fine structure (XAFS) analysis¹⁷ was conducted (see the ESI†) from 0.3 V to −0.8 V versus reversible hydrogen electrode (RHE). Note that all of the potentials (V) in this study are versus RHE. With potentials being applied more negatively, X-ray absorption near edge structure (XANES) spectra show the gradual shift toward lower energy, reflecting the progressive PdD and PdH formation under CO₂RR conditions in 0.1 M NaDCO₃ and NaHCO₃ electrolytes, respectively (Fig. 1A and B). This result implies that the PdD formation likely occurs in the same manner as PdH formation, also evidenced by the interatomic distance profile of Pd and its first neighbor Pd (R_Pd–Pd) obtained from in situ extended XAFS (EXAFS) analysis (Fig. 1C, D and Fig. S2, Table S1, ESI†). In each electrolyte, D⁺(H⁺) diffusion into Pd starts around 0 V and saturates around −0.6 V to form PdD(PdH) with an increase in R_Pd–Pd by ∼2.4%, which is also consistent with the changes in the XANES peak positions.⁷ Meanwhile, the coordination number of Pd remains similar, implying that the particle size is unchanged during the CO₂RR (Fig. S3 in the ESI†).

Fig. 1 In situ XAFS analysis measured at the Pd K-edge in different electrolytes. (A and B) In situ XANES profiles from 0.3 V (bottom) to −0.8 V (top) in 0.1 M (A) NaDCO₃ and (B) NaHCO₃ electrolytes. The dotted yellow lines are vertical lines to visualize the peak shifts. (C and D) In situ EXAFS profiles under the same conditions.

Fig. S4 in the ESI† shows the linear scanning voltammetry (LSV) curves of Pd/C in 0.1 M NaHCO₃ and NaDCO₃ electrolytes saturated with CO₂ or Ar. Under CO₂-saturated conditions (Fig. S4A, ESI†), similar LSV profiles are observed in both electrolytes. In contrast, it can be seen that, under Ar-saturated conditions (Fig. S4B, ESI†), the current density in 0.1 M NaHCO₃ electrolyte is higher than that in a 0.1 M NaDCO₃ electrolyte, indicating that a simple replacement of H₂O with D₂O can significantly reduce the DER activity, as will be discussed later in the product analysis. To investigate the isotope effect of D₂O on CO₂RR behavior, the activity of Pd/C was evaluated by using chronoamperometry (CA) techniques from −0.6 to −1.0 V with a potential interval of 0.1 V in 0.1 M AHCO₃ or ADCO₃ electrolytes as shown in Fig. S5A–D (ESI†). The CA current densities in all of the electrolytes are stable. The gaseous products were analyzed by using gas chromatography, and the faradaic efficiency (FE) of each product was calculated (see the ESI†). As displayed in Fig. 2, CO and H₂ (or D₂) are the main products in H₂O-based (or D₂O-based) electrolytes, with the sum of FE(CO) and FE(H₂/D₂) being near or over 90%. The only exception is at −0.6 V in 0.1 M NaHCO₃, which might be attributed to the formation of oxygenated product (i.e., formate), although the product yield is too low (∼10⁻⁵ M) even after a prolonged CO₂RR period to be detected in the ¹H-NMR analysis (see the ESI†). In addition, the long-term stability tests up to 8 h at −0.8 V in D₂O-based electrolytes suggest that D₂O is an effective media for the electrochemical CO₂RR (Fig. S5E in the ESI†). After 8 h electrolysis, the FE(CO) values decreased by ∼7 and ∼12% in 0.1 M KDCO₃ and NaDCO₃ electrolytes (Fig. S6 in the ESI†), respectively, which seems to originate from catalyst agglomeration. This is also supported by transmission electron microscopy images taken before and after the long-term electrolysis, which show the well-dispersed Pd particles over C with a small increase in size (Fig. S7 in the ESI†).

Fig. 2 Faradaic efficiency of CO and H₂ in different electrolytes. (A) 0.1 M NaHCO₃, (B) 0.1 M NaDCO₃, (C) 0.1 M KHCO₃, and (D) 0.1 M KDCO₃.

However, the resultant product distribution, CO/H₂ or CO/D₂, can be significantly affected by changing H₂O with D₂O over the entire potential range. For example, Pd/C in 0.1 M NaDCO₃ shows a FE(CO) of 76.1% at 0.8 V, which is 1.58 times higher than that (48.2%) at the same potential in 0.1 M NaHCO₃ (Fig. 2A and B). As the pH values of all electrolytes are near ∼6.8, one can exclude the possibility that the CO₂RR performance is influenced by the pH effect. Therefore, such an enhancement can be explained by the different [D⁺] and [H⁺] concentrations in those near neutral-pH electrolytes. When H₂O is replaced with D₂O, [D⁺] is roughly 33% of [H⁺], owing to the higher pK value (14.491) of D₂O than that (13.995) of H₂O.¹⁶ This leads to impaired DER activity in comparison with HER activity. As a consequence, more catalytic sites would be available for the CO₂RR rather than DER, in turn enhancing FE(CO) in 0.1 M NaDCO₃ over 0.1 M NaHCO₃. Such an isotopic effect of D₂O is also observed in K-based electrolytes (Fig. 2C and D). The enhancement in FE(CO) in 0.1 M KHCO₃ is attributed to the different hydrolysis capability of Na and K ions (the so-called, cation effect),^14,15 which is now well-established. Based on the observed FE trend in Fig. 2, one can conclude that, regardless of the salt choice, D₂O plays a role in facilitating CO₂RR activity and decreasing DER activity, mainly due to the lower [D⁺] concentration.

It has been proposed that CO production primarily takes place via a carboxylic intermediate (*HOCO) while HER occurs via a hydrogen intermediate (*H) as shown in Fig. S8 (ESI†).^18,19 DER is expected to undergo a similar intermediate (*D) as the HER. The proposed mechanisms imply that the protonation (deuteration) step is involved in both CO₂RR and HER (DER),¹⁸ which in turn suggests that the total current density (J(total)) would be influenced by the solvent choice. As shown in Fig. 3A, in the case that the same cation salt is selected, H₂O-based electrolytes exhibit slightly higher J(total) values than those in D₂O-based electrolytes, confirming that D⁺/H⁺ are involved in the CO₂RR as well as DER/HER. On the other hand, regardless of the isotopic effect, K-based electrolytes show higher J(total) values than those in Na-based electrolytes, which is attributed to the cation effect described above.

Fig. 3 Isotope and cation effects on the CO₂RR. (A) Total current densities, (B and C) partial current density of (B) CO and (C) D₂ or H₂ with different electrolytes. Inset in (C) compares J(CO) values in 0.1 M KHCO₃ (blue) and KDCO₃ (green) at −0.7 and −0.8 V.

Fig. 3B shows the partial current density of D₂ and H₂ evolutions (J(D₂) or J(H₂)) at different potentials. It is noteworthy that, upon replacing H₂O with D₂O, the DER activity is dramatically decreased compared to the HER activity, which is consistent with the FE trend in Fig. 2. In contrast to the decreased DER activity, it is shown that both the cation and isotopic effects influence the CO₂RR activity (Fig. 3C). In particular, the isotopic effect can be clearly observed in J(CO) profiles at −0.8 V and more negative potentials in Na-based electrolytes. This phenomenon can be understood in a way that DER is less likely to take place than HER, again due to the lower [D⁺] than [H⁺], promoting more catalytic active sites for the CO₂RR and consequently enhancing J(CO). On the other hand, in K-based electrolytes, a relatively small enhancement of J(CO) is observed, indicating that the hydrolysis effect of the K ion is more influential on CO₂RR compared to the isotopic effect of D₂O.

It is interesting that J(CO) at low overpotentials in D₂O-based electrolytes is lower than or comparable to that in H₂O-based electrolytes under the same cation salt conditions. For example, at −0.7 V, the J(CO) value in 0.1 M KDCO₃ is lower than that in 0.1 M KHCO₃. However, with increasing overpotential more negatively, the J(CO) values in 0.1 M KDCO₃ get higher than those in 0.1 M KHCO₃ (inset in Fig. 3C). That is, at low overpotentials, higher [H⁺] is likely to facilitate higher CO yield in 0.1 M KHCO₃ since the protonation step is also essential for the CO₂RR. As the overpotential increases and J(CO) consequently increases, the CO yield is limited by the transport of CO₂. The J(CO) in 0.1 M KDCO₃ surpasses that in 0.1 M KHCO₃ in this case because the DER, a main competing reaction, is less favorable than the HER.

In summary, we demonstrate the isotope influence of D₂O on the CO₂RR using a Pd catalyst. In situ X-ray characterization confirms PdD formation, similar to the case of PdH formation in H₂O during the CO₂RR. It is revealed that a simple replacement of H₂O with D₂O can not only decrease the DER rate due to the low D⁺ concentration but also facilitate the CO₂RR. Results from the current study provide a potentially useful methodology to tune the product distribution in other types of CO₂RR reactions.

This research was supported by the US Department of Energy, Office of Basic Energy Sciences, Catalysis Science Program (Grant No. DE-FG02-13ER16381). The authors acknowledge technical support with 7-BM (QAS) at the National Synchrotron Light Source-II (NSLS-II) and Center for Functional Nanomaterials (CFN) in Brookhaven National Laboratory (Contract No. DE-SC0012704) and 17-BM at Advanced Photon Source (APS) in Argonne National Laboratory (Contract No. DE-AC02-06CH11357). J. H. L. acknowledges the National Research Foundation of Korea (NRF) funded by the Ministry of Education (Grant No. NRF-2017R1A6A3A03004202).

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. Y. Hori, Modern Aspects of Electrochemistry, Springer, 2008, ch. 3, vol. 42, pp. 89–189.
2. M. Meinshausen, N. Meinshausen, W. Hare, S. C. Raper, K. Frieler, R. Knutti, D. J. Frame and M. R. Allen, Nature, 2009, 458, 1158–1162.
3. B. M. Tackett, E. Gomez and J. G. Chen, Nat. Catal., 2019, 2, 381–386.
4. C. W. Lee, K. D. Yang, D. H. Nam, J. H. Jang, N. H. Cho, S. W. Im and K. T. Nam, Adv. Mater., 2018, 30, 1704717.
5. W. Zhu, B. M. Tackett, J. G. Chen and F. Jiao, Top. Curr. Chem., 2018, 376, 41.
6. J. H. Lee, S. Kattel, Z. Jiang, Z. Xie, S. Yao, B. M. Tackett, W. Xu, N. S. Marinkovic and J. G. Chen, Nat. Commun., 2019, 10, 3724.
7. W. Sheng, S. Kattel, S. Yao, B. Yan, Z. Liang, C. J. Hawxhurst, Q. Wu and J. G. Chen, Energy Environ. Sci., 2017, 10, 1180–1185.
8. D. Gao, H. Zhou, F. Cai, D. Wang, Y. Hu, B. Jiang, W.-B. Cai, X. Chen, R. Si, F. Yang, S. Miao, J. Wang, G. Wang and X. Bao, Nano Res., 2017, 10, 2181–2191.
9. A. S. Varela, W. Ju, T. Reier and P. Strasser, ACS Catal., 2016, 6, 2136–2144.
10. J. H. Lee, S. Kattel, Z. Xie, B. M. Tackett, J. Wang, C.-J. Liu and J. G. Chen, Adv. Funct. Mater., 2018, 28, 1804762.
11. C. Kim, H. S. Jeon, T. Eom, M. S. Jee, H. Kim, C. M. Friend, B. K. Min and Y. J. Hwang, J. Am. Chem. Soc., 2015, 137, 13844–13850.
12. M. Cho, J. T. Song, S. Back, Y. Jung and J. Oh, ACS Catal., 2018, 8, 1178–1185.
13. Y.-C. Hsieh, S. D. Senanayake, Y. Zhang, W. Xu and D. E. Polyansky, ACS Catal., 2015, 5, 5349–5356.
14. J. Resasco, L. D. Chen, E. Clark, C. Tsai, C. Hahn, T. F. Jaramillo, K. Chan and A. T. Bell, J. Am. Chem. Soc., 2017, 139, 11277–11287.
15. M. R. Singh, Y. Kwon, Y. Lum, J. W. Ager III and A. T. Bell, J. Am. Chem. Soc., 2016, 138, 13006–13012.
16. A. Kreżel and W. Bal, J. Inorg. Biochem., 2004, 98, 161–166.
17. B. Ravel and M. Newville, J. Synchrotron Radiat., 2005, 12, 537–541.
18. W. Zhu, R. Michalsky, O. N. Metin, H. Lv, S. Guo, C. J. Wright, X. Sun, A. A. Peterson and S. Sun, J. Am. Chem. Soc., 2013, 135, 16833–16836.
19. Y. Katayama, F. Nattino, L. Giordano, J. Hwang, R. R. Rao, O. Andreussi, N. Marzari and Y. Shao-Horn, J. Phys. Chem. C, 2019, 123, 5951–5963.

---

Footnote

† Electronic supplementary information (ESI) available: X-ray diffraction pattern, EXAFS results, electrochemical data, and TEM analyses. See DOI: 10.1039/c9cc07611e

---