Photoactive Zn–air batteries using spinel-type cobalt oxide as a bifunctional photocatalyst at the air cathode

Chanikarn Tomon , Sangchai Sarawutanukul , Salatan Duangdangchote , Atiweena Krittayavathananon and Montree Sawangphruk *
Centre of Excellence for Energy Storage Technology (CEST), Department of Chemical and Biomolecular Engineering, School of Energy Science and Engineering, Vidyasirimedhi Institute of Science and Technology, Rayong 21210, Thailand. E-mail: montree.s@vistec.ac.th

Received 7th March 2019 , Accepted 15th April 2019

First published on 16th April 2019


Spinel-type cobalt oxide (Co3O4) was synthesized and used as a photoactive bifunctional electrocatalyst towards the oxygen evolution reaction (OER) and the oxygen reduction reaction (ORR) at the air cathode of zinc–air batteries (ZABs). The Co3O4 having direct and indirect band gap energies of ca. 2.20 eV and ca. 1.35 eV can absorb visible light, generating photogenerated carriers and photoelectrons via the photoelectric effect. Upon exposure to visible light, the Co3O4 electrode exhibits ca. 30% higher current density than that under dark conditions and provides around 10–20% lower OER and ORR overpotentials than those under the dark conditions. Under visible light, the specific capacity of the as-fabricated photoactive ZAB cell is improved by ca. 10% as compared to that under dark conditions.


To date, rechargeable metal–air batteries have become promising energy storage devices owing to their high theoretical energy densities and environmentally friendly nature.1 Among various types of metal–air batteries, zinc–air batteries (ZABs) provide a high theoretical specific capacity of 820 A h kg−1 and a high theoretical specific energy of 1086 W h kg−1, which is ca. 5-fold higher than that of Li-ion batteries (200–250 W h kg−1).2 Although previous efforts were dedicated towards developing rechargeable ZABs,3–6 their performances are still far from the theoretical value. This is because they are facing many problems.7,8 One of the main problems found in ZABs and other metal–air batteries is the sluggish nature of reaction kinetics towards the oxygen evolution reaction (OER) and the oxygen reduction reaction (ORR) at the air cathode.9 The poor reactivity of the cathode leads to a large potential gap during the charge/discharge process, decreases energy efficiency, and decreases the cycling stability of the battery. Here, in this work, we proposed a new concept of using a photoactive bifunctional electrocatalyst towards the OER and ORR, leading to faster kinetic reactions.

Cobalt oxide, especially spinel-type Co3O4, is an attractive non-precious metal oxide providing a very substantial advance in terms of efficiency as both OER and ORR catalysts.10–12 Its structure typically contains two electronically mixed valences Co2+ and Co3+ placed at the tetrahedral and octahedral sites, respectively, on the close-packed face centred cubic (fcc) lattice formed by the oxide ions. Owing to the mixed valence of the cobalt, Co3O4 becomes an excellent p-type semiconductor, which exhibits highly efficient photocatalytic activity. Typically, Co3O4 has two optical band gaps in the range of 1.5–2.2 eV. They cover the range of visible light (λ ≥ 400 nm), which is ca. 44% of the solar radiation.13 To achieve high capacity ZABs with free photoenergy gain, herein, we introduce a new concept of ZABs using Co3O4 as a photocatalyst at the air cathode. Under light illumination, the Co3O4 is excited by absorbing photon energy () from the light irradiation, generating photoelectrons and simultaneously electron holes (photogenerated carriers, Co3O4*) in the valence band. In principle, the generated Co3O4* is a highly active species or a super-oxidizing agent. The presence of the Co3O4* can also enhance the catalyst activity by (i) reducing the energy barrier and (ii) enhancing the catalytic activity. When increasing the light intensity, the potential drop across the space charge is reduced, resulting in the reduction of the energy barrier of the chemical reaction.14 Besides, the highly active Co3O4* can move along the surface and react with the OH and H2O to form CoOOH and CoO2 as shown in the following reactions (1 and 2):15,16

 
Co3O4* + OH + H2O ↔ 3CoOOH + e(1)
 
3CoOOH + OH ↔ CoO2 + e + H2O(2)
The cell configuration used in this work is a two-electrode system in a quartz container as shown in the schematic in Fig. 1. One side is a quartz window which is placed in between a photoactive cathode having the spinel-type Co3O4 catalyst and a light source. The Co3O4 catalyst was synthesized via a two-step synthesis through a hydrothermal method followed by calcination. More details can be found in the experimental section (see the ESI). SEM and TEM images (Fig. 1(a)) show nanoparticles with a particle size of ca. 25 nm which interconnect along the microrod structures. Its N2 adsorption/desorption isotherm (Fig. S2, ESI) shows an IV-type adsorption isotherm with a H3-type hysteresis loop, which is characteristic of the mesoporous structure.17 The BET surface area is 12.99 m2 g−1, which correlates with a previous publication. The XRD pattern (Fig. 1c) shows the characteristic peaks of cubic spinel-type Co3O4 in the Fd3m space group (JCPDS card no. 43-1003). No characteristic peak of other impurity phases has been detected. According to Bragg's law, its interlayer spacing (d-spacing) is 0.24 nm.


image file: c9cc01876j-f1.tif
Fig. 1 Schematic of a photoactive Zn–air battery setup using spinel-type Co3O4 as an air cathode upon visible light illumination (wavelength (λ) = 400–700 nm) on the left, and the physical characterization of the Co3O4 cathode, i.e., (a) SEM image, (b) TEM image, and (c) XRD pattern.

The optical and electrical properties of the Co3O4 sample were investigated by UV-visible spectroscopy, ultraviolet photoelectron spectroscopy (UPS), and DFT calculations. Fig. 2a shows an absorption spectrum of the Co3O4 sample consisting of two adsorption bands at ca. 440 and 755 nm. The first adsorption band can be assigned to the O2− → Co3+ charge transfer process, whereas the second band relates to the O2− → Co2+ transition.18 The adsorption band gap energy (Eg) can be determined by extrapolating the linear region in a plot of (αhν)nvs. photon energy as shown in Fig. 2b.16,19 Two adsorption bands provide direct and indirect band gap energies, which are ca. 2.20 eV and ca. 1.35 eV, respectively. The values are in reasonable agreement with the DFT calculations (see Fig. S3, ESI) based on the PBE+U approach, where the direct and indirect band gap energies are calculated as 2.24 and 1.36 eV, respectively.


image file: c9cc01876j-f2.tif
Fig. 2 (a) UV-visible spectrum of Co3O4, (b) Tauc plot between photon energy (eV) and (αhv)n, (c) (photoelectron yield)0.5 as a function of photon energy of Co3O4 and (d) photoluminescence spectrum.

UPS was used to determine the valence band (VB) of the Co3O4 sample by emitting ultraviolet photons to bombard the surface of the sample. The VB edge can be determined from the plot of (photoelectron emission yield)0.5vs. photon energy (Fig. 2c). The VB edge of the as-prepared Co3O4 sample is ca. −5.68 eV, while the VB of ITO glass is ca. −4.7 eV. As a result, the energy level of the Co3O4 sample can be summarized as shown in the inset of Fig. 2c. The energy level combines the 3d orbitals of Co2+ and Co3+, which is characteristic of the Co3O4 electronic structure.16 The 2p(O2−) → 3d(Co2+) transition state corresponds to an indirect band gap of 1.35 eV, while the direct band gap is 2.20 eV, corresponding to the 2p(O2−) → 3d(Co3+) transition state. According to this result, a photon energy of 1.35 eV is the minimum energy which is used to excite an electron from the valence band to the conduction band.

To confirm that the Co3O4 is a photoactive material, fluorescence spectroscopy (FLS) was used to measure the fluorescence emission response of the Co3O4 sample (Fig. 2d). The excitation wavelength is 420 nm. The fluorescence emission spectrum indicates a single peak at 575 nm, resulting in the relaxation of the excited electron from 3d(Co3+) to 2p(O2−) as shown in the inset of Fig. 2d. Moreover, the photoelectrochemical properties of Co3O4 towards the OER were studied using a full solar simulator (AM 1.5G/100 mW cm−2) equipped with a motorized monochromator (Oriel Cornerstone 130 1/8 m)20 (Fig. S4, ESI). The visible light provides photocurrent density with an incident-photon-to-current-efficiency (IPCE) of ca. 60.0% at an applied potential of 1.55 V vs. RHE.

The electrochemical photocatalytic activity of the as-prepared Co3O4 material in a basic solution (0.1 M KOH) was investigated by cyclic voltammetry (CV) and linear sweep voltammetry (LSV) by using a three-electrode system under light illumination and dark conditions. The as-prepared Co3O4 material was coated on an ITO electrode with a mass loading of 2 mg cm−2, which was used as a working electrode. A platinum rod and a saturated calomel electrode were used as counter and reference electrodes, respectively. The CV response was recorded at a scan rate of 20 mV s−1 (see Fig. 3a). As expected, the ITO itself shows small current response (Fig. S5, ESI). In comparison between the Pt and graphite rod counter electrodes (Fig. S6, ESI), the CV curves of both cases do not show any significant difference. This indicates that the Pt counter electrode does not affect the electrochemical performance. The observed current response is dominated by the Co3O4 active material. The oxidation peak at ca. 1.5 V vs. RHE is observed during the forward scan, and the reaction consecutively reverses corresponding to the cathodic peak (ca. 1.2 V vs. RHE) in the backward scan. In nature, the electrochemical reaction of the cobalt oxide involves multiple reversible redox processes of Co2+/Co3+/Co4+. These two reversible processes are overlapped and merged into a redox peak as revealed in the CV.21 In a real system, the actual oxidation number of the Co atom on the electrode during the electrochemical reaction was measured using ex situ XANES (Fig. 3d). The average oxidation number of the Co3O4 sample positively shifted from ca. +2.60 to ca. +3.24 eV during the forward scan from 0 to 1.55 V vs. RHE, and nearly levelled off even the increasing applied potential.


image file: c9cc01876j-f3.tif
Fig. 3 (a) CVs of Co3O4/ITO under light illumination and dark conditions, (b) schematic of the Co3O4 electrode being excited with visible light, (c) LSVs of the OER for the Co3O4 electrode, (d) plot between the applied potential and the oxidation number of the Co3O4 electrode during the OER under light illumination and dark conditions, (e) LSVs of the ORR for the Co3O4 electrode, (f) plot between the applied potential and the oxidation number of the Co3O4 electrode during the ORR under light illumination and dark conditions in 0.1 M KOH at a scan rate of 20 mV s−1.

The CV was also used together with a differential electrochemical mass spectrometry (DEMS) to detect the generated oxygen gas due to the OER (Fig. S7, ESI). The ionic current corresponding to the O2 (m/z = 32) detected by DEMS is observed at a potential higher than 1.55 V vs. RHE, which is near the onset potential of the OER. XANES indicates that ca. Co3.24+–Co3.45+ is the OER-active state in this work. Therefore, the observed current at a high potential (>1.55 V vs. RHE) mainly corresponds to the OER (see reaction (3)).22

 
4OH ↔ O2 + 2H2O + 4e(3)
We then further studied the effect of visible light illumination. Under the open circuit voltage (OCV), the average oxidation number of Co in the electrode in 0.1 M KOH solution, according to XANES (Fig. 3d) shifted by +0.2 (from ca. +2.6 to ca. +2.8) after the electrode was illuminated with visible light for 2 h. This indicates that the photoenergy can induce transfer of charge carriers on the electrode surface. During the electrochemical process, the light illumination has greatly influenced the OER and ORR activities and the onset potential. The current responses for the OER and ORR after subtracting the capacitive component from the overall measured current are shown in Fig. 3c and e. Under the light illumination, the anodic current density increases up to ca. 30.0% when compared with that under the dark conditions. Regarding OER activity, the Co3O4 electrode under the light illumination shows an onset potential of 1.52 V vs. RHE, which is 0.07 vs. RHE lower than that of the Co3O4 electrode under the dark conditions. Tafel plots (inset image of Fig. 3c) reveal that Co3O4 under the light illumination has a Tafel value of 74.1 mV dec−1, which is smaller than that of the electrode under the dark conditions (109.3 mV dec−1). The process with the Tafel values in the range of 60–120 mV dec−1 usually involves 2–4 electron transfers, which correlates with the XANES result. Furthermore, the activity for the OER is usually judged by the potential required to oxidize water at a current density of 10 mA cm−2. The Co3O4 electrode under light requires an overpotential of 420 mV based on the equilibrium potential of 1.23 V vs. RHE, and is more favourably active than other cobalt based electrodes (see Table S2, ESI).23,24 The excellent ORR activity of the catalyst under the light illumination was further confirmed by a smaller Tafel slope of 92.6 mV dec−1 at a low onset potential of 0.74 V vs. RHE than that under the dark conditions (102.2 mV dec−1) (Fig. 3e).

To clarify the origin of the catalytic improvement under light illumination, the XAS technique was used to observe Co in Co3O4 after the OER and ORR in the presence and absence of light illumination (Fig. S8, ESI). Like in the absence of light, the oxidation number of Co under the light illumination also increased when applying the potential up to 1.55 V vs. RHE (∼the onset potential of the OER) and remained nearly constant. In the presence of light, the oxidation number increased with a slope of 0.59, which is higher than that under the dark conditions (0.38).

To fabricate a photoactive ZAB, we used a two-electrode system (Fig. 1) consisting of the Co3O4 material coated on a transparent ITO electrode as the air cathode and a Zn plate as the anode. The distance between the two electrodes is ca. 1 cm. The initial charge/discharge profile at a current density of 2 mA cm−2 is shown in Fig. 4a. An open circuit voltage (OCV) of the battery is ca. 1.41 V vs. Zn, which is close to the theoretical value of the equivalent potential between Zn and O2 (1.65 V vs. Zn).25 The difference here is due to the potential drop along the distance between the Co3O4 cathode and the Zn anode. During the charging process, the oxidation reaction of Co3O4 with OH to COOH and CoO2 follows the reactions (1) and (2) (see Fig. 4a).22 The discharging process is its reversible process reducing oxygen atoms from the CoO2 structure together with H2O to produce Co3O4 and OH – namely the ORR process.26 The charge/discharge potential gap of the ZAB under light illumination is ca. 0.82 V vs. Zn, which is lower than that under dark conditions around 0.19 V vs. Zn. At a current density of 20 mA cm−2 (Fig. 4b), the specific discharge capacity under the visible light is 769 mA h gzn−1, which shifts closer to the theoretical value of the ZAB (820 mA h gzn−1).8 The battery in the absence and presence of light demonstrates high cycling stability without an obvious voltage change during the whole 70 h (Fig. 4c). This is because the Co3O4 electrode is excited by absorbing visible light, generating the photoelectrons in the conduction band and simultaneously photogenerated carriers or holes (Co3O4*) in the valence band. The highly active photogenerated carriers can easily react with OH at the solid–liquid interface, enhancing the photocatalytic evolution of O2.


image file: c9cc01876j-f4.tif
Fig. 4 (a) Initial charge/discharge profile of the Zn–air battery, (b) the specific capacity at 20 mA cm−2, and (c) long-term stability of the Zn–air battery under light illumination and dark conditions.

In summary, we introduced photoactive Co3O4 nanostructured microrods as a bifunctional photo-electrocatalyst at the air cathode of ZABs. Photoenergy can enhance the electrocatalyst activity by reducing the energy barrier towards the OER/ORR processes at the cathode. The optical absorption of the Co3O4 electrode shows direct and indirect optical bandgaps of ca. 2.20 eV and ca. 1.35 eV, respectively, which cover the visible light region (1.8–3.1 eV). Under light illumination, the anodic current density of Co3O4 towards the OER/ORR processes increases up to ca. 30% when compared with that under dark conditions. This is due to the photogenerated charge carriers (electron holes) on the electrode surface; the generated Co3O4* is known as a super oxidizing agent or super active agent, which can rapidly react with OH and H2O. The photoenergy also reduces the space charge region at the solid–liquid interface that can reduce the overpotentials of the OER (ca. 100 mV at 10 mA cm−2) and ORR (ca. 110 mV at −0.7 mA cm−2) processes. Under visible light, the specific capacity of a ZAB cell increases ca. 8–10% as compared to that under dark conditions. This finding may lead to an ideal energy conversion and storage device for renewable energy applications.

This work was financially supported by the Thailand Research Fund and Vidyasirimedhi Institute of Science and Technology (RSA6180031) as well as the Energy Policy and Planning Office (EPPO), Ministry of Energy, Thailand. The authors would like to thank the Synchrotron Light Research Institute (Thailand) BL-8 for XAS beam time and NANOTEC (NSTDA, Thailand) through its program of Research Network NANOTEC (RTA6080005). C. T. acknowledge Assist. Prof. Dr Taweesak Sudyoasuk and Dr Saran Kalasina for the assistance of the IPCE measurement.

Conflicts of interest

There are no conflicts to declare.

Notes and references

  1. X. Yu and A. Manthiram, ACS Energy Lett., 2017, 2, 1050–1055 CrossRef.
  2. J. Pan, Y. Y. Xu, H. Yang, Z. Dong, H. Liu and B. Y. Xia, Adv. Sci., 2018, 5, 1700691 CrossRef PubMed.
  3. X. Liu, L. Wang, P. Yu, C. Tian, F. Sun, J. Ma, W. Li and H. Fu, Angew. Chem., 2018, 57, 16166–16170 CrossRef CAS PubMed.
  4. J. T. Li, H. Y. Liu, M. M. Wang, C. Lin, W. Yang, J. S. Meng, Y. A. Xu, K. A. Owusu, B. L. Jiang, C. X. Chen, D. N. Fan, L. Zhou and L. Q. Mai, Chem. Commun., 2019, 55, 334–337 RSC.
  5. H. Cheng, J. M. Chen, Q. J. Li, C. Y. Su, A. N. Chen, J. X. Zhang, Z. Q. Liu and Y. X. Tong, Chem. Commun., 2017, 53, 11596–11599 RSC.
  6. C. Zhu, Z. Yin, W. Lai, Y. Sun, L. Liu, X. Zhang, Y. Chen and S.-L. Chou, Adv. Energy Mater., 2018, 8, 1802327 CrossRef.
  7. J. Zhang, Z. Zhao, Z. Xia and L. Dai, Nat. Nanotechnol., 2015, 10, 444 CrossRef CAS PubMed.
  8. J.-S. Lee, S. Tai Kim, R. Cao, N.-S. Choi, M. Liu, K. T. Lee and J. Cho, Adv. Energy Mater., 2011, 1, 34–50 CrossRef CAS.
  9. Z.-L. Wang, D. Xu, J.-J. Xu and X.-B. Zhang, Chem. Soc. Rev., 2014, 43, 7746–7786 RSC.
  10. C. P. Plaisance and R. A. van Santen, J. Am. Chem. Soc., 2015, 137, 14660–14672 CrossRef CAS PubMed.
  11. J. Xu, P. Gao and T. S. Zhao, Energy Environ. Sci., 2012, 5, 5333–5339 RSC.
  12. S. Dou, X. Y. Li, L. Tao, J. Huo and S. Y. Wang, Chem. Commun., 2016, 52, 9727–9730 RSC.
  13. J. Yuan, J. Wen, Q. Gao, S. Chen, J. Li, X. Li and Y. Fang, Dalton Trans., 2015, 44, 1680–1689 RSC.
  14. R. Beranek, Adv. Phys. Chem., 2011, 2011, 20 Search PubMed.
  15. K. Takanabe, ACS Catal., 2017, 7, 8006–8022 CrossRef CAS.
  16. S. Kalasina, N. Phattharasupakun, M. Suksomboon, K. Kongsawatvoragul and M. Sawangphruk, Electrochim. Acta, 2018, 283, 1125–1133 CrossRef CAS.
  17. N. Venugopal, A. K. Pullur, W.-S. Kim and H. P. J. C. L. Ha, Catal. Lett., 2014, 144, 2151–2156 CrossRef CAS.
  18. P. H. T. Ngamou and N. Bahlawane, Chem. Mater., 2010, 22, 4158–4165 CrossRef CAS.
  19. N. A. M. Barakat, M. S. Khil, F. A. Sheikh and H. Y. Kim, J. Phys. Chem. C, 2008, 112, 12225–12233 CrossRef CAS.
  20. G. Dong, H. Hu, X. Huang, Y. Zhang and Y. Bi, J. Mater. Chem. A, 2018, 6, 21003–21009 RSC.
  21. R. P. Raj, P. Ragupathy and S. Mohan, J. Mater. Chem. A, 2015, 3, 24338–24348 RSC.
  22. P. Tan, B. Chen, H. Xu, W. Cai, W. He, M. Liu, Z. Shao and M. Ni, Small, 2018, 14, 1800225 CrossRef PubMed.
  23. G. Du, X. Liu, Y. Zong, T. S. A. Hor, A. Yu and Z. Liu, Nanoscale, 2013, 5, 4657–4661 RSC.
  24. M. Al-Mamun, Y. Wang, P. Liu, Y. L. Zhong, H. Yin, X. Su, H. Zhang, H. Yang, D. Wang, Z. Tang and H. Zhao, J. Mater. Chem. A, 2016, 4, 18314–18321 RSC.
  25. Y. Li and H. Dai, Chem. Soc. Rev., 2014, 43, 5257–5275 RSC.
  26. V. G. Anju, R. Manjunatha, P. M. Austeria and S. Sampath, J. Mater. Chem. A, 2016, 4, 5258–5264 RSC.

Footnote

Electronic supplementary information (ESI) available: Experimental details, supporting figures and tables, and calculations. See DOI: 10.1039/c9cc01876j

This journal is © The Royal Society of Chemistry 2019