Accelerated service life test of electrodeposited NiSn alloys as bifunctional catalysts for alkaline water electrolysis under industrial operating conditions

https://doi.org/10.1016/j.jelechem.2017.06.011Get rights and content

Highlights

  • Electrodeposited NiSn alloy coatings were tested as the alkaline water electrolysis catalysts.

  • Accelerated stability tests were performed for oxygen evolution and hydrogen evolution reaction.

  • The cell voltage saving with the NiSn alloy electrodes was ~ 0.435 V compared to Ni electrodes.

Abstract

Electrodeposited NiSn alloy coatings onto Ni 40 mesh substrate were tested for application as cathodes and anodes in the cell for alkaline water electrolysis in 30 wt% KOH at 80 °C. The “accelerated service life test” (ASLT) was performed for the hydrogen evolution reaction (HER), as well as for the oxygen evolution reaction (OER), and compared to that recorded for the Ni coating (Ni-dep) and Ni-mesh for both reactions. The morphology and chemical compositions of the NiSn and Ni coatings were investigated by scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDS), while their surface composition was investigated by X-ray photoelectron spectroscopy (XPS) before and after the ASLT for both reactions, respectively. By measuring the potential at j = 0.3 A cm 2 it was shown that during the ASLT the NiSn alloy coating catalytic activity for the HER decreases (about 24 mV after 25 cycles), while the catalytic activity for the OER increases (about 50 mV after 25 cycles), so that the cell voltage decreases for about 26 mV. The Ni-dep and Ni-mesh electrodes catalytic activity was found to increase for the HER (for about 103 mV), as well as for the OER (for about 52 mV) during the ASLT. Hence, the cell voltage for the Ni-dep and Ni-mesh electrodes decreased from 2.402 V to 2.245 V during the ASLT, while that for the NiSn electrode decreased from 1.967 V to 1.941 V. The cell voltage saving with the NiSn electrodes amounts to about 435 mV before the ASLT and about 304 mV after the ASLT. SEM results showed that no changes in the morphology of as prepared samples could be detected after the ASLTs for both reactions. EDS analysis confirmed that some changes occurred during the ASLT, particularly for the oxygen content in the surface layer. Similar conclusions were made from the XPS analysis.

Introduction

In the review paper of Santos et al. [1], among other basics of the process, a detailed history of the hydrogen and oxygen production by water electrolysis was given. Discovery of the phenomenon of electrolytic splitting of water into hydrogen and oxygen is known for > 200 years [2], [3], while the technique for the industrial production of hydrogen and oxygen in alkaline solutions was developed by Dmitry Lachinov in 1888 and during the next 16 years > 400 industrial water electrolyzers were already in operation [4]. Although water electrolysis has the advantage of producing extremely pure hydrogen, its applications are often limited to the small scale and unique situations, such as marine, rockets, spacecrafts, electronic industry, food industry as well as medical applications. Nowadays only 4% of the total world hydrogen production is accomplished by the alkaline water electrolysis [5], [6].

The current density used in the industrial plants varies from 1 to 3 A cm 2, at a temperature between 70 and 90 °C in 25–30 wt% KOH [7] and here are presented only the references where the investigations were performed under these conditions.

The electrode materials should satisfy following demands: good corrosion resistance, high conductivity, high catalytic effect, and low price [8]. Ni was found to be an electroactive cathode and anode material with good corrosion resistance in the alkaline solutions [9], [10], [11].

In the review paper of Zeng et al. [12] a more detailed overview was given for: different industrial systems for the alkaline water electrolysis [7], [13], [14] (Tables 5 and 6 of Ref. [12] respectively); kinetic parameters for the HER and the OER at different metals (Tables 1 and 2 of Ref. [12], [15], [16]) and different Ni based alloys (Table 8 of Ref. [12], [17], [18], [19]); overpotential for the OER at different electrode materials (Table 9 of Ref. [12], [20]); overpotential for the HER at different electrode materials (Table 10 of Ref. [12], [21], [22], [23], [24], [25]).

In Table 5 of Ref. [12] are compared cell operating conditions for the alkaline water electrolysis in five companies [13]: De Nora S.A.P., Norsk Hydro, Electrolyzer Corp. Ltd., Teledyne Energy systems and General Electric. Only Electrolyzer Corp. Ltd. uses monopolar cell type, while all others are using bipolar cell type. The PTFE bonded noble metals are cathodes and anodes in the cells of General Electric (operating at 5 A cm 2 with the cell voltage of 1.7 V), while others are using either Ni-coated steel, or activated Ni-coated steel. The operation current density varies from 1.34 A cm 2 to 2 A cm 2 with the operating cell voltage varying from 1.75 V to 1.9 V. In Ref. [14] (Table 6 of Ref. [12]) are given characteristics of two types of commercialized electrolyzers: a monopolar alkaline electrolyzer and a PEM electrolyzer/cell. The PEM electrolyzer/cell operates at 4 times higher current density producing 10 times lower amount of hydrogen and oxygen in comparison with that in the monopolar alkaline electrolyzer.

Annealed Ni foils coated with 0.04 μm thick Fe layer using magnetron sputtering were found to be more stable during the HER. A thin Fe coating on a Ni cathode prevents its deactivation during alkaline water electrolysis by preventing Ni-hydride formation [9]. In Ref. [10] it was shown that the overpotential for the HER at the Ni electrode increases significantly with the time of the HER at the cathodic current densities higher than − 0.1 A cm 2. Such effect was ascribed to the hydride formation at active Ni cathode surfaces. After applying high cathodic current densities (− 0.3 to − 1.0 A cm 2) for > 2 h in 30 wt% KOH at 80 °C the Ni electrodes were taken from the solution, washed, dried and after 20 min. Their surfaces were analyzed by X-ray diffraction, showing the presence of two small peaks corresponding to (111) NiH and (200) NiH.

The HER on the polished Ni cathodes was investigated in 30 wt% KOH at 70 °C in the presence of 5 ppm dissolved Fe [17]. The deactivation (increase of cathodic potential) of the Ni was attributed to the penetration of atomic hydrogen in the metal lattice by the diffusion mechanism, being more pronounced at high cathodic current densities. Fe deposits observed at a prolonged hydrogen evolution were found to have detrimental effect on the HER [17].

The HER was investigated on different Ni-based alloys [18], the Zr-based alloys [18], the nanocrystalline fcc NixMo(1  x) metallic powders [19], adherent Ni-Fe-Mo-(Zn) coatings [21], electrodeposited amorphous Ni-S-Co alloy [22], the hydrogen storage alloys [23], and the Ni-Mo alloy coating [24]. Several binary composites [25], Ni-Mo, Ni-Zn, Ni-Co, Ni-W, Ni-Cr and Ni-Fe, electrodeposited onto mild steel substrate were investigated in 6 M KOH at 80 °C as cathodes for the water electrolysis. Their activity was found to follow the order: Ni-Mo > Ni-Zn > Ni-Co > Ni-W > Ni-Fe > Ni-Cr, while all binary composites showed significantly higher catalytic activity in comparison with the mild steel cathodes.

High catalytic activity for the HER in alkaline solutions on the electrodeposited NiSn alloys was discovered in nineties [27], [28]. In our previous work [29], [30], [31], [32] the NiSn alloy coatings were electrodeposited onto Ni plates and Ni 40 meshes from the pyrophosphate-glycine bath containing the same concentrations of Sn2 + and Ni2 + ions (0.1 M) at different current densities and their morphology, chemical and phase compositions were investigated. Four crystalline phases of low crystallinity were detected in total: face centered cubic (fcc) Ni phase, hexagonal close packed (hcp) Ni3Sn phase, hexagonal Ni(1 + x)Sn (0 < x < 0.5) phase, adopting NiAs type structure [29], [30] and monoclinic Ni3Sn4 phase with CoSn type structure, with Ni(1 + x)Sn (0 < x < 0.5) phase being dominant in samples electrodeposited at the cathodic current densities higher than − 20 mA cm 2. The catalytic activity for the HER in alkaline solutions was shown to be the consequence of the presence of high amount (about 80%) of the hexagonal Ni(1 + x)Sn (0 < x < 0.5) phase and of the high roughness of these NiSn coatings.

The influence of surface finishing (sandblasting and/or chemical pickling) of the Ni on the OER was investigated in Ref. [11]. It was concluded that the presence of the NiO2 on the surface of sandblasted/pickled electrodes inhibits their electrocatalytic properties for the OER. The influence of solution temperature for the HER and the OER were investigated at the Ni electrode in 50 wt% KOH [16] (kinetic parameters are also given in Table 2 of Ref. [12]). It was concluded that the temperature effect is more pronounced for the OER. The OER was investigated at different metal oxides prepared by thermal decomposition at the Ti substrate [15]. Their activity was found to follow the order: Ru > Ir ~ Pt ~ Rh ~ Pd ~ Ni ~ Os  Co  Fe (In Table 2 of [15] are given kinetic parameters – jo and Tafel slope – for Pt, Co and Fe). The OER was investigated at a mixture NiO/CoO [20], and at electrodeposited Ni-Fe alloys [26]. It was also shown in our previous work that the NiSn coatings could be good catalysts for the OER in 1 M NaOH at room temperature [33]. These coatings showed high stability during the extended oxidation at j = 0.3 A cm 2 in comparison with the Ni electrode which underwent to additional polarization under such condition. Hence, it appeared that the same NiSn coating could be used as a cathode for the HER and an anode for the OER in alkaline solution with low overpotential for both reactions.

For the application as cathodes and anodes in the industrial process of water electrolysis both electrodes must fulfill the most important factor, low overpotential at the high current densities (0.3–0.6 A cm 2) for both reactions, hydrogen and oxygen evolution. The next, very important property, is high stability of the electrodes in industrial cells. In the case of chlor-alkali electrolysis the loss of catalytic activity and stability of cathodes during long term operation is the result of the so-called polarity inversion of the electrodes, which takes place during the replacement of old electrodes of an electrolyzer with the new ones in the zero-gap cells. During this operation in one set of the cells, the anodes and cathodes of the rest of the cells are short-circuited, causing a reverse current flow which may damage the cathodes and negatively affect their activity for the HER [34]. The manufacturers can predict how often in a certain period of time such operation should be performed and, in accordance with that, design appropriate ASLT for cathodes, as it was the case for cathodes used in the industrial chlor-alkali electrolysis [32], [34], [35], [36]. The procedure of the ASLT is based on a sequence of galvanostatic polarizations in the HER range and cyclic voltammetries (CVs) in a wide potential range between the potential of hydrogen evolution and the potential of oxygen evolution (reproducing the conditions of polarity inversion [34]). Using the same procedure for the ASLT in our previous works [35], [36] the Ni-MoO2 cathodes, as well as the NiSn cathodes [32], were tested and compared with that obtained for the commercial De Nora's (Ni-RuO2) cathode. Such ASLT could be applied at the electrodes for the industrial water electrolysis, since it represents the most severe conditions of their long term operation.

Recently [37] similar ASLT for the HER was applied on the bilayer Ni/Cu porous cathodes under the conditions of industrial water electrolysis (j =  0.3 A cm 2, 30 wt% KOH at 80 °C). Two types of ASLT were performed: one designed by the authors (1ASLT) and another one taken from already published works [32], [34], [35], [36] (2ASLT). The cathode with the lowest amount of the electrodeposited Ni (33.41% of Cu) showed the highest catalytic activity for the HER, “mainly due to the enhancement of the real electrochemical surface area.” Anodic potential of 0.75 V caused significant damage of the cathodes, so that the authors concluded that “cathodes for alkaline water electrolysis based on the Ni must be subjected to potentials below 0.75 V under the conditions of inversion polarization.”

Since our previous results [29], [30], [31], [32] confirmed that the NiSn coating could be promising cathode, as well as anode [33], in alkaline solutions, in this work the ASLT of the NiSn coating, Ni coating and Ni 40 mesh were tested and compared for the HER and the OER under the conditions of industrial water electrolysis.

Section snippets

Experimental

All solutions were made from p.a. chemicals (Aldrich-Sigma) in extra pure (18.2 MΩ) UV water (Smart2, Pure UV, TKA).

Electrochemical characterization of the NiSn coatings electrodeposition

In this work lower concentration of the SnCl2 (0.03 M) was used since it was discovered that the solutions used in our previous works [29], [30], [31], [32], [33] were not stable for > 2–3 weeks and the precipitation of Sn-hydroxide was observed. The solution used in this work was stable (no precipitate was observed) for > 8 months. The polarization characteristic for NiSn electrodeposition onto Ni 40 mesh in 0.6 M K4P2O7 + 0.1 M NiCl2·6H2O + 0.03 M SnCl2·2H2O + 0.3 M NH2CH2COOH at 25 °C is shown in Fig. 1a.

Conclusions

Electrodeposited NiSn alloy coatings were tested as hydrogen and oxygen evolution catalysts for alkaline water electrolysis under industrial operating conditions, in 30 wt% KOH at 80 °C. The ASLT was performed for the HER, as well as for the OER, and compared to that recorded for the Ni coating (Ni-dep) and Ni-mesh for both reactions. It was shown that at j = 0.3 A cm 2 during the ASLT the NiSn alloy coating catalytic activity for the HER decreases (about 24 mV after 25 cycles), while the catalytic

Acknowledgement

The authors are indebted to the Ministry of Education, Science and Technological Development of the Republic of Serbia (Project No. 172054) for the financial support of this work. The authors would like to acknowledge the COST MP 1407 action for networking support. V.D. Jović thanks to Juan Feliu, Phil Bartlett, Jacek Lipkowski and David Schiffrin for invitation to contribute in a special issue of the Journal of Electroanalytical Chemistry in honor of Roger Parsons.

References (59)

  • D.L. Stojic et al.

    Sources

    (2003)
  • W. Kreuter et al.

    Electrolysis: the important energy transformer in a world of sustainable energy

    Int. J. Hydrog. Energy

    (1998)
  • S. Dunn

    Int. J. Hydrog. Energy

    (2002)
  • R.F. de Souza et al.

    J. Power Sources

    (2007)
  • A. Marshall et al.

    Energy

    (2007)
  • A.E. Mauer et al.

    Electrochim. Acta

    (2007)
  • C. Bocca et al.

    Int. J. Hydrog. Energy

    (1998)
  • K. Zeng et al.

    Prog. Energy Combust. Sci.

    (2010)
  • J.Y. Huot et al.

    Int. J. Hydrog. Energy

    (1987)
  • J.Y. Huor et al.

    Int. J. Hydrog. Energy

    (1989)
  • H. Wendt et al.

    Mater. Chem. Phys.

    (1989)
  • F.C. Crnkovic et al.

    Int. J. Hydrog. Energy

    (2004)
  • Q. Han et al.

    Int. J. Hydrog. Energy

    (2003)
  • W. Hu

    Int. J. Hydrog. Energy

    (2000)
  • W. Hu et al.

    Int. J. Hydrog. Energy

    (1998)
  • E. Potvin et al.

    Mater. Chem. Phys.

    (1992)
  • M.B.F. Santos et al.

    Electrochim. Acta

    (1992)
  • V.D. Jović et al.

    Int. J. Hydrog. Energy

    (2012)
  • B.M. Jović et al.

    Electrochim. Acta

    (2013)
  • B.M. Jović et al.

    Int. J. Hydrog. Energy

    (2014)
  • B.M. Jović et al.

    J. Electroanal. Chem.

    (2015)
  • A.L. Antozzi et al.

    Electrochim. Acta

    (2008)
  • V.D. Jović et al.

    Electrochim. Acta

    (2012)
  • C. Valero-Vidal et al.

    Appl. Catal. B Environ.

    (2016)
  • A. Lasia et al.

    J. Electroanal. Chem.

    (1990)
  • N. Krstajić et al.

    J. Electroanal. Chem.

    (2001)
  • H. Bode et al.

    Electrochim. Acta

    (1966)
  • F. Hahn et al.

    Electrochim. Acta

    (1986)
  • M.C. Biesinger et al.

    Appl. Surf. Sci.

    (2011)
  • Cited by (0)

    View full text