Controllably Alloyed, Low Density, Free-standing Ni-Co and Ni-Graphene Sponges for Electrocatalytic Water Splitting

Synthesis of low cost, durable and efficient electrocatalysts that support oxygen evolution reaction (OER) and hydrogen evolution reaction (HER) are the bottlenecks in water electrolysis. Here we propose a strategy for the development of controllably alloyed, porous, and low density nickel (Ni) and cobalt (Co) based alloys - whose electrocatalytic properties can be tuned to make them multifunctional. Ni and Co based alloy with the chemical structure of Ni1Co2 is identified as an efficient OER catalyst among other stoichiometric structures in terms of over potential @ 10 mAcm−2 (1.629 V), stability, low tafel slope (87.3 mV/dec), and high Faradaic efficiency (92%), and its OER performance is also found to be on par with the benchmarked IrO2. Tunability in the porous metal synthesis strategy allowed the incorporation of graphene during the Ni sponge formation, and the Ni- incorporated nitrogen doped graphene sponge (Ni-NG) is found to have very high HER activity. A water electrolysis cell fabricated and demonstrated with these freestanding electrodes is found to have high stability (>10 hours) and large current density (10 mAcm−2 @ 1.6 V), opening new avenues in the design and development of cost effective and light weight energy devices.

Supporting Information Video S1 (attached) Figure S1. XRF spectrum ofNi and Ni 1 Co 2 alloy. The column shows the percentage of Ni and Co in the different types of alloy sponges.

Calculation of Faradaic efficiency from RRDE voltammetry
In order to calculate Faradaic efficiency, the collection efficiency of the ring electrode (Pt) needs to be determined. RRDE experiments were carried out using a glassy carbon disk-Pt ring RRDE electrode connected with BioLogic work station.

RRDE-Collection efficiency calculation:
In a typical Rotating Ring Disc Electrode (RRDE) collection efficiency experiment, the complete product generated at the disk electrode will not reach the ring electrode. The percentage of material which is collected at the ring electrode is often called the "collection efficiency" of the RRDE. One may empirically measure the collection efficiency of a specific RRDE before using it for any quantitative work. This is normally carried out using a welldefined electrochemical system such as the ferricyanide-ferrocyanide redox couple. This system can be used to measure stable collection efficiency at rates between 400 and 2000 rpm.
The ring collection efficiency was calculated to be 37% (or expressed as 0.37), as measured using a [Fe(CN) 6 ] 3-/4redox couple. The disc electrode is swept from 0.6V to -0.2V vs.
Ag/AgCl and the ring potential is kept at 0.23V vs. Ag/AgCl. The collection efficiency (N CL ), calculated from the current response of K 3 Fe(CN) 6 (5 mM) in 0.1M KCl was found to be 0.37 ± 1%. Ar gas was purged into the electrochemical cell for 30 min before the experiment and remained the Ar flow throughout the experiment.

Faradaic efficiency calculation:
In order to calculate the faradaic efficiency, 10 μL of the ink solution containing 4mg/mL Ni 1 Co 2 catalyst was drop-cast on the glassy carbon (GC) disk electrode. When the Ni 1 Co 2 loaded GC disk electrode (working electrode) potential was swept in anodic potentials, oxygen is generated from the working electrode and the produced oxygen will moved towards the platinum ring electrodes due to the rotating action (laminar flow) of the electrode 8 (1600rpm). The platinum ring electrode is held at -0.5V vs. Ag/AgCl, which will facilitates the reduction of oxygen 2 . The Faradaic efficiency can be calculated based on the ratio of disk and ring currents using the following equation: The faradaic efficiency calculated from the RRDE study (figure given below) is found to be 92%. Figure S10. Determination of Faradaic efficiency of Ni 1 Co 2 catalyst.

'Blue Bottle' Experiment:
To further analyse the gas produced in anode (OER reaction), an alkaline solution [1M solution (pH ~14)] of KOH containing glucose (Sigma Aldrich] is taken. Here glucose will act as reducing agent to change the colour of added methylene blue (sigma Aldrich) dye from blue to colourless leuco-methylene blue. Shaking/ introduction (purging) of external oxygen to this solution containing leuco-methylene blue will lead to the oxidation back to methylene blue -leading to regaining of its parent blue colour. When the dissolved oxygen has been fully consumed, methylene blue will slowly reduce back to its colourless form by the remaining glucose. In the present experiment, the above colourless solution is saturated with N 2 gas and showed that even shaking the solution does not give any colour change. Further, oxygen gas  Further, UV-Vis spectroscopy is used to confirm the transformation of leucomethylene blue to methylene blue due to the presence of oxygen. The UV Vis spectra of leuco-methylene blue solution after various times (black to violet) of oxygen purging are given in the figure S11. In the presence of oxygen (the gas coming out from our MEA setup) there is decrease in intensity of peak corresponds to leuco-methylene blue (peak a) centred at (664 nm). This clearly confirms the production of oxygen gas from the water electrolyser set up. Figure S12. HER activity of different types of materials used in this study.