Efficient hydrogen evolution by ternary molybdenum sulfoselenide particles on self-standing porous nickel diselenide foam

With the massive consumption of fossil fuels and its detrimental impact on the environment, methods of generating clean power are urgent. Hydrogen is an ideal carrier for renewable energy; however, hydrogen generation is inefficient because of the lack of robust catalysts that are substantially cheaper than platinum. Therefore, robust and durable earth-abundant and cost-effective catalysts are desirable for hydrogen generation from water splitting via hydrogen evolution reaction. Here we report an active and durable earth-abundant transition metal dichalcogenide-based hybrid catalyst that exhibits high hydrogen evolution activity approaching the state-of-the-art platinum catalysts, and superior to those of most transition metal dichalcogenides (molybdenum sulfide, cobalt diselenide and so on). Our material is fabricated by growing ternary molybdenum sulfoselenide particles on self-standing porous nickel diselenide foam. This advance provides a different pathway to design cheap, efficient and sizable hydrogen-evolving electrode by simultaneously tuning the number of catalytic edge sites, porosity, heteroatom doping and electrical conductivity.


Supplementary
. Here j 0 represents the exchange current density, while , η 10 , η 20   The synthesis of porous NiSe foam from commercial Ni foam. 2 foam was performed via direct selenization in a tube furnace. The commercial Ni foam was cut into pieces with an area of 1.0 cm 2 . Selenium powder (99.5%, Alfa Aesar) was used to supply Se vapor at the upstream of the furnace, which was transferred to the center region by Ar gas and reacted with Ni foam. System purging was performed before heating by introducing high-flow Ar gas (99.999%, ultrahigh purity). The furnace was programmed and heated to 600 o C in a short time and kept unchanged for 1h. Finally, the tube furnace was automatically turned off and cooled down with the protection of Ar gas.

Supplementary Note 5: Calculation of turn over frequency (TOF).
For rough estimation of the active surface site density and per-site turn over frequency (TOF) in the MoS 2(1-x) Se 2x /NiSe 2 hybrid catalyst, we suppose that the contribution of the MoS 2(1-x) Se 2x particles plays a dominant role. This is reasonable since the surface of NiSe 2 foam is nearly fully covered by the MoS 2(1-x) Se 2x particles, and the catalytic performance of ternary MoS 2(1-x) Se 2x /NiSe 2 hybrid is far better than that of pure NiSe 2 foam. According to this approach adopted by Jaramillo et al 20,21 , we carried out a similar calculation method by considering the relative roughness factor of the catalyst, the geometry of a MoS 2(1-x) Se 2x surface, and the hydrogen evolution current density. As shown in Fig. 3e, we have determined the specific capacitance to be 319 mF cm -2 , which can be directly used to estimate the relevant electrochemical active surface area (ECSA) by using the specific capacitance value for a flat electrode with real surface area 1 cm 2 . We assume 60 μF cm -2 for a flat electrode provided in Jaramillo et al 2,21 and Kim et al 1 for calculation here, and use 20 and 60 μF cm -2 for evaluating a lower and upper limit of the TOFs 20 (Supplementary Table 3). Thus, the number of electrochemically effective surface sites on the MoS 2(1-x) Se 2x catalyst was calculated as the following: Compared to the flat standard electrode (60 μF cm -2 ), the relative roughness factor of the investigated catalyst is determined to be ~ 5316 based on the electrochemically double-layer capacitance measurement.
As a result, the number of surface active sites for the MoS 2(1-x) Se 2x /NiSe 2 hybrid catalyst is estimated to be 5.85 × 10 18 surface sites/cm 2 To further get insights into the per-site TOF, the following formula is utilized: from the above formula, indicating a large number of active sites introduced by our special experimental design.
The total number of hydrogen turn overs is related to the current density, and is calculated based on the following conversion: While for the MoS 2 /NiSe 2 hybrid catalyst, given that its double-layer capacitance is around 30.9 mFcm -2 , and the current densities are 4.9, 13.0, and 31.0 mAcm -2 at η = 100, 125, and 150 mV, respectively, we can get the corresponding TOF values to be 0.027, 0.071, and 0.170 H 2 /s per surface site.
These results help us to conclude that the MoS 2(1-x) Se 2x /NiSe 2 hybrid catalyst shows a faster TOF value compared to that of the MoS 2 /NiSe 2 hybrid catalyst. As summarized above in Supplementary Table 3, the MoS 2(1-x) Se 2x /NiSe 2 hybrid catalyst has a faster TOF than that of MoS 2 on 3D carbon fiber paper, 3 MoS 2 particles on graphene oxide, 18 and double-gyroid MoS 2 . 2 Meanwhile, by considering the loading (4.5 mg cm -2 ) of MoS 2(1-x) Se 2x particles, we can also make a rough estimation of the TOFs, which are 0.014, 0.041 and 0.099 s -1 at the overpotentials of 100, 125 and 150 mV, respectively. 14,22 These TOFs are in the range of the above values calculated by the capacitance method (Supplementary Table 3).