Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Non-noble metal-nitride based electrocatalysts for high-performance alkaline seawater electrolysis

Non-noble metal-nitride based electrocatalysts for high-performance alkaline seawater electrolysis ARTICLE https://doi.org/10.1038/s41467-019-13092-7 OPEN Non-noble metal-nitride based electrocatalysts for high-performance alkaline seawater electrolysis 1,2 2,3 2,3 2 2 4 4 Luo Yu , Qing Zhu , Shaowei Song , Brian McElhenny , Dezhi Wang , Chunzheng Wu , Zhaojun Qin , 4 1 2 2 Jiming Bao , Ying Yu *, Shuo Chen * & Zhifeng Ren * Seawater is one of the most abundant natural resources on our planet. Electrolysis of sea- water is not only a promising approach to produce clean hydrogen energy, but also of great significance to seawater desalination. The implementation of seawater electrolysis requires robust and efficient electrocatalysts that can sustain seawater splitting without chloride corrosion, especially for the anode. Here we report a three-dimensional core-shell metal- nitride catalyst consisting of NiFeN nanoparticles uniformly decorated on NiMoN nanorods supported on Ni foam, which serves as an eminently active and durable oxygen evolution reaction catalyst for alkaline seawater electrolysis. Combined with an efficient hydrogen evolution reaction catalyst of NiMoN nanorods, we have achieved the industrially required −2 current densities of 500 and 1000 mA cm at record low voltages of 1.608 and 1.709 V, respectively, for overall alkaline seawater splitting at 60 °C. This discovery significantly advances the development of seawater electrolysis for large-scale hydrogen production. 1 2 College of Physical Science and Technology, Central China Normal University, Wuhan 430079, China. Department of Physics and TcSUH, University of 3 4 Houston, Houston, TX 77204, USA. Materials Science and Engineering Program, University of Houston, Houston, TX 77204, USA. Department of Electrical and Computer Engineering, University of Houston, Houston, TX 77204, USA. *email: yuying01@mail.ccnu.edu.cn; schen34@uh.edu; zren@uh.edu NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications 1 1234567890():,; ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 ydrogen (H ) is playing an increasingly important role as (NiMoN@NiFeN) for exceptional alkaline seawater electrolysis. an ideal energy source owing to its high energy density The 3D core-shell catalyst yields large current densities of 500 −1 1–5 −2 H(142 MJ kg ) and pollution-free use . Splitting water and 1000 mA cm at overpotentials of 369 and 398 mV, into H and oxygen (O ) by electricity produced from waste heat respectively, for OER in 1 M KOH + natural seawater at 25 °C. 2 2 or from renewable but intermittent wind or solar energy is one of In-depth studies show that in situ evolved amorphous layers of the most efficient and sustainable routes for high-purity H NiFe oxide and NiFe oxy(hydroxide) on the anode surface are the 6–11 production . Over the past decade, many low-cost water real active sites that are not only responsible for the excellent electrolyzers with electrolytes consisting of high-purity freshwater OER performance, but also contribute to the superior chlorine have been developed, and some achieve performance even better corrosion-resistance. Additionally, the integrated 3D core-shell than that of the benchmark platinum (Pt) and iridium dioxide TMN nanostructures with multiple levels of porosity offer 12–15 (IrO ) catalysts . However, large-scale freshwater electrolysis numerous active sites, efficient charge transfer, and rapid gaseous would put a heavy strain on vital water resources. Seawater is one product releasing, which also account for the promoted OER of the most abundant natural resources on our planet and performance. An outstanding two-electrode seawater electrolyzer accounts for 96.5% of the world’s total water resources . Direct has subsequently been fabricated by pairing this OER catalyst electrolysis of seawater rather than freshwater is highly sig- with another efficient HER catalyst of NiMoN, where the current −2 nificant, especially for the arid zones, since this technology not densities of 500 and 1000 mA cm are achieved at record low only stores clean energy, but also produces fresh drinking water voltages of 1.608 and 1.709 V, respectively, for overall alkaline from seawater. Nevertheless, the implementation of seawater seawater splitting at 60 °C, along with superior stability. splitting remains highly challenging, especially for the anodic Impressively, our electrolyzer can be driven by an AA battery or a reaction. commercial thermoelectric (TE) module, demonstrating great The major challenge in seawater splitting is the chlorine evolu- potential and flexibility in utilizing a broad range of power tion reaction (CER), which occurs on the anode due to the exis- sources. Overall, this work greatly boosts the science and tech- tence of chloride anions (∼0.5 M) in seawater, and competes with nology of seawater electrolysis. 17,18 the oxygen evolution reaction (OER) . For the CER in alkaline media, chlorine would further react with OH for hypochlorite formation with an onset potential of about 490 mV higher than Results that of OER, and thus highly active OER catalysts are demanded to Electrocatalyst preparation and characterization. Figure 1a −2 deliver large current densities (500 and 1000 mA cm )atover- presents a schematic illustration of the synthesis procedures for 18,19 potentials well below 490 mV to avoid hypochlorite formation . the 3D core-shell NiMoN@NiFeN catalyst, where commercial Ni Another bottleneck hindering the progress of seawater splitting is foam (Supplementary Fig. 1) is used as the conductive support the formation of insoluble precipitates, such as magnesium due to its high surface area, good electrical conductivity, and low hydroxide, on the electrode surface, which may poison the OER cost .We first used a hydrothermal method to synthesize and hydrogen evolution reaction (HER) catalysts . To alleviate this NiMoO nanorod arrays on Ni foam, which was then soaked in a issue, catalysts possessing large surface areas with numerous active NiFe precursor ink and air-dried, followed by a one-step thermal sites are more favorable. In addition, the aggressive chloride anions nitridation. The stable construction and the hydrophilic nature of in seawater also corrode the electrodes, further restricting the the NiMoO nanorod arrays (Supplementary Fig. 2) facilitate the development of seawater splitting . Because of these intractable uniform coverage of the nanorods by the NiFe precursor ink. The obstacles, only a few studies on electrocatalysts for seawater split- pure NiMoN catalyst was prepared by nitridation of NiMoO ting have been reported, with limited progress made thus far. without soaking in the precursor ink, and scanning electron Recently, Kuang et al. reported an impressive anode catalyst microscopy (SEM) images show that numerous nanorods with composed of a nickel-iron hydroxide layer coated on a nickel smooth surfaces were uniformly and vertically grown on the sulfide layer for active and stable alkaline seawater electrolysis, in surface of the Ni foam (Fig. 1b and its inset, and Supplementary −2 which a current density of 400 mA cm was achieved at 1.72 V for Fig. 3). After soaking in the precursor ink and thermal nitrida- two-electrode electrolysis in 6 M KOH+ 1.5 M NaCl electrolyte at tion, the NiMoN@NiFeN shows a well-preserved nanorod mor- 80 °C . Other non-precious electrocatalysts, including transition phology with rough and dense surfaces (Fig. 1c and its inset). The metal hexacyanometallate, cobalt selenide, cobalt borate, and cobalt high-magnification SEM image in Fig. 1d clearly shows that the phosphate, have been well studied for OER in NaCl-containing surfaces of the nanorods were uniformly decorated with many 17,20,21 electrolytes , but the overpotentials needed to deliver large nanoparticles, forming a unique 3D core-shell nanostructure that −2 current densities (500 and 1000 mA cm ) are much higher than offers an extremely large surface area with a huge quantity of 490 mV, not to mention the activity for overall seawater splitting. active sites, even with the formation of insoluble precipitates Therefore, it is highly desirable to develop other robust and inex- during seawater electrolysis. For comparison, pure NiFeN nano- pensive electrocatalysts to expedite the sluggish seawater splitting particles (Supplementary Fig. 4) were also synthesized on the Ni process, especially for OER at large current densities, so as to boost foam by soaking bare Ni foam in the NiFe precursor ink, followed research on large-scale seawater electrolysis. by thermal nitridation. We also studied the morphology variation Transition metal-nitride (TMN) is highly corrosion-resistant, of NiMoN@NiFeN with different loading amounts of NiFeN electrically conductive, and mechanically strong, which makes it a nanoparticles by controlling the concentration of NiFe precursor very promising candidate for electrolytic seawater splitting . ink (Supplementary Fig. 5). It was determined that the optimized −1 Recent studies on Ni N/Ni, NiMoN, and Ni-Fe-Mo trimetallic concentration is 0.25 g ml , so this concentration was used for nitride catalysts have established TMN-based materials to be further analyses unless otherwise indicated. efficient non-noble metal electrocatalysts for freshwater splitting Transmission electron microscopy (TEM) images of NiMoN@- 23–25 in alkaline media (1 M KOH) . Considering the need for NiFeN in Fig. 1e, f further detail the desired core-shell catalysts with large surface areas and high-density active sites for morphology of the nanoparticle-decorated nanorods, showing seawater splitting, here we report the design and synthesis of a that the thickness of the NiFeN shell is about 100 nm. Figure 1g three-dimensional (3D) core-shell TMN-based OER electro- displays a high-resolution TEM (HRTEM) image taken from the catalyst, in which NiFeN nanoparticles are uniformly decorated tip of the NiMoN@NiFeN nanorod presented in Fig. 1f, showing on NiMoN nanorods supported on porous Ni foam that the NiFeN nanoparticles are highly mesoporous and 2 NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 ARTICLE Soaked in Ni(NO ) 3 2 + Fe(NO ) solution 3 3 Hydrothermal NH annealing b c d NiMoN (100) e g h NiMoN (110) 100 nm NiFeN (110) 0.186 nm NiFeN (002) NiFeN (002) NiFeN (112) Ni Mo Fe N i j Fig. 1 Synthesis and microscopic characterization of the as-prepared NiMoN@NiFeN catalyst. a Schematic illustration of the synthesis procedures for the self-supported 3D core-shell NiMoN@NiFeN catalyst. b–d SEM images of (b) NiMoN and (c, d) NiMoN@NiFeN at different magnifications. e, f TEM images of NiMoN@NiFeN core-shell nanorods at different magnifications. g HRTEM image, h SAED pattern, i EDS line scan, and j dark field scanning transmission electron microscopy (DF-STEM) image and corresponding elemental mapping of the NiMoN@NiFeN catalyst. Scale bars: b, c 30 µm; insets of (b, c)3 µm; d, e 500 nm; f 200 nm; g 20 nm; inset of (g) 1 nm; h 2 1/nm; i 250 nm; j 1 µm interconnected with one another to form a 3D porous network, We then conducted X-ray diffraction (XRD) and X-ray which is beneficial for seawater diffusion and gaseous product photoelectron spectroscopy (XPS) measurements to study the 27,28 release . The HRTEM image in the Fig. 1g inset reveals chemical compositions and surface element states of the catalysts. distinctive lattice fringes with interplanar spacings of 0.186 nm, Typical XRD patterns (Fig. 2a) reveal the successful formation of which is assigned to the (002) plane of NiFeN. The selected area NiMoN and NiFeN compositions after corresponding thermal electron diffraction (SAED) pattern (Fig. 1h) recorded from the nitridation. Figure 2b shows the XPS survey spectra, demonstrat- NiMoN@NiFeN core-shell nanorod exhibits apparent diffraction ing the presence of Ni, Mo, and N in the NiMoN nanorods; Ni, rings of NiMoN and NiFeN, confirming the existence of NiMoN Fe, and N in the NiFeN nanoparticles; and Ni, Mo, Fe, and N in and NiFeN phases. The energy dispersive X-ray spectroscopy the core-shell NiMoN@NiFeN nanorods. For the high-resolution (EDS) line scan result (Fig. 1i) and EDS mapping analysis (Fig. 1j) XPS of Ni 2p of the three catalysts (Fig. 2c), the two peaks at further verify the quintessential core-shell nanostructure, clearly 853.4 and 870.8 eV are attributed to the Ni 2p and Ni 2p of 3/2 1/2 displaying that Mo and Fe are distributed in the central nanorod Ni species in Ni-N, respectively, while the peaks located at 856.3 and edge nanoparticles, respectively, while Ni and N are and 873.9 eV are assigned to the Ni 2p and Ni 2p of the 3/2 1/2 homogeneously distributed throughout the entire core-shell oxidized Ni species (Ni–O), respectively . The two additional nanorod. peaks at 862.0 and 880.1 eV are the relevant satellite peaks (Sat.). NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications 3 NiMoO Ni foam NiMoN@NiFeN Ni Fe Mo ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 a bc Ni NiFeN NiMoN Ni 2p NiMoN@NiFeN 3/2 Ni 2p 1/2 Ni-N Ni-O Sat. Ni-O Ni-N Sat. NiMoN@NiFeN NiFeN NiFeN NiMoN@NiFeN NiFeN NiMoN NiMoN NiMoN 35 40 45 50 55 60 65 70 75 80 900 800 700 600 500 400 300 200 888 882 876 870 864 858 852 2θ (deg.) Binding energy (eV) Binding energy (eV) de f NiMoN@NiFeN NiMoN@NiFeN NiMoN@NiFeN Metal-N Mo 3d 5/2 Mo 3d Mo 3d 3/2 3/2 Fe 2p Fe 2p 3/2 1/2 0.38 eV 0.34 eV Sat. 0.38 eV N-H 6+ 6+ 6+ MO MO MO NiFeN NiMoN NiFeN NiMoN 730 725 720 715 710 705 700 238 236 234 232 230 228 226 402 400 398 396 394 392 Binding energy (eV) Binding energy (eV) Binding energy (eV) Fig. 2 Structural characterization of as-prepared catalysts. a XRD, and b XPS survey, and c–f high-resolution XPS of (c) Ni 2p, (d) Fe 2p, (e) Mo 3d, and (f) N 1 s of the NiMoN, NiFeN, and NiMoN@NiFeN catalysts The Fe 2p XPS of NiFeN and NiMoN@NiFeN in Fig. 2d show (348 and 417 mV), NiMoN (350 and 458 mV), and the bench- two peaks of Fe 2p and Fe 2p at 711.0 and 723.6 eV, mark IrO electrodes (430 and 542 mV). This performance is also 3/2 1/2 2 respectively, as well as a tiny peak at 720.5 corresponding to the superior to that of most non-precious OER catalysts in 1 M KOH satellite peak . In Fig. 2e, the Mo 3d XPS of NiMoN and (Supplementary Table 1), including the recently reported ZnCo 3+ 6+ 33 34 NiMoN@NiFeN show two valence states of Mo and Mo . For oxyhydroxide , Se-doped FeOOH , NiCoFe-MOF (metal- NiMoN, the peak located at 229.6 eV (Mo 3d ) is ascribed to organic frameworks) , and FeNiP/NCH (nitrogen-doped carbon 5/2 3+ 36 Mo in the metal-nitride, which is recognized to be active for hollow framework) . The polarization curves of the CV backward HER . The peaks at 232.7 (Mo 3d ) and 235.3 eV are attributed scan, the CV without and with iR compensation are presented for 3/2 6+ 31 to Mo due to the surface oxidation of NiMoN . However, the comparison in Supplementary Figs. 7, 8, and 9a, respectively. We 3+ 6+ two main peaks of Mo 3d (Mo ) and Mo 3d (Mo ) show also investigated the redox behaviors of the different metal-nitride 5/2 3/2 an apparent negative shift in binding energy for the NiMoN@- catalysts by analyzing their CV curves in the range of about 1.125 NiFeN, indicating the strong electronic interactions between ~1.525 V vs. RHE, and the results are displayed in Supplementary NiMoN and NiFeN. For the N 1 s XPS (Fig. 2f), the main peak is Fig. 9b–d. In addition, the OER activity of other NiMoN@NiFeN located at 397.4 eV, which is ascribed to the N species in metal- catalysts with different loading amounts of NiFeN was also studied nitrides, and another peak at 399.6 eV originates from the (Supplementary Fig. 10), and the one prepared with a precursor 23,32 −1 incomplete reaction of NH . Additionally, the Mo 3p peak ink concentration of 0.25 g ml exhibits the highest OER activity. 3 3/2 also appears for the NiMoN and NiMoN@NiFeN, and a negative Tafel plots in Fig. 3b show that the NiMoN@NiFeN catalyst has a −1 shift in binding energy still exists for the NiMoN@NiFeN, which relatively smaller Tafel slope of 58.6 mV dec in comparison with −1 −1 is in good agreement with the results in Fig. 2e. that of the NiFeN (68.9 mV dec ), NiMoN (82.1 mV dec ), and −1 IrO electrodes (86.7 mV dec ), verifying its rapid OER catalytic kinetics. We further calculated TOF to assess the intrinsic OER activity of the NiMoN@NiFeN catalyst, which presents a TOF Oxygen and hydrogen evolution catalysis.We first evaluated the −1 OER activity of the as-prepared catalysts in 1 M KOH electrolyte value of 0.09 s at an overpotential of 300 mV. This value is not the best among the OER catalysts listed in Supplementary Table 1, in freshwater at room temperature (25 °C). The benchmark IrO catalyst on Ni foam was also included for comparison. All of the but still larger than that of the very good OER catalysts of (Ni,Fe) 12 37 38 OOH ,Fe Co OOH , and NiFe-OH/NiFeP . Impressively, measured potentials vs. Hg/HgO were converted to the reversible x 1−x hydrogen electrode (RHE) according to the reference electrode our 3D core-shell NiMoN@NiFeN catalyst shows very good durability as well for OER in 1 M KOH electrolyte. As revealed in calibration (Supplementary Fig. 6, E = E + 0.925). All RHE Hg/HgO −2 data were measured after cyclic voltammetry (CV) activation and Fig. 3c, the current densities of 100 and 500 mA cm at constant overpotentials show negligible decrease over 48 h OER catalysis, reported with iR compensation (85%). The current density was normalized by the geometrical surface area unless otherwise and the CV polarization curves (inset of Fig. 3c) after the stability test remain almost the same as prior to the test. It should be noted mentioned. As the CV forward scan results in Fig. 3a show, our −2 that for the stability test at 500 mA cm , the current density 3D core-shell NiMoN@NiFeN catalyst exhibits significantly −2 improved OER activity, requiring overpotentials as low as 277 and slightly decreases from 499.5 to 480.9 mA cm with a degrada- −2 −1 −2 tion rate of 0.775 mA cm h , which is mainly attributed to the 337 mV to achieve current densities of 100 and 500 mA cm , respectively, which are considerably smaller than that of NiFeN strong adsorption of bubbles blocking the active sites. Moreover, 4 NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications Intensity (a.u.) Intensity (a.u.) (100) (110) (002) (110) (112) Intensity (a.u.) Intensity (a.u.) Ni 2p Fe LMM Fe 2p Ni LMM O 1s Mo 3p N 1s C 1s Mo 3d Intensity (a.u.) Intensity (a.u.) NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 ARTICLE ab c 500 0.4 337 417 542 mV NiFeN –2 458 500 mA cm @337 mV NiMoN 400 500 NiMoN@NiFeN IrO Initial 0.3 400 After stability test 0.2 200 277 348 350 200 430 mV 100 –2 100 mA cm @277 mV 1.2 1.3 1.4 1.5 1.6 E (V vs. RHE) 0.1 1.2 1.8 1.4 1.5 1.6 1.7 1.8 0.5 1.0 1.5 2.0 0 8 16 24 32 40 48 –2 E (V vs. RHE) log|j (mA cm )| Time (h) de f –2 0.20 100 mA cm @56 mV –100 205 96 84 –100 56 mV Initial 0.15 –200 After stability test –200 –200 –400 –300 0.10 NiFeN –600 –300 NiMoN –400 –800 –2 0.05 500 mA cm @127 mV NiMoN@NiFeN –400 –1000 Pt/C –500 –0.3 –0.2 –0.1 0.0 E (V vs. RHE) 299 252 180 127 mV 0.00 –500 –600 –0.5 –0.4 –0.3 –0.2 –0.1 0.0 1.0 1.5 2.0 0 8 16 24 32 40 48 –2 E ( V vs. RHE) log|j (mA cm ) Time (h) gh –2 100 mA cm NiMoN@NiFeN for OER 1 M KOH 400 400 1 M KOH + 0.5 M NaCl –2 500 mA cm 1 M KOH + seawater –2 300 300 1000 mA cm NiMoN for HER NiMoN for HER 200 200 NiMoN@NiFeN for OER 100 100 0 0 –0.2 –0.1 0.0 1.3 1.4 1.5 1.6 KOH KOH + Nacl KOH + seawater KOH KOH + Nacl KOH + seawater E (V vs. RHE) Fig. 3 Oxygen and hydrogen evolution catalysis. a OER polarization curves in 1 M KOH, and b corresponding Tafel plots of different catalysts. c OER chronoamperometry curves of NiMoN@NiFeN at overpotentials of 277 and 337 mV in 1 M KOH. Inset: CV curves of NiMoN@NiFeN before and after the stability test. d HER polarization curves tested in 1 M KOH, and e corresponding Tafel plots of different catalysts. f HER chronoamperometry curves of NiMoN at overpotentials of 56 and 127 mV in 1 M KOH. Inset: LSV curves of NiMoN before and after the stability test. g OER and HER polarization curves of NiMoN@NiFeN and NiMoN, respectively, in different electrolytes. h Comparison of the overpotentials required to achieve current densities of 100, 500, −2 and 1000 mA cm for NiMoN@NiFeN (OER) and NiMoN (HER) in different electrolytes SEM images after OER stability tests (Supplementary Fig. 11) the NiMoN@NiFeN core-shell catalyst, the highly conductive core demonstrate the high integrity of the 3D core-shell nanostructures of NiMoN nanorods and the robust contact between the NiFeN of the NiMoN@NiFeN catalyst. Thus, the long-term robustness nanoparticles and NiMoN nanorods facilitate the charge transfer mostly originates from its integral 3D core-shell nanostructure between the catalyst and electrolyte, as manifested by the results with different levels of porosity, which benefits rapid gaseous from electrochemical impedance spectroscopy (EIS, Supplemen- product release, and the strong adhesion between the TMN cat- tary Fig. 15), which show that the charge-transfer resistance (R ) ct alysts and the Ni foam substrate. To investigate the origins of of this 3D core-shell electrode is only 1.0 Ω, significantly smaller promoted OER activity in the NiMoN@NiFeN catalyst, we cal- than 9.6 Ω for NiFeN. Additionally, the NiMoN catalyst also has a culated the electrochemical active surface area (ECSA) for the small R of 1.7 Ω,confirming its good electronic conductivity and ct different catalysts by double-layer capacitance (C ) from their CV fast charge transfer. Hence, the rational design of 3D core-shell dl curves (Supplementary Fig. 12) . Clearly, the C values of the TMN catalysts offers a large surface area and efficient charge dl NiMoN and NiMoN@NiFeN catalysts are as large as 188.3 and transfer, both of which contribute to the improved OER activity. −2 238.7 mF cm (Supplementary Fig. 13), respectively, which are To seek a good HER catalyst to combine with our nearly 2.9 and 3.6 times that of the pure NiFeN nanoparticles NiMoN@NiFeN catalyst for overall seawater splitting, we tested −2 (65.4 mF cm ), respectively, demonstrating the highly improved the HER performance of different catalysts, including the ECSA and the increased number of active sites achieved by dec- benchmark Pt/C on Ni foam, in 1 M KOH in freshwater. orating NiFeN nanoparticles on the NiMoN nanorods to form a Strikingly, both the NiMoN@NiFeN and NiMoN catalysts exhibit 3D core-shell nanoarchitecture, which benefits seawater adsorp- exceptional HER activity (Fig. 3d) that is even better than that of 40,41 tion and offers rich active sites for catalytic reactions .We the benchmark Pt/C catalyst, especially the NiMoN catalyst, further normalized current density by the ECSA, and the which requires very low overpotentials of 56 and 127 mV for −2 NiMoN@NiFeN catalyst still shows better OER activity than that current densities of 100 and 500 mA cm , respectively. The of NiFeN (Supplementary Fig. 14), indicating that factors other overpotentials required to achieve the same current densities by than the ECSA also contribute to the enhanced OER activity. For our NiMoN@NiFeN catalyst (84 and 180 mV) are slightly higher, NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications 5 –1 Pt/C: 51.8 mV dec –1 NiMoN@NiFeN: 50.7 mV dec –1 NiMoN@NiFeN: 58.6 mV dec –1 IrO : 86.7 mV dec –1 NiFeN: 68.9 mV dec –1 NiMoN: 45.6 mV dec –1 NiMoN: 82.1 mV dec –1 NiFeN: 96.7 mV dec –2 –2 –2 j (mA cm ) j (mA cm ) j (mA cm ) Overpotential (mV) Overpotential (V) Overpotential (V) –2 –2 j (mA cm ) j (mA cm ) –2 j (mA cm ) Overpotential (mV) ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 but superior to those needed for the Pt/C (96 and 252 mV) and natural seawater, the cell voltages for the corresponding current NiFeN (205 and 299 mV) catalysts. NiMoN has been demon- densities are only 1.774 and 1.901 V. Such performance even strated to be an efficient HER catalyst in alkaline media because outperforms that of most non-noble metal catalysts for alkaline of its excellent electronic conductivity and low adsorption free freshwater splitting, as well as that of the benchmark of Pt/C 24,42,43 15 energy of H* . Fig. 3e reveals that the NiMoN catalyst also and IrO catalysts in 1 M KOH . To boost the industrial appli- −1 exhibits a much smaller Tafel slope of 45.6 mV dec in cations of this electrolyzer, the cell voltages are further decreased comparison to the other catalysts measured. Moreover, the to 1.454, 1.608, and 1.709 V for current densities of 100, 500, and −2 NiMoN catalyst shows good stability at current densities of 100 1000 mA cm , respectively, in 1 M KOH + Seawater electrolyte −2 and 500 mA cm over 48 h HER testing (Fig. 3f). Therefore, our by heating the electrolyte to 60 °C, which can be easily achieved NiMoN@NiFeN and NiMoN catalysts are highly active and by employing a solar thermal hot water system. These values robust for OER and HER, respectively, during freshwater represent the current record-high performance indices for overall electrolysis in alkaline media. alkaline seawater splitting. The overall seawater splitting perfor- We then studied the OER and HER activity in an alkaline mance without iR compensation was also tested in 1 M KOH + simulated seawater electrolyte (1 M KOH + 0.5 M NaCl). As Seawater at 25 °C for comparison (Supplementary Fig. 19), and shown in Fig. 3g, the 3D core-shell NiMoN@NiFeN catalyst still was found to be worse than that with iR compensation. We exhibits outstanding catalytic activity for OER, requiring over- attempted to split pure natural seawater as well, but the perfor- potentials of 286 and 347 mV to achieve current densities of 100 mance is unsatisfactory due to the low ionic conductivity and −2 and 500 mA cm , respectively. This performance is very close to strong corrosiveness of the natural seawater (Supplementary that in the 1 M KOH electrolyte (Fig. 3g), suggesting selective Fig. 20). We then evaluated the Faradaic efficiency of the elec- OER in the alkaline adjusted salty water. We also collected trolyzer in 1 M KOH + 0.5 M NaCl at room temperature by natural seawater from Galveston Bay near Houston, Texas, USA collecting the evolved gaseous products over the cathode and (Supplementary Fig. 16) and prepared an alkaline natural anode (Supplementary Fig. 21). As shown in Fig. 4c, only H and seawater electrolyte (1 M KOH + Seawater), in which the OER O gases with a molar ratio close to 2:1 are detected, and the activity of the NiMoN@NiFeN catalyst shows only slight decay Faradaic efficiency is determined to be around 97.8% during compared with that in the other two electrolytes (Fig. 3g). The seawater electrolysis, demonstrating the high selectivity of OER slight decrease in activity may be due to some insoluble on the anode. precipitates [e.g., Mg(OH) and Ca(OH) ] covering the surface The operating durability is also a very important metric to 2 2 of the electrode, and thus burying some surface active sites assess the performance of an electrolyzer. As shown in Fig. 4d, this (Supplementary Figs. 17 and 18). Even so, the NiMoN@NiFeN electrolyzer can retain outstanding overall seawater splitting −2 catalyst still delivers current densities of 100 and 500 mA cm at performance with no noticeable degradation over 100 h operation −2 small overpotentials of 307 and 369 mV, respectively, in the at a constant current density of 100 mA cm in both the alkaline alkaline natural seawater electrolyte (Fig. 3h). In addition, at an simulated and natural seawater electrolytes. More importantly, −2 even larger current density of 1000 mA cm , the demanded the voltage needed to achieve a very large current density of −2 overpotential is only 398 mV, which is well below the 490 mV 500 mA cm also shows very little increase (<10%) after 100 h overpotential required to trigger chloride oxidation to hypo- water electrolysis in either of the two electrolytes (Fig. 4d), chlorite. Moreover, this overpotential is also much lower than verifying the superior durability of this electrolyzer. The anode of that of any of the other reported non-precious OER catalysts in the NiMoN@NiFeN catalyst further demonstrates good structural alkaline adjusted salty water (Supplementary Table 2). The HER integrity after long-term seawater electrolysis (Supplementary catalyst of NiMoN also exhibits excellent activity in both the Fig. 22). In addition, the electrolyzer exhibits very good activity alkaline simulated and natural seawater electrolytes (Fig. 3g). To and stability (over 600 h electrolysis) for overall seawater splitting −2 deliver current densities of 100, 500, and 1000 mA cm in the in a very harsh condition of 6 M KOH+ Seawater (Supplementary alkaline natural seawater, the required overpotentials are as low as Fig. 23), demonstrating its great potential for large-scale applica- 82, 160, and 218 mV, respectively (Fig. 3h). Consequently, our tions. Given its excellent catalytic performance, this electrolyzer NiMoN@NiFeN and NiMoN catalysts are not only efficient for can be easily actuated by a 1.5 V AA battery (Supplementary freshwater electrolysis, but also highly active for alkaline seawater Fig. 24). Moreover, we also demonstrated the harvesting of waste splitting. heat, the major energy loss in various activities and device operations, by our seawater electrolyzer powered with a commer- cial TE device that directly coverts heat into electricity (Fig. 4e) . Overall seawater splitting. Considering the outstanding catalytic As shown in Fig. 4f, when the temperature gradient between the performance of both the NiMoN@NiFeN and NiMoN catalysts, hot and cold sides of the TE module is 40, 50, and 60 °C, we further investigated the overall seawater splitting performance the corresponding output voltage can expeditiously drive the by integrating the two catalysts into a two-electrode alkaline electrolyzer for stable delivery of current density of 30, 100, and −2 electrolyzer (without a diaphragm or membrane), in which 200 mA cm , respectively. Even when the temperature gradient NiMoN@NiFeN is used as the anode for OER and NiMoN as the through the TE module is decreased to 40 °C, the electrolyzer can −2 cathode for HER (Fig. 4a). Remarkably, this electrolyzer shows still supply a current density of ~30 mA cm with good excellent overall seawater splitting activity in both the alkaline recyclability, suggesting that we can efficiently use the waste heat simulated and natural seawater electrolytes. As displayed in to produce H fuel by the electrolysis of seawater. Fig. 4b, at room temperature (25 °C), the cell voltages needed to −2 produce a current density of 100 mA cm are as low as 1.564 and 1.581 V in 1 M KOH + 0.5 M NaCl and 1 M KOH + Sea- Active sites for oxygen evolution catalysis. To gain a deeper water electrolytes, respectively. In particular, our electrolyzer insight into the real catalytic active sites for the extraordinary can generate extremely large current densities of 500 and OER activity of the NiMoN@NiFeN catalyst, we further studied −2 1000 mA cm at 1.735 and 1.841 V, respectively, in 1 M KOH + its nanostructure, surface composition, and chemical state during 0.5 M NaCl electrolyte, which is slightly better than the recently and after OER tests. The TEM image in Fig. 5a shows that the 3D reported anion exchange membrane (AEM) based electrolyzer in core-shell nanostructure of NiMoN@NiFeN is intact after OER an alkaline simulated seawater electrolyte . Even in the alkaline tests, which is consistent with the SEM results (Supplementary 6 NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications NiMoN@NiFeN NiMoN NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 ARTICLE a be 1 M KOH + 0.5 M NaCl@25° Hot side 1 M KOH + seawater@25° 800 1 M KOH + 0.5 M NaCl@60° PN H O 2 1 M KOH + seawater@60° –2 500 mA cm H O –2 100 mA cm Cold side 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 Voltage (V) cd f 3.0 2.0 Calculated H 2 1 M KOH + 0.5 M NaCl 1.9 Calculated O ΔT = 60 °C –2 1.8 500 mA cm 2.5 Measured H 1.7 –2 Measured O 100 mA cm 1.6 2.0 1.5 1.4 1.5 ΔT = 50 °C 2.0 1 M KOH + seawater –2 1.9 500 mA cm 1.0 1.8 –2 1.7 ΔT = 40 °C 100 mA cm 0.5 1.6 ΔT = 40 °C 1.5 0 0.0 1.4 0 20 40 60 80 100 0 20406080 100 0 100 200 300 400 500 600 Time (min) Time (h) Time (min) Fig. 4 Overall seawater splitting performance. a Schematic illustration of an overall seawater splitting electrolyzer with NiMoN and NiMoN@NiFeN as the cathode and anode, respectively. b Polarization curves after iR compensation of NiMoN and NiMoN@NiFeN coupled catalysts in a two-electrode electrolyzer tested in alkaline simulated (1 M KOH + 0.5 M NaCl, resistance: ~1.1 Ω) and natural seawater (1 M KOH + Seawater, resistance: ~1.2 Ω) electrolytes under different temperatures. c Comparison between the amount of collected and theoretical gaseous products (H and O ) by the two- 2 2 −2 electrode electrolyzer at a constant current density of 100 mA cm in 1 M KOH + 0.5 M NaCl at 25 °C. d Durability tests of the electrolyzer at constant −2 current densities of 100 and 500 mA cm in different electrolytes at 25 °C. e Schematic illustration of the principle for power generation between the hot and cold sides of a TE device. f Real-time dynamics of current densities for the electrolyzer in 1 M KOH + 0.5 M NaCl at 25 °C driven by a TE device when the temperature gradient (ΔT) between its hot and cold sides is 40, 50, 60, and 40 °C Fig. 11). The TEM image in Fig. 5b reveals that many nano- main peaks at 531.9 and 530.1 eV and the appearance of a new particles are closely attached on the nanorod, and there seems to peak at 532.3 eV . To confirm the formation of NiFe oxides/oxy be some very thin layers on the nanoparticle surface. The (hydroxides), we further performed in situ Raman measurements HRTEM image in Fig. 5c confirms the existence of thin amor- (Supplementary Fig. 28) to elucidate the real-time evolution of phous layers and Ni(OH) . We suspect that the thin layers are the NiMoN@NiFeN catalyst during the OER process. As the in situ generated amorphous NiFe oxides and NiFe oxy(hydro- results in Fig. 5g show, the spectrum for the as-prepared xides), which have been verified by elemental mapping and XPS NiMoN@NiFeN exhibits a sharp and broad peak at around −1 analyses following OER testing. Figure 5d displays the DF-STEM 80.3 cm , which is probably due to the metal-N stretching and corresponding elemental mapping images, which show the modes. The transformation into NiOOH starts at 1.4 V according −1 12 absence of N and the increased O content on the NiMoN@NiFeN to a new Raman band located at 480.1 cm . When the surface after OER due to the intense oxidation process. The high- potential reaches to 1.6 and 1.7 V, two additional Raman bands −1 resolution XPS of N 1s (Supplementary Fig. 25) also corroborates are generated. The one located at 324.7 cm is assigned to the 50 −1 this point (the surface N content in the NiMoN@NiFeN catalyst Fe-O vibrations in Fe O , and the other at 693.1 cm belongs 2 3 was reduced from 10.3% in the fresh sample to 0.36% after OER). to the Fe-O vibrations in amorphous FeOOH . Therefore, by For the high-resolution XPS of Ni 2p (Fig. 5e), the two peaks combining these results with the XPS results, we conclude that attributed to Ni-N species at 853.4 and 870.8 eV also disappear thin amorphous layers of NiFe oxide and NiFe oxy(hydroxide) after OER because of surface oxidation. A new peak at 868.9 eV, are evolved from the NiFeN nanoparticles at the surface during which is assigned to Ni(OH) , shows up,. Besides, the two peaks OER electrocatalysis, and that these serve as the real active sites at 856.3 (Ni-O) and 862.0 eV (Sat.) positively shift toward higher participating in the OER process. The formation of a metal binding energy, which is also observed in the XPS of Fe 2p nitride-metal oxide/oxy(hydroxide) core-shell structure may also 2+ 2+ (Fig. 5f), indicating the oxidation of Ni and Fe to the higher facilitate electron transfer from the NiFeN core to the oxidized 3+ 3+ valence states of Ni and Fe (Supplementary Fig. 26), species (Supplementary Fig. 29). This observation is consistent respectively, resulting from the formation of NiFe oxides/oxy with the results of other reported OER catalysts, including metal 46–48 14,46 (hydroxides) . The O 1s XPS (Supplementary Fig. 27) also selenides and phosphides . However, the structure of the 2+ 2+ proves the increased valence states of Ni and Fe after OER, in situ formed NiFe oxides/oxy(hydroxides) is different from that as well as showing the appearance of Fe-OH from the NiFe oxy of the (Ni,Fe)OOH thin-film catalyst reported by Zhou et al. , (hydroxides), which can be seen from the negative shift of the which undergoes a rapid self-reconstruction due to the partial NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications 7 Gas amount (mmol) –2 j (mA cm ) Voltage (V) –2 j (mA cm ) Heat flow ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 Ni Mo ac b d Amorphous NiFeOOH layer Fe N O 0.261 nm Ni(OH) (101) e f g Ni 2p Before OER Before OER In situ raman 3/2 –1 Fe 2p Ni 2p Fe 2p 80.3 cm 3/2 –1 1/2 1/2 693.1 cm Ni-N Sat. Ni-O –1 324.7 cm –1 1.7 V Ni-O 480.1 cm Sat. Ni-N Sat. 1.6 V 3+ Ni 3+ 3+ Fe Fe After OER 3+ Fe After OER Ni(OH) 3+ Ni 1.5 V 1.4 V KOH As prepared 888 882 876 870 864 858 852 730 725 720 715 710 705 700 100 200 300 400 500 600 700 800 –1 Binding energy (eV) Binding energy (eV) Raman shift (cm ) Fig. 5 Material characterization to study the OER active sites. a, b TEM images of NiMoN@NiFeN core-shell nanorods at different magnifications after OER tests. c HRTEM image, and d DF-STEM image and corresponding elemental mapping of the NiMoN@NiFeN catalyst after OER tests. Scale bars: a 500 nm; b 50 nm; c 5 nm; d 1 µm. High-resolution XPS of (e), Ni 2p and (f), Fe 2p of NiMoN@NiFeN after OER tests in comparison with those before OER tests. g In situ Raman spectra of the NiMoN@NiFeN catalyst at various potentials for the OER process dissolution of FeOOH in KOH solution, forming amorphous step in the development of a robust and active catalyst to utilize the NiOOH nanoarrays mixed with a small amount of FeOOH world’s abundant seawater feedstock for large-scale hydrogen nanoparticles after OER. Notably, such in situ generated amor- production by renewable energy sources. phous NiFe oxide and NiFe oxy(hydroxide) layers also play a positive role in improving the resistance to corrosion by chloride Methods anions in seawater (Supplementary Fig. 30), which contributes to Chemicals. Ethanol (C H OH, Decon Labs, Inc.), ammonium heptmolybdate 2 5 the superior stability during seawater electrolysis. [(NH ) Mo O ·4H O, 98%, Sigma-Aldrich], nickel(II) nitrate hexahydrate (Ni 4 6 7 24 2 (NO ) ·6H O, 98%, Sigma-Aldrich), iron (III) nitrate hexahydrate (Fe 3 2 2 (NO ) ·9H O, 98%, Sigma-Aldrich), N, N Dimethylformamide [DMF, (CH ) NC 3 3 2 3 2 (O)H, anhydrous, 99.8%, Sigma-Aldrich], platinum powder (Pt, nominally 20% on Discussion carbon black, Alfa Aesar), iridium oxide powder (IrO2, 99%, Alfa Aesar), Nafion In summary, we have developed a 3D core-shell OER catalyst of (117 solution, 5% wt, Sigma-Aldrich), sodium chloride (NaCl, Fisher Chemical), NiMoN@NiFeN for active and stable alkaline seawater splitting. potassium hydroxide (KOH, 50% w/v, Alfa Aesar), and Ni foam (thickness: 1.6 mm, porosity: ~95%) were used as received. Deionized (DI) water (resistivity: The interior NiMoN nanorods are highly conductive and afford a 18.3 MΩ·cm) was used for the preparation of all aqueous solutions. large surface area, which ensure efficient charge transfer and numerous active sites. The outer NiFeN nanoparticles in situ evolve Synthesis of NiMoO nanorods on Ni foam. NiMoO nanorods were synthesized thin amorphous layers of NiFe oxide and NiFe oxy(hydroxide) 4 4 on nickel foam through a hydrothermal method . A piece of commercial Ni foam during OER catalysis, which are not only responsible for the (2 × 5 cm ) was cleaned by ultrasonication with ethanol and DI water for several selective OER activity, but also beneficial for the corrosion resis- minutes, and the substrate was then transferred into a polyphenyl (PPL)-lined tance to chloride anions in seawater. At the same time, the 3D core- stainless-steel autoclave (100 ml) containing a homogenous solution of Ni shell nanostructures with multiple levels of porosity are favorable (NO ) ·6H O (0.04 M) and (NH ) Mo O ·4H O (0.01 M) in 50 ml H O. After- 3 2 2 4 6 7 24 2 2 ward, the autoclave was sealed and maintained at 150 °C for 6 h. The sample was for seawater diffusion and H /O gases releasing. Thus, this OER 2 2 then taken out and washed with DI water and ethanol several times before being catalyst requires very low overpotentials of 369 and 398 mV to fully dried at 60 °C overnight under vacuum. −2 deliver large current densities of 500 and 1000 mA cm , respec- tively, in alkaline natural seawater at 25 °C. Additionally, by pairing Synthesis of NiMoN@NiFeN core-shell nanorods. The metal nitrides were it with another efficient HER catalyst of NiMoN, we assembled an synthesized by one-step nitridation of the NiMoO nanorods in a tube furnace. For outstanding water electrolyzer for overall seawater splitting, which the synthesis of NiMoN nanorods, a piece of NiMoO /Ni foam (~1 cm ) was −2 placed at the middle of a tube furnace and thermal nitridation was conducted at outputs current densities of 500 and 1000 mA cm at record low 500 °C under a flow of 120 standard cubic centimeters (sccm) NH and 30 sccm Ar voltages of 1.608 and 1.709 V, respectively, in alkaline natural for 1 h. The furnace was then automatically turned off and naturally cooled down seawater at 60 °C. The electrolyzer also shows excellent durability at to room temperature under Ar atmosphere. For the synthesis of NiMoN@NiFeN −2 current densities of 100 and 500 mA cm during up to 100 h 2 core-shell nanorods, a piece of NiMoO /Ni foam (~1 cm ) was first soaked in a alkaline seawater electrolysis. This discovery represents a significant NiFe precursor ink, which was prepared by dissolving Ni(NO ) ·6H O and Fe 3 2 2 8 NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications Intensity (a.u.) Intensity (a.u.) Intensity (a.u.) NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 ARTICLE (NO ) ·9H O with a mole ratio of 1:1 in DMF, and the NiMoO /Ni foam coated Overall seawater splitting driven by a TE module. We purchased a commercial 3 3 2 4 with the NiFe precursor ink was then dried at ambient condition. The dried sample TE module from Amazon and used it as a power generator to drive our two- was subjected to thermal nitridation under the same conditions as for NiMoN. To electrode electrolyzer according to our previous work . During the test, the hot study the effect of the NiFeN loading amount on the morphology of the core-shell side of the TE module was covered by a large flat copper plate, which was in direct nanorods, we prepared four different NiMoN@NiFeN core-shell nanorods with contact with a heater on top. The hot-side temperature was maintained relatively different loading amounts of NiFeN by controlling the concentration of Ni and Fe constant by tuning the DC power supply to the heater, while the cold-side tem- −1 precursors. Specifically, 0.1, 0.25, 0.5, and 0.75 g ml concentrations of precursor perature was controlled by placing it in direct contact with a cooling system, where ink were used. For comparison, pure NiFeN nanoparticles were also prepared on the water inside was adjusted to remain at a constant temperature. Thus, the TE the Ni foam by replacing the NiMoO /Ni foam with Ni foam. The concentration of module generated a relatively stable open circuit voltage between the hot and cold −1 precursor ink in this case was 0.25 g ml , and all other synthesis conditions were sides. A nano-voltmeter and an ammeter were embedded into the circuit for real- the same as for NiMoN@NiFeN. time monitoring of the voltage and current, respectively, between the two elec- trodes of the water-splitting cell. Preparation of IrO and Pt/C catalysts on Ni foam. To prepare the IrO elec- 2 2 trode for comparison , 40 mg of IrO and 60 μLofNafion were dispersed in Data availability 540 μL of ethanol and 400 μL of DI water, and the mixture was ultrasonicated for The source data underlying Figs. 1i, 2a–f, 3c, f and h, 4c and f, 5e–g, and Supplementary 30 min. The dispersion was then coated onto a Ni foam substrate, which was dried Figs. 2a, 6, 12, 20, 21a, 23a, 25, 26, and 27 are provided as a Source Data file. The other in air overnight. Pt/C electrodes were obtained by the same method. data that support the findings of this work are available from the corresponding authors upon reasonable request. Materials characterization. The morphology and nanostructure of the samples were determined by scanning electron microscopy (SEM, LEO 1525) and trans- Code availability mission electron microscopy (TEM, JEOL 2010F) coupled with energy dispersive All plots and data analysis were performed with OriginPro 8.5 software, and will be made X-ray (EDX) spectroscopy. The phase composition of the samples was character- available by the corresponding authors upon reasonable request. ized by X-ray diffraction (PANalytical X’pert PRO diffractometer with a Cu Ka radiation source) and XPS (PHI Quantera XPS) using a PHI Quantera SXM scanning X-ray microprobe. Received: 18 June 2019; Accepted: 18 October 2019; Electrochemical tests. The electrochemical performance was tested on an elec- trochemical station (Gamry, Reference 600). The two half reactions of OER and HER were each carried out at room temperature (~25 °C) in a standard three- electrode system with our prepared sample as the working electrode, a graphite rod as the counter electrode, and a standard Hg/HgO electrode as the reference elec- References trode. Four different electrolytes, including 1 M KOH, 1 M KOH + 0.5 M NaCl, 1. Tiwari, J. N. et al. Multicomponent electrocatalyst with ultralow Pt loading 1 M KOH + Seawater, and natural seawater, were used, and the pH was around 14 and high hydrogen evolution activity. Nat. Energy 3, 773 (2018). −2 except for the natural seawater (pH~7.2). The electrode size is around 1 cm , and 2. Turner, J. A. Sustainable hydrogen production. Science 305, 972–974 (2004). −2 the effective parts in the electrolyte are 0.3 ~ 0.45 cm for different samples. Both 3. Cabán-Acevedo, M. et al. Efficient hydrogen evolution catalysis using ternary the anodes (NiMoN@NiFeN) and cathodes (NiMoN) were cycled ~100 times by pyrite-type cobalt phosphosulphide. Nat. Mater. 14, 1245 (2015). CV until a stable polarization curve was developed prior to measuring each 4. Luo, J. et al. Water photolysis at 12.3% efficiency via perovskite photovoltaics polarization curve. OER polarization curve measurements were performed by CV and earth-abundant catalysts. Science 345, 1593–1596 (2014). −1 at a scan rate of 2 mV s and stability tests were carried out under constant 5. Pu, Z. et al. A universal synthesis strategy for P-rich noble metal diphosphide overpotentials. HER polarization curve measurements were performed by linear based electrocatalyst for hydrogen evolution reactions. Energy Environ. Sci. 12, −1 sweep voltammetry (LSV) at a scan rate of 5 mV s and stability tests were carried 952–957 (2019). out under constant overpotentials. Electrochemical impedance spectra (EIS) were 6. Roger, I., Shipman, M. A. & Symes, M. D. Earth-abundant catalysts for measured at an overpotential of 270 mV from 0.1 Hz to 100 KHz with an ampli- electrochemical and photoelectrochemical water splitting. Nat. Rev. Chem. 1, tude of 10 mV. For the two-electrode seawater electrolysis, the as-prepared 0003 (2017). NiMoN@NiFeN and NiMoN catalysts (after CV activation) were used as the anode 7. Seh, Z. W. et al. Combining theory and experiment in electrocatalysis: Insights and cathode, respectively. The polarization curves were measured in different into materials design. Science 355, eaad4998 (2017). electrolytes at different temperatures (25 and 60 °C), and stability tests were carried −2 8. Ledezma-Yanez, I. et al. Interfacial water reorganization as a pH-dependent out under constant current densities of 100 and 500 mA cm at room descriptor of the hydrogen evolution rate on platinum electrodes. Nat. Energy temperature. 2, 17031 (2017). 9. Ding, Q., Song, B., Xu, P. & Jin, S. Efficient electrocatalytic and Turnover frequency (TOF) calculation. The TOF value is calculated from the photoelectrochemical hydrogen generation using MoS and related 54,55 Eq. (1) : compounds. Chem 1, 699–726 (2016). j  A  η 10. Landman, A. et al. Photoelectrochemical water splitting in separate oxygen TOF ¼ ð1Þ 4  F  n and hydrogen cells. Nat. Mater. 16, 646 (2017). 11. Fabbri, E. et al. Dynamic surface self-reconstruction is the key of highly active where j is the current density at a given overpotential, A is the surface area of the perovskite nano-electrocatalysts for water splitting. Nat. Mater. 16, 925 electrode, η is the OER Faradaic efficiency, F is the Faraday constant, and n is the (2017). total number of active sites on the electrode. To calculate TOF, the most significant 12. Zhou, H. et al. Water splitting by electrolysis at high current densities under challenge is to identify the number of active sites. Since the real active sites for the 1.6 volts. Energy Environ. Sci. 11, 2858–2864 (2018). NiMoN@NiFeN catalyst are in situ formed NiFe oxide and NiFe oxy(hydroxide), 13. Hui, L. et al. Overall water splitting by graphdiyne-exfoliated and- we calculated the number of active sites by integrating the Ni redox peaks from the sandwiched layered double-hydroxide nanosheet arrays. Nat. Commun. 9, CV curve according to the following Eq. (2) : 5309 (2018). Q  N 14. Yu, F. et al. High-performance bifunctional porous non-noble metal n ¼ ð2Þ F  R Ni=Fe phosphide catalyst for overall water splitting. Nat. Commun. 9, 2551 (2018). 15. Yu, L. et al. Cu nanowires shelled with NiFe layered double hydroxide where Q is the integration of Ni redox features from the CV curve (Supplementary nanosheets as bifunctional electrocatalysts for overall water splitting. Energy Fig. 9d), N is Avogadro’s constant, F is the Faraday constant, and R is the A Ni/Fe 2+ 3+ Environ. Sci. 10, 1820–1827 (2017). molar ratio of Ni/Fe, assuming that Ni /Ni is a one-electron process. For the −2 16. Dresp, S., Dionigi, F., Klingenhof, M. & Strasser, P. Direct electrolytic splitting NiMoN@NiFeN catalyst at overpotential of 300 mV, j = 206 mA cm , and thus −1 TOF was calculated to be 0.09 s at overpotential of 300 mV. of seawater: Opportunities and challenges. ACS Energy Lett. 4, 933–942 (2019). 17. Hsu, S. H. et al. An earth‐abundant catalyst‐based seawater photoelectrolysis Gas chromatography measurement. Overall seawater splitting for gas chroma- system with 17.9% solar‐to‐hydrogen efficiency. Adv. Mater. 30, 1707261 tography (GC, GOW-MAC 350 TCD) tests were performed in a gas-tight elec- (2018). trochemical cell with 1 M KOH + 0.5 M NaCl as the electrolyte at room 18. Kuang, Y. et al. Solar-driven, highly sustained splitting of seawater into temperature (25 °C). Chronopotentiometry was applied with a constant current hydrogen and oxygen fuels. Proc. Natl Acad. Sci. USA 116, 6624–6629 (2019). −2 density of 100 mA cm to maintain oxygen and hydrogen generation. For each 19. Dionigi, F., Reier, T., Pawolek, Z., Gliech, M. & Strasser, P. Design criteria, measurement over an interval of 10 min, 0.3 µL gas sample was carefully extracted operating conditions, and nickel-iron hydroxide catalyst materials for selective from the sealed cell and injected into the GC instrument using a glass syringe seawater electrolysis. ChemSusChem 9, 962–972 (2016). (Hamilton Gastight 1002). NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 20. Zhao, Y. et al. Charge state manipulation of cobalt selenide catalyst for overall 48. Hu, C. et al. Synergism of geometric construction and electronic regulation: seawater electrolysis. Adv. Energy Mater. 8, 1801926 (2018). 3D Se‐(NiCo) S /(OH) nanosheets for highly efficient overall water splitting. x x 21. Esswein, A. J., Surendranath, Y., Reece, S. Y. & Nocera, D. G. Highly active Adv. Mater. 30, 1705538 (2018). cobalt phosphate and borate based oxygen evolving catalysts operating in 49. Zhou, H. et al. Highly active catalyst derived from a 3D foam of Fe(PO ) / 3 2 neutral and natural waters. Energy Environ. Sci. 4, 499–504 (2011). Ni P for extremely efficient water oxidation. Proc. Natl Acad. Sci. USA 114, 22. Zhang, Y. et al. 3D porous hierarchical nickel-molybdenum nitrides 5607–5611 (2017). synthesized by RF plasma as highly active and stable hydrogen‐evolution‐ 50. Song, F. et al. An unconventional iron nickel catalyst for the oxygen evolution reaction electrocatalysts. Adv. Energy Mater. 6, 1600221 (2016). reaction. ACS Cent. Sci. 5, 558–568 (2019). 23. Song, F. et al. Interfacing nickel nitride and nickel boosts both electrocatalytic 51. Chemelewski, W. D., Lee, H.-C., Lin, J.-F., Bard, A. J. & Mullins, C. B. hydrogen evolution and oxidation reactions. Nat. Commun. 9, 4531 (2018). Amorphous FeOOH oxygen evolution reaction catalyst for 24. Wang, Y. et al. Self-supported NiMo-based nanowire arrays as bifunctional photoelectrochemical water splitting. J. Am. Chem. Soc. 136, 2843–2850 electrocatalysts for full water splitting. J. Mater. Chem. A 6, 8479–8487 (2018). (2014). 25. Zhu, C. et al. Fe‐Ni‐Mo nitride porous nanotubes for full water splitting and 52. Zhang, J. et al. Efficient hydrogen production on MoNi electrocatalysts with Zn‐air batteries. Adv. Energy Mater. 8, 1802327 (2018). fast water dissociation kinetics. Nat. Commun. 8, 15437 (2017). 26. Feng, L.-L. et al. High-index faceted Ni S nanosheet arrays as highly active 53. Yu, F. et al. Three-dimensional nanoporous iron nitride film as an efficient 3 2 and ultrastable electrocatalysts for water splitting. J. Am. Chem. Soc. 137, electrocatalyst for water oxidation. ACS Catal. 7, 2052–2057 (2017). 14023–14026 (2015). 54. Song, F. & Hu, X. Exfoliation of layered double hydroxides for enhanced 27. Xiao, M. et al. Hollow nanostructures for photocatalysis: Advantages and oxygen evolution catalysis. Nat. Commun. 5, 4477 (2014). challenges. Adv. Mater. 31, 1801369 (2018). 55. Zhang, B. et al. Homogeneously dispersed multimetal oxygen-evolving 28. Yu, L. et al. Ternary Ni Mo P nanowire arrays toward efficient and stable catalysts. Science 352, 333–337 (2016). 2(1-x) 2x hydrogen evolution electrocatalysis under large-current-density. Nano Energy 53, 492–500 (2018). Acknowledgements 29. Qiu, B. et al. Fabrication of nickel-cobalt bimetal phosphide nanocages The work performed in China is financially supported by the National Natural Science for enhanced oxygen evolution catalysis. Adv. Funct. Mater. 28, 1706008 Foundation of China (Nos. 21573085 and 51872108), the Wuhan Planning Project of (2018). Science and Technology (No. 2018010401011294), and self-determined research funds of 30. Wang, T. et al. NiFe (oxy) hydroxides derived from NiFe disulfides as an CCNU from the college’s basic research and operation through the Chinese Ministry of efficient oxygen evolution catalyst for rechargeable Zn-air batteries: the effect Education (No. CCNU18TS034). J.B. acknowledges the support from the Robert A. of surface S residues. Adv. Mater. 30, 1800757 (2018). Welch Foundation (E-1728). Z.R. acknowledges the Research Award from the Alexander 31. Wu, A. et al. Integrating the active OER and HER components as the von Humboldt Foundation and Prof. Kornelius Nielsch at IFW Dresden Germany. heterostructures for the efficient overall water splitting. Nano Energy 44, 353–363 (2018). 32. Yan, H. et al. Anion‐modulated HER and OER activities of 3D Ni-V‐based Author contributions interstitial compound heterojunctions for high‐efficiency and stable overall Z.F.R. and Y.Y. led the project. L.Y. designed and performed most of the experiments and water splitting. Adv. Mater. 31, 1901174 (2019). analyzed most of the data including material synthesis, characterization, and electro- 33. Huang, Z.-F. et al. Chemical and structural origin of lattice oxygen oxidation chemical tests. Q.Z. carried out the thermoelectric measurements. S.W.S. performed the in Co-Zn oxyhydroxide oxygen evolution electrocatalysts. Nat. Energy 4, XPS characterization. B.M. and D.Z.W. took the TEM images. Z.J.Q. and J.M.B. per- 329–338 (2019). formed the in situ Raman tests. C.Z.W. and J.M.B. performed the GC tests. L.Y., S.C., Y. 34. Niu, S. et al. Se-doping activates FeOOH for cost-effective and efficient Y., and Z.F.R. wrote the paper. All of the authors have discussed the results and revised electrochemical water oxidation. J. Am. Chem. Soc. 141, 7005–7013 (2019). the paper together. 35. Qian, Q., Li, Y., Liu, Y., Yu, L. & Zhang, G. Ambient fast synthesis and active sites deciphering of hierarchical foam‐like trimetal-organic framework nanostructures as a platform for highly efficient oxygen evolution Competing interests The authors declare no competing interests. electrocatalysis. Adv. Mater. 31, 1901139 (2019). 36. Wei, Y.-S. et al. A single-crystal open-capsule metal-organic framework. J. Am. Chem. Soc. 141, 7906–7916 (2019). Additional information 37. Ye, S. H., Shi, Z. X., Feng, J. X., Tong, Y. X. & Li, G. R. Activating CoOOH Supplementary information is available for this paper at https://doi.org/10.1038/s41467- porous nanosheet arrays by partial iron substitution for efficient oxygen 019-13092-7. evolution reaction. Angew. Chem. Int. Ed. 57, 2672–2676 (2018). 38. Liang, H. et al. Amorphous NiFe-OH/NiFeP electrocatalyst fabricated at low Correspondence and requests for materials should be addressed to Y.Y., S.C. or Z.R. temperature for water oxidation applications. ACS Energy Lett. 2, 1035–1042 (2017). Peer review information Nature Communications thanks the anonymous reviewers for 39. Gao, S. et al. Partially oxidized atomic cobalt layers for carbon dioxide their contributions to the peer review of this work. electroreduction to liquid fuel. Nature 529, 68 (2016). 40. Zhang, J., Zhao, Z., Xia, Z. & Dai, L. A metal-free bifunctional electrocatalyst Reprints and permission information is available at http://www.nature.com/reprints for oxygen reduction and oxygen evolution reactions. Nat. Nanotech. 10, 444 (2015). Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in 41. Yu, F. et al. Recent developments in earth-abundant and non-noble published maps and institutional affiliations. electrocatalysts for water electrolysis. Mater. Today Phys. 7, 121–138 (2018). 42. Chang, B. et al. Bimetallic NiMoN nanowires with a preferential reactive facet: An ultraefficient bifunctional electrocatalyst for overall water splitting. ChemSusChem 11, 3198–3207 (2018). Open Access This article is licensed under a Creative Commons 43. Jia, J. et al. Nickel molybdenum nitride nanorods grown on Ni foam as Attribution 4.0 International License, which permits use, sharing, efficient and stable bifunctional electrocatalysts for overall water splitting. ACS adaptation, distribution and reproduction in any medium or format, as long as you give Appl. Mater. Interfaces 10, 30400–30408 (2018). appropriate credit to the original author(s) and the source, provide a link to the Creative 44. Dresp, S. et al. Direct electrolytic splitting of seawater: activity, selectivity, Commons license, and indicate if changes were made. The images or other third party degradation, and recovery studied from the molecular catalyst structure to the material in this article are included in the article’s Creative Commons license, unless electrolyzer cell level. Adv. Energy Mater. 8, 1800338 (2018). indicated otherwise in a credit line to the material. If material is not included in the 45. He, J. & Tritt, T. M. Advances in thermoelectric materials research: looking article’s Creative Commons license and your intended use is not permitted by statutory back and moving forward. Science 357, eaak9997 (2017). regulation or exceeds the permitted use, you will need to obtain permission directly from 46. Xu, X., Song, F. & Hu, X. A nickel iron diselenide-derived efficient oxygen- the copyright holder. To view a copy of this license, visit http://creativecommons.org/ evolution catalyst. Nat. Commun. 7, 12324 (2016). licenses/by/4.0/. 47. He, Q. et al. Highly defective Fe-based oxyhydroxides from electrochemical reconstruction for efficient oxygen evolution catalysis. ACS Energy Lett. 3, © The Author(s) 2019 861–868 (2018). 10 NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

Non-noble metal-nitride based electrocatalysts for high-performance alkaline seawater electrolysis

Loading next page...
 
/lp/springer-journals/non-noble-metal-nitride-based-electrocatalysts-for-high-performance-00dq9x2EQE

References (68)

Publisher
Springer Journals
Copyright
Copyright © 2019 by The Author(s)
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2041-1723
DOI
10.1038/s41467-019-13092-7
Publisher site
See Article on Publisher Site

Abstract

ARTICLE https://doi.org/10.1038/s41467-019-13092-7 OPEN Non-noble metal-nitride based electrocatalysts for high-performance alkaline seawater electrolysis 1,2 2,3 2,3 2 2 4 4 Luo Yu , Qing Zhu , Shaowei Song , Brian McElhenny , Dezhi Wang , Chunzheng Wu , Zhaojun Qin , 4 1 2 2 Jiming Bao , Ying Yu *, Shuo Chen * & Zhifeng Ren * Seawater is one of the most abundant natural resources on our planet. Electrolysis of sea- water is not only a promising approach to produce clean hydrogen energy, but also of great significance to seawater desalination. The implementation of seawater electrolysis requires robust and efficient electrocatalysts that can sustain seawater splitting without chloride corrosion, especially for the anode. Here we report a three-dimensional core-shell metal- nitride catalyst consisting of NiFeN nanoparticles uniformly decorated on NiMoN nanorods supported on Ni foam, which serves as an eminently active and durable oxygen evolution reaction catalyst for alkaline seawater electrolysis. Combined with an efficient hydrogen evolution reaction catalyst of NiMoN nanorods, we have achieved the industrially required −2 current densities of 500 and 1000 mA cm at record low voltages of 1.608 and 1.709 V, respectively, for overall alkaline seawater splitting at 60 °C. This discovery significantly advances the development of seawater electrolysis for large-scale hydrogen production. 1 2 College of Physical Science and Technology, Central China Normal University, Wuhan 430079, China. Department of Physics and TcSUH, University of 3 4 Houston, Houston, TX 77204, USA. Materials Science and Engineering Program, University of Houston, Houston, TX 77204, USA. Department of Electrical and Computer Engineering, University of Houston, Houston, TX 77204, USA. *email: yuying01@mail.ccnu.edu.cn; schen34@uh.edu; zren@uh.edu NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications 1 1234567890():,; ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 ydrogen (H ) is playing an increasingly important role as (NiMoN@NiFeN) for exceptional alkaline seawater electrolysis. an ideal energy source owing to its high energy density The 3D core-shell catalyst yields large current densities of 500 −1 1–5 −2 H(142 MJ kg ) and pollution-free use . Splitting water and 1000 mA cm at overpotentials of 369 and 398 mV, into H and oxygen (O ) by electricity produced from waste heat respectively, for OER in 1 M KOH + natural seawater at 25 °C. 2 2 or from renewable but intermittent wind or solar energy is one of In-depth studies show that in situ evolved amorphous layers of the most efficient and sustainable routes for high-purity H NiFe oxide and NiFe oxy(hydroxide) on the anode surface are the 6–11 production . Over the past decade, many low-cost water real active sites that are not only responsible for the excellent electrolyzers with electrolytes consisting of high-purity freshwater OER performance, but also contribute to the superior chlorine have been developed, and some achieve performance even better corrosion-resistance. Additionally, the integrated 3D core-shell than that of the benchmark platinum (Pt) and iridium dioxide TMN nanostructures with multiple levels of porosity offer 12–15 (IrO ) catalysts . However, large-scale freshwater electrolysis numerous active sites, efficient charge transfer, and rapid gaseous would put a heavy strain on vital water resources. Seawater is one product releasing, which also account for the promoted OER of the most abundant natural resources on our planet and performance. An outstanding two-electrode seawater electrolyzer accounts for 96.5% of the world’s total water resources . Direct has subsequently been fabricated by pairing this OER catalyst electrolysis of seawater rather than freshwater is highly sig- with another efficient HER catalyst of NiMoN, where the current −2 nificant, especially for the arid zones, since this technology not densities of 500 and 1000 mA cm are achieved at record low only stores clean energy, but also produces fresh drinking water voltages of 1.608 and 1.709 V, respectively, for overall alkaline from seawater. Nevertheless, the implementation of seawater seawater splitting at 60 °C, along with superior stability. splitting remains highly challenging, especially for the anodic Impressively, our electrolyzer can be driven by an AA battery or a reaction. commercial thermoelectric (TE) module, demonstrating great The major challenge in seawater splitting is the chlorine evolu- potential and flexibility in utilizing a broad range of power tion reaction (CER), which occurs on the anode due to the exis- sources. Overall, this work greatly boosts the science and tech- tence of chloride anions (∼0.5 M) in seawater, and competes with nology of seawater electrolysis. 17,18 the oxygen evolution reaction (OER) . For the CER in alkaline media, chlorine would further react with OH for hypochlorite formation with an onset potential of about 490 mV higher than Results that of OER, and thus highly active OER catalysts are demanded to Electrocatalyst preparation and characterization. Figure 1a −2 deliver large current densities (500 and 1000 mA cm )atover- presents a schematic illustration of the synthesis procedures for 18,19 potentials well below 490 mV to avoid hypochlorite formation . the 3D core-shell NiMoN@NiFeN catalyst, where commercial Ni Another bottleneck hindering the progress of seawater splitting is foam (Supplementary Fig. 1) is used as the conductive support the formation of insoluble precipitates, such as magnesium due to its high surface area, good electrical conductivity, and low hydroxide, on the electrode surface, which may poison the OER cost .We first used a hydrothermal method to synthesize and hydrogen evolution reaction (HER) catalysts . To alleviate this NiMoO nanorod arrays on Ni foam, which was then soaked in a issue, catalysts possessing large surface areas with numerous active NiFe precursor ink and air-dried, followed by a one-step thermal sites are more favorable. In addition, the aggressive chloride anions nitridation. The stable construction and the hydrophilic nature of in seawater also corrode the electrodes, further restricting the the NiMoO nanorod arrays (Supplementary Fig. 2) facilitate the development of seawater splitting . Because of these intractable uniform coverage of the nanorods by the NiFe precursor ink. The obstacles, only a few studies on electrocatalysts for seawater split- pure NiMoN catalyst was prepared by nitridation of NiMoO ting have been reported, with limited progress made thus far. without soaking in the precursor ink, and scanning electron Recently, Kuang et al. reported an impressive anode catalyst microscopy (SEM) images show that numerous nanorods with composed of a nickel-iron hydroxide layer coated on a nickel smooth surfaces were uniformly and vertically grown on the sulfide layer for active and stable alkaline seawater electrolysis, in surface of the Ni foam (Fig. 1b and its inset, and Supplementary −2 which a current density of 400 mA cm was achieved at 1.72 V for Fig. 3). After soaking in the precursor ink and thermal nitrida- two-electrode electrolysis in 6 M KOH+ 1.5 M NaCl electrolyte at tion, the NiMoN@NiFeN shows a well-preserved nanorod mor- 80 °C . Other non-precious electrocatalysts, including transition phology with rough and dense surfaces (Fig. 1c and its inset). The metal hexacyanometallate, cobalt selenide, cobalt borate, and cobalt high-magnification SEM image in Fig. 1d clearly shows that the phosphate, have been well studied for OER in NaCl-containing surfaces of the nanorods were uniformly decorated with many 17,20,21 electrolytes , but the overpotentials needed to deliver large nanoparticles, forming a unique 3D core-shell nanostructure that −2 current densities (500 and 1000 mA cm ) are much higher than offers an extremely large surface area with a huge quantity of 490 mV, not to mention the activity for overall seawater splitting. active sites, even with the formation of insoluble precipitates Therefore, it is highly desirable to develop other robust and inex- during seawater electrolysis. For comparison, pure NiFeN nano- pensive electrocatalysts to expedite the sluggish seawater splitting particles (Supplementary Fig. 4) were also synthesized on the Ni process, especially for OER at large current densities, so as to boost foam by soaking bare Ni foam in the NiFe precursor ink, followed research on large-scale seawater electrolysis. by thermal nitridation. We also studied the morphology variation Transition metal-nitride (TMN) is highly corrosion-resistant, of NiMoN@NiFeN with different loading amounts of NiFeN electrically conductive, and mechanically strong, which makes it a nanoparticles by controlling the concentration of NiFe precursor very promising candidate for electrolytic seawater splitting . ink (Supplementary Fig. 5). It was determined that the optimized −1 Recent studies on Ni N/Ni, NiMoN, and Ni-Fe-Mo trimetallic concentration is 0.25 g ml , so this concentration was used for nitride catalysts have established TMN-based materials to be further analyses unless otherwise indicated. efficient non-noble metal electrocatalysts for freshwater splitting Transmission electron microscopy (TEM) images of NiMoN@- 23–25 in alkaline media (1 M KOH) . Considering the need for NiFeN in Fig. 1e, f further detail the desired core-shell catalysts with large surface areas and high-density active sites for morphology of the nanoparticle-decorated nanorods, showing seawater splitting, here we report the design and synthesis of a that the thickness of the NiFeN shell is about 100 nm. Figure 1g three-dimensional (3D) core-shell TMN-based OER electro- displays a high-resolution TEM (HRTEM) image taken from the catalyst, in which NiFeN nanoparticles are uniformly decorated tip of the NiMoN@NiFeN nanorod presented in Fig. 1f, showing on NiMoN nanorods supported on porous Ni foam that the NiFeN nanoparticles are highly mesoporous and 2 NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 ARTICLE Soaked in Ni(NO ) 3 2 + Fe(NO ) solution 3 3 Hydrothermal NH annealing b c d NiMoN (100) e g h NiMoN (110) 100 nm NiFeN (110) 0.186 nm NiFeN (002) NiFeN (002) NiFeN (112) Ni Mo Fe N i j Fig. 1 Synthesis and microscopic characterization of the as-prepared NiMoN@NiFeN catalyst. a Schematic illustration of the synthesis procedures for the self-supported 3D core-shell NiMoN@NiFeN catalyst. b–d SEM images of (b) NiMoN and (c, d) NiMoN@NiFeN at different magnifications. e, f TEM images of NiMoN@NiFeN core-shell nanorods at different magnifications. g HRTEM image, h SAED pattern, i EDS line scan, and j dark field scanning transmission electron microscopy (DF-STEM) image and corresponding elemental mapping of the NiMoN@NiFeN catalyst. Scale bars: b, c 30 µm; insets of (b, c)3 µm; d, e 500 nm; f 200 nm; g 20 nm; inset of (g) 1 nm; h 2 1/nm; i 250 nm; j 1 µm interconnected with one another to form a 3D porous network, We then conducted X-ray diffraction (XRD) and X-ray which is beneficial for seawater diffusion and gaseous product photoelectron spectroscopy (XPS) measurements to study the 27,28 release . The HRTEM image in the Fig. 1g inset reveals chemical compositions and surface element states of the catalysts. distinctive lattice fringes with interplanar spacings of 0.186 nm, Typical XRD patterns (Fig. 2a) reveal the successful formation of which is assigned to the (002) plane of NiFeN. The selected area NiMoN and NiFeN compositions after corresponding thermal electron diffraction (SAED) pattern (Fig. 1h) recorded from the nitridation. Figure 2b shows the XPS survey spectra, demonstrat- NiMoN@NiFeN core-shell nanorod exhibits apparent diffraction ing the presence of Ni, Mo, and N in the NiMoN nanorods; Ni, rings of NiMoN and NiFeN, confirming the existence of NiMoN Fe, and N in the NiFeN nanoparticles; and Ni, Mo, Fe, and N in and NiFeN phases. The energy dispersive X-ray spectroscopy the core-shell NiMoN@NiFeN nanorods. For the high-resolution (EDS) line scan result (Fig. 1i) and EDS mapping analysis (Fig. 1j) XPS of Ni 2p of the three catalysts (Fig. 2c), the two peaks at further verify the quintessential core-shell nanostructure, clearly 853.4 and 870.8 eV are attributed to the Ni 2p and Ni 2p of 3/2 1/2 displaying that Mo and Fe are distributed in the central nanorod Ni species in Ni-N, respectively, while the peaks located at 856.3 and edge nanoparticles, respectively, while Ni and N are and 873.9 eV are assigned to the Ni 2p and Ni 2p of the 3/2 1/2 homogeneously distributed throughout the entire core-shell oxidized Ni species (Ni–O), respectively . The two additional nanorod. peaks at 862.0 and 880.1 eV are the relevant satellite peaks (Sat.). NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications 3 NiMoO Ni foam NiMoN@NiFeN Ni Fe Mo ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 a bc Ni NiFeN NiMoN Ni 2p NiMoN@NiFeN 3/2 Ni 2p 1/2 Ni-N Ni-O Sat. Ni-O Ni-N Sat. NiMoN@NiFeN NiFeN NiFeN NiMoN@NiFeN NiFeN NiMoN NiMoN NiMoN 35 40 45 50 55 60 65 70 75 80 900 800 700 600 500 400 300 200 888 882 876 870 864 858 852 2θ (deg.) Binding energy (eV) Binding energy (eV) de f NiMoN@NiFeN NiMoN@NiFeN NiMoN@NiFeN Metal-N Mo 3d 5/2 Mo 3d Mo 3d 3/2 3/2 Fe 2p Fe 2p 3/2 1/2 0.38 eV 0.34 eV Sat. 0.38 eV N-H 6+ 6+ 6+ MO MO MO NiFeN NiMoN NiFeN NiMoN 730 725 720 715 710 705 700 238 236 234 232 230 228 226 402 400 398 396 394 392 Binding energy (eV) Binding energy (eV) Binding energy (eV) Fig. 2 Structural characterization of as-prepared catalysts. a XRD, and b XPS survey, and c–f high-resolution XPS of (c) Ni 2p, (d) Fe 2p, (e) Mo 3d, and (f) N 1 s of the NiMoN, NiFeN, and NiMoN@NiFeN catalysts The Fe 2p XPS of NiFeN and NiMoN@NiFeN in Fig. 2d show (348 and 417 mV), NiMoN (350 and 458 mV), and the bench- two peaks of Fe 2p and Fe 2p at 711.0 and 723.6 eV, mark IrO electrodes (430 and 542 mV). This performance is also 3/2 1/2 2 respectively, as well as a tiny peak at 720.5 corresponding to the superior to that of most non-precious OER catalysts in 1 M KOH satellite peak . In Fig. 2e, the Mo 3d XPS of NiMoN and (Supplementary Table 1), including the recently reported ZnCo 3+ 6+ 33 34 NiMoN@NiFeN show two valence states of Mo and Mo . For oxyhydroxide , Se-doped FeOOH , NiCoFe-MOF (metal- NiMoN, the peak located at 229.6 eV (Mo 3d ) is ascribed to organic frameworks) , and FeNiP/NCH (nitrogen-doped carbon 5/2 3+ 36 Mo in the metal-nitride, which is recognized to be active for hollow framework) . The polarization curves of the CV backward HER . The peaks at 232.7 (Mo 3d ) and 235.3 eV are attributed scan, the CV without and with iR compensation are presented for 3/2 6+ 31 to Mo due to the surface oxidation of NiMoN . However, the comparison in Supplementary Figs. 7, 8, and 9a, respectively. We 3+ 6+ two main peaks of Mo 3d (Mo ) and Mo 3d (Mo ) show also investigated the redox behaviors of the different metal-nitride 5/2 3/2 an apparent negative shift in binding energy for the NiMoN@- catalysts by analyzing their CV curves in the range of about 1.125 NiFeN, indicating the strong electronic interactions between ~1.525 V vs. RHE, and the results are displayed in Supplementary NiMoN and NiFeN. For the N 1 s XPS (Fig. 2f), the main peak is Fig. 9b–d. In addition, the OER activity of other NiMoN@NiFeN located at 397.4 eV, which is ascribed to the N species in metal- catalysts with different loading amounts of NiFeN was also studied nitrides, and another peak at 399.6 eV originates from the (Supplementary Fig. 10), and the one prepared with a precursor 23,32 −1 incomplete reaction of NH . Additionally, the Mo 3p peak ink concentration of 0.25 g ml exhibits the highest OER activity. 3 3/2 also appears for the NiMoN and NiMoN@NiFeN, and a negative Tafel plots in Fig. 3b show that the NiMoN@NiFeN catalyst has a −1 shift in binding energy still exists for the NiMoN@NiFeN, which relatively smaller Tafel slope of 58.6 mV dec in comparison with −1 −1 is in good agreement with the results in Fig. 2e. that of the NiFeN (68.9 mV dec ), NiMoN (82.1 mV dec ), and −1 IrO electrodes (86.7 mV dec ), verifying its rapid OER catalytic kinetics. We further calculated TOF to assess the intrinsic OER activity of the NiMoN@NiFeN catalyst, which presents a TOF Oxygen and hydrogen evolution catalysis.We first evaluated the −1 OER activity of the as-prepared catalysts in 1 M KOH electrolyte value of 0.09 s at an overpotential of 300 mV. This value is not the best among the OER catalysts listed in Supplementary Table 1, in freshwater at room temperature (25 °C). The benchmark IrO catalyst on Ni foam was also included for comparison. All of the but still larger than that of the very good OER catalysts of (Ni,Fe) 12 37 38 OOH ,Fe Co OOH , and NiFe-OH/NiFeP . Impressively, measured potentials vs. Hg/HgO were converted to the reversible x 1−x hydrogen electrode (RHE) according to the reference electrode our 3D core-shell NiMoN@NiFeN catalyst shows very good durability as well for OER in 1 M KOH electrolyte. As revealed in calibration (Supplementary Fig. 6, E = E + 0.925). All RHE Hg/HgO −2 data were measured after cyclic voltammetry (CV) activation and Fig. 3c, the current densities of 100 and 500 mA cm at constant overpotentials show negligible decrease over 48 h OER catalysis, reported with iR compensation (85%). The current density was normalized by the geometrical surface area unless otherwise and the CV polarization curves (inset of Fig. 3c) after the stability test remain almost the same as prior to the test. It should be noted mentioned. As the CV forward scan results in Fig. 3a show, our −2 that for the stability test at 500 mA cm , the current density 3D core-shell NiMoN@NiFeN catalyst exhibits significantly −2 improved OER activity, requiring overpotentials as low as 277 and slightly decreases from 499.5 to 480.9 mA cm with a degrada- −2 −1 −2 tion rate of 0.775 mA cm h , which is mainly attributed to the 337 mV to achieve current densities of 100 and 500 mA cm , respectively, which are considerably smaller than that of NiFeN strong adsorption of bubbles blocking the active sites. Moreover, 4 NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications Intensity (a.u.) Intensity (a.u.) (100) (110) (002) (110) (112) Intensity (a.u.) Intensity (a.u.) Ni 2p Fe LMM Fe 2p Ni LMM O 1s Mo 3p N 1s C 1s Mo 3d Intensity (a.u.) Intensity (a.u.) NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 ARTICLE ab c 500 0.4 337 417 542 mV NiFeN –2 458 500 mA cm @337 mV NiMoN 400 500 NiMoN@NiFeN IrO Initial 0.3 400 After stability test 0.2 200 277 348 350 200 430 mV 100 –2 100 mA cm @277 mV 1.2 1.3 1.4 1.5 1.6 E (V vs. RHE) 0.1 1.2 1.8 1.4 1.5 1.6 1.7 1.8 0.5 1.0 1.5 2.0 0 8 16 24 32 40 48 –2 E (V vs. RHE) log|j (mA cm )| Time (h) de f –2 0.20 100 mA cm @56 mV –100 205 96 84 –100 56 mV Initial 0.15 –200 After stability test –200 –200 –400 –300 0.10 NiFeN –600 –300 NiMoN –400 –800 –2 0.05 500 mA cm @127 mV NiMoN@NiFeN –400 –1000 Pt/C –500 –0.3 –0.2 –0.1 0.0 E (V vs. RHE) 299 252 180 127 mV 0.00 –500 –600 –0.5 –0.4 –0.3 –0.2 –0.1 0.0 1.0 1.5 2.0 0 8 16 24 32 40 48 –2 E ( V vs. RHE) log|j (mA cm ) Time (h) gh –2 100 mA cm NiMoN@NiFeN for OER 1 M KOH 400 400 1 M KOH + 0.5 M NaCl –2 500 mA cm 1 M KOH + seawater –2 300 300 1000 mA cm NiMoN for HER NiMoN for HER 200 200 NiMoN@NiFeN for OER 100 100 0 0 –0.2 –0.1 0.0 1.3 1.4 1.5 1.6 KOH KOH + Nacl KOH + seawater KOH KOH + Nacl KOH + seawater E (V vs. RHE) Fig. 3 Oxygen and hydrogen evolution catalysis. a OER polarization curves in 1 M KOH, and b corresponding Tafel plots of different catalysts. c OER chronoamperometry curves of NiMoN@NiFeN at overpotentials of 277 and 337 mV in 1 M KOH. Inset: CV curves of NiMoN@NiFeN before and after the stability test. d HER polarization curves tested in 1 M KOH, and e corresponding Tafel plots of different catalysts. f HER chronoamperometry curves of NiMoN at overpotentials of 56 and 127 mV in 1 M KOH. Inset: LSV curves of NiMoN before and after the stability test. g OER and HER polarization curves of NiMoN@NiFeN and NiMoN, respectively, in different electrolytes. h Comparison of the overpotentials required to achieve current densities of 100, 500, −2 and 1000 mA cm for NiMoN@NiFeN (OER) and NiMoN (HER) in different electrolytes SEM images after OER stability tests (Supplementary Fig. 11) the NiMoN@NiFeN core-shell catalyst, the highly conductive core demonstrate the high integrity of the 3D core-shell nanostructures of NiMoN nanorods and the robust contact between the NiFeN of the NiMoN@NiFeN catalyst. Thus, the long-term robustness nanoparticles and NiMoN nanorods facilitate the charge transfer mostly originates from its integral 3D core-shell nanostructure between the catalyst and electrolyte, as manifested by the results with different levels of porosity, which benefits rapid gaseous from electrochemical impedance spectroscopy (EIS, Supplemen- product release, and the strong adhesion between the TMN cat- tary Fig. 15), which show that the charge-transfer resistance (R ) ct alysts and the Ni foam substrate. To investigate the origins of of this 3D core-shell electrode is only 1.0 Ω, significantly smaller promoted OER activity in the NiMoN@NiFeN catalyst, we cal- than 9.6 Ω for NiFeN. Additionally, the NiMoN catalyst also has a culated the electrochemical active surface area (ECSA) for the small R of 1.7 Ω,confirming its good electronic conductivity and ct different catalysts by double-layer capacitance (C ) from their CV fast charge transfer. Hence, the rational design of 3D core-shell dl curves (Supplementary Fig. 12) . Clearly, the C values of the TMN catalysts offers a large surface area and efficient charge dl NiMoN and NiMoN@NiFeN catalysts are as large as 188.3 and transfer, both of which contribute to the improved OER activity. −2 238.7 mF cm (Supplementary Fig. 13), respectively, which are To seek a good HER catalyst to combine with our nearly 2.9 and 3.6 times that of the pure NiFeN nanoparticles NiMoN@NiFeN catalyst for overall seawater splitting, we tested −2 (65.4 mF cm ), respectively, demonstrating the highly improved the HER performance of different catalysts, including the ECSA and the increased number of active sites achieved by dec- benchmark Pt/C on Ni foam, in 1 M KOH in freshwater. orating NiFeN nanoparticles on the NiMoN nanorods to form a Strikingly, both the NiMoN@NiFeN and NiMoN catalysts exhibit 3D core-shell nanoarchitecture, which benefits seawater adsorp- exceptional HER activity (Fig. 3d) that is even better than that of 40,41 tion and offers rich active sites for catalytic reactions .We the benchmark Pt/C catalyst, especially the NiMoN catalyst, further normalized current density by the ECSA, and the which requires very low overpotentials of 56 and 127 mV for −2 NiMoN@NiFeN catalyst still shows better OER activity than that current densities of 100 and 500 mA cm , respectively. The of NiFeN (Supplementary Fig. 14), indicating that factors other overpotentials required to achieve the same current densities by than the ECSA also contribute to the enhanced OER activity. For our NiMoN@NiFeN catalyst (84 and 180 mV) are slightly higher, NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications 5 –1 Pt/C: 51.8 mV dec –1 NiMoN@NiFeN: 50.7 mV dec –1 NiMoN@NiFeN: 58.6 mV dec –1 IrO : 86.7 mV dec –1 NiFeN: 68.9 mV dec –1 NiMoN: 45.6 mV dec –1 NiMoN: 82.1 mV dec –1 NiFeN: 96.7 mV dec –2 –2 –2 j (mA cm ) j (mA cm ) j (mA cm ) Overpotential (mV) Overpotential (V) Overpotential (V) –2 –2 j (mA cm ) j (mA cm ) –2 j (mA cm ) Overpotential (mV) ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 but superior to those needed for the Pt/C (96 and 252 mV) and natural seawater, the cell voltages for the corresponding current NiFeN (205 and 299 mV) catalysts. NiMoN has been demon- densities are only 1.774 and 1.901 V. Such performance even strated to be an efficient HER catalyst in alkaline media because outperforms that of most non-noble metal catalysts for alkaline of its excellent electronic conductivity and low adsorption free freshwater splitting, as well as that of the benchmark of Pt/C 24,42,43 15 energy of H* . Fig. 3e reveals that the NiMoN catalyst also and IrO catalysts in 1 M KOH . To boost the industrial appli- −1 exhibits a much smaller Tafel slope of 45.6 mV dec in cations of this electrolyzer, the cell voltages are further decreased comparison to the other catalysts measured. Moreover, the to 1.454, 1.608, and 1.709 V for current densities of 100, 500, and −2 NiMoN catalyst shows good stability at current densities of 100 1000 mA cm , respectively, in 1 M KOH + Seawater electrolyte −2 and 500 mA cm over 48 h HER testing (Fig. 3f). Therefore, our by heating the electrolyte to 60 °C, which can be easily achieved NiMoN@NiFeN and NiMoN catalysts are highly active and by employing a solar thermal hot water system. These values robust for OER and HER, respectively, during freshwater represent the current record-high performance indices for overall electrolysis in alkaline media. alkaline seawater splitting. The overall seawater splitting perfor- We then studied the OER and HER activity in an alkaline mance without iR compensation was also tested in 1 M KOH + simulated seawater electrolyte (1 M KOH + 0.5 M NaCl). As Seawater at 25 °C for comparison (Supplementary Fig. 19), and shown in Fig. 3g, the 3D core-shell NiMoN@NiFeN catalyst still was found to be worse than that with iR compensation. We exhibits outstanding catalytic activity for OER, requiring over- attempted to split pure natural seawater as well, but the perfor- potentials of 286 and 347 mV to achieve current densities of 100 mance is unsatisfactory due to the low ionic conductivity and −2 and 500 mA cm , respectively. This performance is very close to strong corrosiveness of the natural seawater (Supplementary that in the 1 M KOH electrolyte (Fig. 3g), suggesting selective Fig. 20). We then evaluated the Faradaic efficiency of the elec- OER in the alkaline adjusted salty water. We also collected trolyzer in 1 M KOH + 0.5 M NaCl at room temperature by natural seawater from Galveston Bay near Houston, Texas, USA collecting the evolved gaseous products over the cathode and (Supplementary Fig. 16) and prepared an alkaline natural anode (Supplementary Fig. 21). As shown in Fig. 4c, only H and seawater electrolyte (1 M KOH + Seawater), in which the OER O gases with a molar ratio close to 2:1 are detected, and the activity of the NiMoN@NiFeN catalyst shows only slight decay Faradaic efficiency is determined to be around 97.8% during compared with that in the other two electrolytes (Fig. 3g). The seawater electrolysis, demonstrating the high selectivity of OER slight decrease in activity may be due to some insoluble on the anode. precipitates [e.g., Mg(OH) and Ca(OH) ] covering the surface The operating durability is also a very important metric to 2 2 of the electrode, and thus burying some surface active sites assess the performance of an electrolyzer. As shown in Fig. 4d, this (Supplementary Figs. 17 and 18). Even so, the NiMoN@NiFeN electrolyzer can retain outstanding overall seawater splitting −2 catalyst still delivers current densities of 100 and 500 mA cm at performance with no noticeable degradation over 100 h operation −2 small overpotentials of 307 and 369 mV, respectively, in the at a constant current density of 100 mA cm in both the alkaline alkaline natural seawater electrolyte (Fig. 3h). In addition, at an simulated and natural seawater electrolytes. More importantly, −2 even larger current density of 1000 mA cm , the demanded the voltage needed to achieve a very large current density of −2 overpotential is only 398 mV, which is well below the 490 mV 500 mA cm also shows very little increase (<10%) after 100 h overpotential required to trigger chloride oxidation to hypo- water electrolysis in either of the two electrolytes (Fig. 4d), chlorite. Moreover, this overpotential is also much lower than verifying the superior durability of this electrolyzer. The anode of that of any of the other reported non-precious OER catalysts in the NiMoN@NiFeN catalyst further demonstrates good structural alkaline adjusted salty water (Supplementary Table 2). The HER integrity after long-term seawater electrolysis (Supplementary catalyst of NiMoN also exhibits excellent activity in both the Fig. 22). In addition, the electrolyzer exhibits very good activity alkaline simulated and natural seawater electrolytes (Fig. 3g). To and stability (over 600 h electrolysis) for overall seawater splitting −2 deliver current densities of 100, 500, and 1000 mA cm in the in a very harsh condition of 6 M KOH+ Seawater (Supplementary alkaline natural seawater, the required overpotentials are as low as Fig. 23), demonstrating its great potential for large-scale applica- 82, 160, and 218 mV, respectively (Fig. 3h). Consequently, our tions. Given its excellent catalytic performance, this electrolyzer NiMoN@NiFeN and NiMoN catalysts are not only efficient for can be easily actuated by a 1.5 V AA battery (Supplementary freshwater electrolysis, but also highly active for alkaline seawater Fig. 24). Moreover, we also demonstrated the harvesting of waste splitting. heat, the major energy loss in various activities and device operations, by our seawater electrolyzer powered with a commer- cial TE device that directly coverts heat into electricity (Fig. 4e) . Overall seawater splitting. Considering the outstanding catalytic As shown in Fig. 4f, when the temperature gradient between the performance of both the NiMoN@NiFeN and NiMoN catalysts, hot and cold sides of the TE module is 40, 50, and 60 °C, we further investigated the overall seawater splitting performance the corresponding output voltage can expeditiously drive the by integrating the two catalysts into a two-electrode alkaline electrolyzer for stable delivery of current density of 30, 100, and −2 electrolyzer (without a diaphragm or membrane), in which 200 mA cm , respectively. Even when the temperature gradient NiMoN@NiFeN is used as the anode for OER and NiMoN as the through the TE module is decreased to 40 °C, the electrolyzer can −2 cathode for HER (Fig. 4a). Remarkably, this electrolyzer shows still supply a current density of ~30 mA cm with good excellent overall seawater splitting activity in both the alkaline recyclability, suggesting that we can efficiently use the waste heat simulated and natural seawater electrolytes. As displayed in to produce H fuel by the electrolysis of seawater. Fig. 4b, at room temperature (25 °C), the cell voltages needed to −2 produce a current density of 100 mA cm are as low as 1.564 and 1.581 V in 1 M KOH + 0.5 M NaCl and 1 M KOH + Sea- Active sites for oxygen evolution catalysis. To gain a deeper water electrolytes, respectively. In particular, our electrolyzer insight into the real catalytic active sites for the extraordinary can generate extremely large current densities of 500 and OER activity of the NiMoN@NiFeN catalyst, we further studied −2 1000 mA cm at 1.735 and 1.841 V, respectively, in 1 M KOH + its nanostructure, surface composition, and chemical state during 0.5 M NaCl electrolyte, which is slightly better than the recently and after OER tests. The TEM image in Fig. 5a shows that the 3D reported anion exchange membrane (AEM) based electrolyzer in core-shell nanostructure of NiMoN@NiFeN is intact after OER an alkaline simulated seawater electrolyte . Even in the alkaline tests, which is consistent with the SEM results (Supplementary 6 NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications NiMoN@NiFeN NiMoN NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 ARTICLE a be 1 M KOH + 0.5 M NaCl@25° Hot side 1 M KOH + seawater@25° 800 1 M KOH + 0.5 M NaCl@60° PN H O 2 1 M KOH + seawater@60° –2 500 mA cm H O –2 100 mA cm Cold side 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 Voltage (V) cd f 3.0 2.0 Calculated H 2 1 M KOH + 0.5 M NaCl 1.9 Calculated O ΔT = 60 °C –2 1.8 500 mA cm 2.5 Measured H 1.7 –2 Measured O 100 mA cm 1.6 2.0 1.5 1.4 1.5 ΔT = 50 °C 2.0 1 M KOH + seawater –2 1.9 500 mA cm 1.0 1.8 –2 1.7 ΔT = 40 °C 100 mA cm 0.5 1.6 ΔT = 40 °C 1.5 0 0.0 1.4 0 20 40 60 80 100 0 20406080 100 0 100 200 300 400 500 600 Time (min) Time (h) Time (min) Fig. 4 Overall seawater splitting performance. a Schematic illustration of an overall seawater splitting electrolyzer with NiMoN and NiMoN@NiFeN as the cathode and anode, respectively. b Polarization curves after iR compensation of NiMoN and NiMoN@NiFeN coupled catalysts in a two-electrode electrolyzer tested in alkaline simulated (1 M KOH + 0.5 M NaCl, resistance: ~1.1 Ω) and natural seawater (1 M KOH + Seawater, resistance: ~1.2 Ω) electrolytes under different temperatures. c Comparison between the amount of collected and theoretical gaseous products (H and O ) by the two- 2 2 −2 electrode electrolyzer at a constant current density of 100 mA cm in 1 M KOH + 0.5 M NaCl at 25 °C. d Durability tests of the electrolyzer at constant −2 current densities of 100 and 500 mA cm in different electrolytes at 25 °C. e Schematic illustration of the principle for power generation between the hot and cold sides of a TE device. f Real-time dynamics of current densities for the electrolyzer in 1 M KOH + 0.5 M NaCl at 25 °C driven by a TE device when the temperature gradient (ΔT) between its hot and cold sides is 40, 50, 60, and 40 °C Fig. 11). The TEM image in Fig. 5b reveals that many nano- main peaks at 531.9 and 530.1 eV and the appearance of a new particles are closely attached on the nanorod, and there seems to peak at 532.3 eV . To confirm the formation of NiFe oxides/oxy be some very thin layers on the nanoparticle surface. The (hydroxides), we further performed in situ Raman measurements HRTEM image in Fig. 5c confirms the existence of thin amor- (Supplementary Fig. 28) to elucidate the real-time evolution of phous layers and Ni(OH) . We suspect that the thin layers are the NiMoN@NiFeN catalyst during the OER process. As the in situ generated amorphous NiFe oxides and NiFe oxy(hydro- results in Fig. 5g show, the spectrum for the as-prepared xides), which have been verified by elemental mapping and XPS NiMoN@NiFeN exhibits a sharp and broad peak at around −1 analyses following OER testing. Figure 5d displays the DF-STEM 80.3 cm , which is probably due to the metal-N stretching and corresponding elemental mapping images, which show the modes. The transformation into NiOOH starts at 1.4 V according −1 12 absence of N and the increased O content on the NiMoN@NiFeN to a new Raman band located at 480.1 cm . When the surface after OER due to the intense oxidation process. The high- potential reaches to 1.6 and 1.7 V, two additional Raman bands −1 resolution XPS of N 1s (Supplementary Fig. 25) also corroborates are generated. The one located at 324.7 cm is assigned to the 50 −1 this point (the surface N content in the NiMoN@NiFeN catalyst Fe-O vibrations in Fe O , and the other at 693.1 cm belongs 2 3 was reduced from 10.3% in the fresh sample to 0.36% after OER). to the Fe-O vibrations in amorphous FeOOH . Therefore, by For the high-resolution XPS of Ni 2p (Fig. 5e), the two peaks combining these results with the XPS results, we conclude that attributed to Ni-N species at 853.4 and 870.8 eV also disappear thin amorphous layers of NiFe oxide and NiFe oxy(hydroxide) after OER because of surface oxidation. A new peak at 868.9 eV, are evolved from the NiFeN nanoparticles at the surface during which is assigned to Ni(OH) , shows up,. Besides, the two peaks OER electrocatalysis, and that these serve as the real active sites at 856.3 (Ni-O) and 862.0 eV (Sat.) positively shift toward higher participating in the OER process. The formation of a metal binding energy, which is also observed in the XPS of Fe 2p nitride-metal oxide/oxy(hydroxide) core-shell structure may also 2+ 2+ (Fig. 5f), indicating the oxidation of Ni and Fe to the higher facilitate electron transfer from the NiFeN core to the oxidized 3+ 3+ valence states of Ni and Fe (Supplementary Fig. 26), species (Supplementary Fig. 29). This observation is consistent respectively, resulting from the formation of NiFe oxides/oxy with the results of other reported OER catalysts, including metal 46–48 14,46 (hydroxides) . The O 1s XPS (Supplementary Fig. 27) also selenides and phosphides . However, the structure of the 2+ 2+ proves the increased valence states of Ni and Fe after OER, in situ formed NiFe oxides/oxy(hydroxides) is different from that as well as showing the appearance of Fe-OH from the NiFe oxy of the (Ni,Fe)OOH thin-film catalyst reported by Zhou et al. , (hydroxides), which can be seen from the negative shift of the which undergoes a rapid self-reconstruction due to the partial NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications 7 Gas amount (mmol) –2 j (mA cm ) Voltage (V) –2 j (mA cm ) Heat flow ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 Ni Mo ac b d Amorphous NiFeOOH layer Fe N O 0.261 nm Ni(OH) (101) e f g Ni 2p Before OER Before OER In situ raman 3/2 –1 Fe 2p Ni 2p Fe 2p 80.3 cm 3/2 –1 1/2 1/2 693.1 cm Ni-N Sat. Ni-O –1 324.7 cm –1 1.7 V Ni-O 480.1 cm Sat. Ni-N Sat. 1.6 V 3+ Ni 3+ 3+ Fe Fe After OER 3+ Fe After OER Ni(OH) 3+ Ni 1.5 V 1.4 V KOH As prepared 888 882 876 870 864 858 852 730 725 720 715 710 705 700 100 200 300 400 500 600 700 800 –1 Binding energy (eV) Binding energy (eV) Raman shift (cm ) Fig. 5 Material characterization to study the OER active sites. a, b TEM images of NiMoN@NiFeN core-shell nanorods at different magnifications after OER tests. c HRTEM image, and d DF-STEM image and corresponding elemental mapping of the NiMoN@NiFeN catalyst after OER tests. Scale bars: a 500 nm; b 50 nm; c 5 nm; d 1 µm. High-resolution XPS of (e), Ni 2p and (f), Fe 2p of NiMoN@NiFeN after OER tests in comparison with those before OER tests. g In situ Raman spectra of the NiMoN@NiFeN catalyst at various potentials for the OER process dissolution of FeOOH in KOH solution, forming amorphous step in the development of a robust and active catalyst to utilize the NiOOH nanoarrays mixed with a small amount of FeOOH world’s abundant seawater feedstock for large-scale hydrogen nanoparticles after OER. Notably, such in situ generated amor- production by renewable energy sources. phous NiFe oxide and NiFe oxy(hydroxide) layers also play a positive role in improving the resistance to corrosion by chloride Methods anions in seawater (Supplementary Fig. 30), which contributes to Chemicals. Ethanol (C H OH, Decon Labs, Inc.), ammonium heptmolybdate 2 5 the superior stability during seawater electrolysis. [(NH ) Mo O ·4H O, 98%, Sigma-Aldrich], nickel(II) nitrate hexahydrate (Ni 4 6 7 24 2 (NO ) ·6H O, 98%, Sigma-Aldrich), iron (III) nitrate hexahydrate (Fe 3 2 2 (NO ) ·9H O, 98%, Sigma-Aldrich), N, N Dimethylformamide [DMF, (CH ) NC 3 3 2 3 2 (O)H, anhydrous, 99.8%, Sigma-Aldrich], platinum powder (Pt, nominally 20% on Discussion carbon black, Alfa Aesar), iridium oxide powder (IrO2, 99%, Alfa Aesar), Nafion In summary, we have developed a 3D core-shell OER catalyst of (117 solution, 5% wt, Sigma-Aldrich), sodium chloride (NaCl, Fisher Chemical), NiMoN@NiFeN for active and stable alkaline seawater splitting. potassium hydroxide (KOH, 50% w/v, Alfa Aesar), and Ni foam (thickness: 1.6 mm, porosity: ~95%) were used as received. Deionized (DI) water (resistivity: The interior NiMoN nanorods are highly conductive and afford a 18.3 MΩ·cm) was used for the preparation of all aqueous solutions. large surface area, which ensure efficient charge transfer and numerous active sites. The outer NiFeN nanoparticles in situ evolve Synthesis of NiMoO nanorods on Ni foam. NiMoO nanorods were synthesized thin amorphous layers of NiFe oxide and NiFe oxy(hydroxide) 4 4 on nickel foam through a hydrothermal method . A piece of commercial Ni foam during OER catalysis, which are not only responsible for the (2 × 5 cm ) was cleaned by ultrasonication with ethanol and DI water for several selective OER activity, but also beneficial for the corrosion resis- minutes, and the substrate was then transferred into a polyphenyl (PPL)-lined tance to chloride anions in seawater. At the same time, the 3D core- stainless-steel autoclave (100 ml) containing a homogenous solution of Ni shell nanostructures with multiple levels of porosity are favorable (NO ) ·6H O (0.04 M) and (NH ) Mo O ·4H O (0.01 M) in 50 ml H O. After- 3 2 2 4 6 7 24 2 2 ward, the autoclave was sealed and maintained at 150 °C for 6 h. The sample was for seawater diffusion and H /O gases releasing. Thus, this OER 2 2 then taken out and washed with DI water and ethanol several times before being catalyst requires very low overpotentials of 369 and 398 mV to fully dried at 60 °C overnight under vacuum. −2 deliver large current densities of 500 and 1000 mA cm , respec- tively, in alkaline natural seawater at 25 °C. Additionally, by pairing Synthesis of NiMoN@NiFeN core-shell nanorods. The metal nitrides were it with another efficient HER catalyst of NiMoN, we assembled an synthesized by one-step nitridation of the NiMoO nanorods in a tube furnace. For outstanding water electrolyzer for overall seawater splitting, which the synthesis of NiMoN nanorods, a piece of NiMoO /Ni foam (~1 cm ) was −2 placed at the middle of a tube furnace and thermal nitridation was conducted at outputs current densities of 500 and 1000 mA cm at record low 500 °C under a flow of 120 standard cubic centimeters (sccm) NH and 30 sccm Ar voltages of 1.608 and 1.709 V, respectively, in alkaline natural for 1 h. The furnace was then automatically turned off and naturally cooled down seawater at 60 °C. The electrolyzer also shows excellent durability at to room temperature under Ar atmosphere. For the synthesis of NiMoN@NiFeN −2 current densities of 100 and 500 mA cm during up to 100 h 2 core-shell nanorods, a piece of NiMoO /Ni foam (~1 cm ) was first soaked in a alkaline seawater electrolysis. This discovery represents a significant NiFe precursor ink, which was prepared by dissolving Ni(NO ) ·6H O and Fe 3 2 2 8 NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications Intensity (a.u.) Intensity (a.u.) Intensity (a.u.) NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 ARTICLE (NO ) ·9H O with a mole ratio of 1:1 in DMF, and the NiMoO /Ni foam coated Overall seawater splitting driven by a TE module. We purchased a commercial 3 3 2 4 with the NiFe precursor ink was then dried at ambient condition. The dried sample TE module from Amazon and used it as a power generator to drive our two- was subjected to thermal nitridation under the same conditions as for NiMoN. To electrode electrolyzer according to our previous work . During the test, the hot study the effect of the NiFeN loading amount on the morphology of the core-shell side of the TE module was covered by a large flat copper plate, which was in direct nanorods, we prepared four different NiMoN@NiFeN core-shell nanorods with contact with a heater on top. The hot-side temperature was maintained relatively different loading amounts of NiFeN by controlling the concentration of Ni and Fe constant by tuning the DC power supply to the heater, while the cold-side tem- −1 precursors. Specifically, 0.1, 0.25, 0.5, and 0.75 g ml concentrations of precursor perature was controlled by placing it in direct contact with a cooling system, where ink were used. For comparison, pure NiFeN nanoparticles were also prepared on the water inside was adjusted to remain at a constant temperature. Thus, the TE the Ni foam by replacing the NiMoO /Ni foam with Ni foam. The concentration of module generated a relatively stable open circuit voltage between the hot and cold −1 precursor ink in this case was 0.25 g ml , and all other synthesis conditions were sides. A nano-voltmeter and an ammeter were embedded into the circuit for real- the same as for NiMoN@NiFeN. time monitoring of the voltage and current, respectively, between the two elec- trodes of the water-splitting cell. Preparation of IrO and Pt/C catalysts on Ni foam. To prepare the IrO elec- 2 2 trode for comparison , 40 mg of IrO and 60 μLofNafion were dispersed in Data availability 540 μL of ethanol and 400 μL of DI water, and the mixture was ultrasonicated for The source data underlying Figs. 1i, 2a–f, 3c, f and h, 4c and f, 5e–g, and Supplementary 30 min. The dispersion was then coated onto a Ni foam substrate, which was dried Figs. 2a, 6, 12, 20, 21a, 23a, 25, 26, and 27 are provided as a Source Data file. The other in air overnight. Pt/C electrodes were obtained by the same method. data that support the findings of this work are available from the corresponding authors upon reasonable request. Materials characterization. The morphology and nanostructure of the samples were determined by scanning electron microscopy (SEM, LEO 1525) and trans- Code availability mission electron microscopy (TEM, JEOL 2010F) coupled with energy dispersive All plots and data analysis were performed with OriginPro 8.5 software, and will be made X-ray (EDX) spectroscopy. The phase composition of the samples was character- available by the corresponding authors upon reasonable request. ized by X-ray diffraction (PANalytical X’pert PRO diffractometer with a Cu Ka radiation source) and XPS (PHI Quantera XPS) using a PHI Quantera SXM scanning X-ray microprobe. Received: 18 June 2019; Accepted: 18 October 2019; Electrochemical tests. The electrochemical performance was tested on an elec- trochemical station (Gamry, Reference 600). The two half reactions of OER and HER were each carried out at room temperature (~25 °C) in a standard three- electrode system with our prepared sample as the working electrode, a graphite rod as the counter electrode, and a standard Hg/HgO electrode as the reference elec- References trode. Four different electrolytes, including 1 M KOH, 1 M KOH + 0.5 M NaCl, 1. Tiwari, J. N. et al. Multicomponent electrocatalyst with ultralow Pt loading 1 M KOH + Seawater, and natural seawater, were used, and the pH was around 14 and high hydrogen evolution activity. Nat. Energy 3, 773 (2018). −2 except for the natural seawater (pH~7.2). The electrode size is around 1 cm , and 2. Turner, J. A. Sustainable hydrogen production. Science 305, 972–974 (2004). −2 the effective parts in the electrolyte are 0.3 ~ 0.45 cm for different samples. Both 3. Cabán-Acevedo, M. et al. Efficient hydrogen evolution catalysis using ternary the anodes (NiMoN@NiFeN) and cathodes (NiMoN) were cycled ~100 times by pyrite-type cobalt phosphosulphide. Nat. Mater. 14, 1245 (2015). CV until a stable polarization curve was developed prior to measuring each 4. Luo, J. et al. Water photolysis at 12.3% efficiency via perovskite photovoltaics polarization curve. OER polarization curve measurements were performed by CV and earth-abundant catalysts. Science 345, 1593–1596 (2014). −1 at a scan rate of 2 mV s and stability tests were carried out under constant 5. Pu, Z. et al. A universal synthesis strategy for P-rich noble metal diphosphide overpotentials. HER polarization curve measurements were performed by linear based electrocatalyst for hydrogen evolution reactions. Energy Environ. Sci. 12, −1 sweep voltammetry (LSV) at a scan rate of 5 mV s and stability tests were carried 952–957 (2019). out under constant overpotentials. Electrochemical impedance spectra (EIS) were 6. Roger, I., Shipman, M. A. & Symes, M. D. Earth-abundant catalysts for measured at an overpotential of 270 mV from 0.1 Hz to 100 KHz with an ampli- electrochemical and photoelectrochemical water splitting. Nat. Rev. Chem. 1, tude of 10 mV. For the two-electrode seawater electrolysis, the as-prepared 0003 (2017). NiMoN@NiFeN and NiMoN catalysts (after CV activation) were used as the anode 7. Seh, Z. W. et al. Combining theory and experiment in electrocatalysis: Insights and cathode, respectively. The polarization curves were measured in different into materials design. Science 355, eaad4998 (2017). electrolytes at different temperatures (25 and 60 °C), and stability tests were carried −2 8. Ledezma-Yanez, I. et al. Interfacial water reorganization as a pH-dependent out under constant current densities of 100 and 500 mA cm at room descriptor of the hydrogen evolution rate on platinum electrodes. Nat. Energy temperature. 2, 17031 (2017). 9. Ding, Q., Song, B., Xu, P. & Jin, S. Efficient electrocatalytic and Turnover frequency (TOF) calculation. The TOF value is calculated from the photoelectrochemical hydrogen generation using MoS and related 54,55 Eq. (1) : compounds. Chem 1, 699–726 (2016). j  A  η 10. Landman, A. et al. Photoelectrochemical water splitting in separate oxygen TOF ¼ ð1Þ 4  F  n and hydrogen cells. Nat. Mater. 16, 646 (2017). 11. Fabbri, E. et al. Dynamic surface self-reconstruction is the key of highly active where j is the current density at a given overpotential, A is the surface area of the perovskite nano-electrocatalysts for water splitting. Nat. Mater. 16, 925 electrode, η is the OER Faradaic efficiency, F is the Faraday constant, and n is the (2017). total number of active sites on the electrode. To calculate TOF, the most significant 12. Zhou, H. et al. Water splitting by electrolysis at high current densities under challenge is to identify the number of active sites. Since the real active sites for the 1.6 volts. Energy Environ. Sci. 11, 2858–2864 (2018). NiMoN@NiFeN catalyst are in situ formed NiFe oxide and NiFe oxy(hydroxide), 13. Hui, L. et al. Overall water splitting by graphdiyne-exfoliated and- we calculated the number of active sites by integrating the Ni redox peaks from the sandwiched layered double-hydroxide nanosheet arrays. Nat. Commun. 9, CV curve according to the following Eq. (2) : 5309 (2018). Q  N 14. Yu, F. et al. High-performance bifunctional porous non-noble metal n ¼ ð2Þ F  R Ni=Fe phosphide catalyst for overall water splitting. Nat. Commun. 9, 2551 (2018). 15. Yu, L. et al. Cu nanowires shelled with NiFe layered double hydroxide where Q is the integration of Ni redox features from the CV curve (Supplementary nanosheets as bifunctional electrocatalysts for overall water splitting. Energy Fig. 9d), N is Avogadro’s constant, F is the Faraday constant, and R is the A Ni/Fe 2+ 3+ Environ. Sci. 10, 1820–1827 (2017). molar ratio of Ni/Fe, assuming that Ni /Ni is a one-electron process. For the −2 16. Dresp, S., Dionigi, F., Klingenhof, M. & Strasser, P. Direct electrolytic splitting NiMoN@NiFeN catalyst at overpotential of 300 mV, j = 206 mA cm , and thus −1 TOF was calculated to be 0.09 s at overpotential of 300 mV. of seawater: Opportunities and challenges. ACS Energy Lett. 4, 933–942 (2019). 17. Hsu, S. H. et al. An earth‐abundant catalyst‐based seawater photoelectrolysis Gas chromatography measurement. Overall seawater splitting for gas chroma- system with 17.9% solar‐to‐hydrogen efficiency. Adv. Mater. 30, 1707261 tography (GC, GOW-MAC 350 TCD) tests were performed in a gas-tight elec- (2018). trochemical cell with 1 M KOH + 0.5 M NaCl as the electrolyte at room 18. Kuang, Y. et al. Solar-driven, highly sustained splitting of seawater into temperature (25 °C). Chronopotentiometry was applied with a constant current hydrogen and oxygen fuels. Proc. Natl Acad. Sci. USA 116, 6624–6629 (2019). −2 density of 100 mA cm to maintain oxygen and hydrogen generation. For each 19. Dionigi, F., Reier, T., Pawolek, Z., Gliech, M. & Strasser, P. Design criteria, measurement over an interval of 10 min, 0.3 µL gas sample was carefully extracted operating conditions, and nickel-iron hydroxide catalyst materials for selective from the sealed cell and injected into the GC instrument using a glass syringe seawater electrolysis. ChemSusChem 9, 962–972 (2016). (Hamilton Gastight 1002). NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13092-7 20. Zhao, Y. et al. Charge state manipulation of cobalt selenide catalyst for overall 48. Hu, C. et al. Synergism of geometric construction and electronic regulation: seawater electrolysis. Adv. Energy Mater. 8, 1801926 (2018). 3D Se‐(NiCo) S /(OH) nanosheets for highly efficient overall water splitting. x x 21. Esswein, A. J., Surendranath, Y., Reece, S. Y. & Nocera, D. G. Highly active Adv. Mater. 30, 1705538 (2018). cobalt phosphate and borate based oxygen evolving catalysts operating in 49. Zhou, H. et al. Highly active catalyst derived from a 3D foam of Fe(PO ) / 3 2 neutral and natural waters. Energy Environ. Sci. 4, 499–504 (2011). Ni P for extremely efficient water oxidation. Proc. Natl Acad. Sci. USA 114, 22. Zhang, Y. et al. 3D porous hierarchical nickel-molybdenum nitrides 5607–5611 (2017). synthesized by RF plasma as highly active and stable hydrogen‐evolution‐ 50. Song, F. et al. An unconventional iron nickel catalyst for the oxygen evolution reaction electrocatalysts. Adv. Energy Mater. 6, 1600221 (2016). reaction. ACS Cent. Sci. 5, 558–568 (2019). 23. Song, F. et al. Interfacing nickel nitride and nickel boosts both electrocatalytic 51. Chemelewski, W. D., Lee, H.-C., Lin, J.-F., Bard, A. J. & Mullins, C. B. hydrogen evolution and oxidation reactions. Nat. Commun. 9, 4531 (2018). Amorphous FeOOH oxygen evolution reaction catalyst for 24. Wang, Y. et al. Self-supported NiMo-based nanowire arrays as bifunctional photoelectrochemical water splitting. J. Am. Chem. Soc. 136, 2843–2850 electrocatalysts for full water splitting. J. Mater. Chem. A 6, 8479–8487 (2018). (2014). 25. Zhu, C. et al. Fe‐Ni‐Mo nitride porous nanotubes for full water splitting and 52. Zhang, J. et al. Efficient hydrogen production on MoNi electrocatalysts with Zn‐air batteries. Adv. Energy Mater. 8, 1802327 (2018). fast water dissociation kinetics. Nat. Commun. 8, 15437 (2017). 26. Feng, L.-L. et al. High-index faceted Ni S nanosheet arrays as highly active 53. Yu, F. et al. Three-dimensional nanoporous iron nitride film as an efficient 3 2 and ultrastable electrocatalysts for water splitting. J. Am. Chem. Soc. 137, electrocatalyst for water oxidation. ACS Catal. 7, 2052–2057 (2017). 14023–14026 (2015). 54. Song, F. & Hu, X. Exfoliation of layered double hydroxides for enhanced 27. Xiao, M. et al. Hollow nanostructures for photocatalysis: Advantages and oxygen evolution catalysis. Nat. Commun. 5, 4477 (2014). challenges. Adv. Mater. 31, 1801369 (2018). 55. Zhang, B. et al. Homogeneously dispersed multimetal oxygen-evolving 28. Yu, L. et al. Ternary Ni Mo P nanowire arrays toward efficient and stable catalysts. Science 352, 333–337 (2016). 2(1-x) 2x hydrogen evolution electrocatalysis under large-current-density. Nano Energy 53, 492–500 (2018). Acknowledgements 29. Qiu, B. et al. Fabrication of nickel-cobalt bimetal phosphide nanocages The work performed in China is financially supported by the National Natural Science for enhanced oxygen evolution catalysis. Adv. Funct. Mater. 28, 1706008 Foundation of China (Nos. 21573085 and 51872108), the Wuhan Planning Project of (2018). Science and Technology (No. 2018010401011294), and self-determined research funds of 30. Wang, T. et al. NiFe (oxy) hydroxides derived from NiFe disulfides as an CCNU from the college’s basic research and operation through the Chinese Ministry of efficient oxygen evolution catalyst for rechargeable Zn-air batteries: the effect Education (No. CCNU18TS034). J.B. acknowledges the support from the Robert A. of surface S residues. Adv. Mater. 30, 1800757 (2018). Welch Foundation (E-1728). Z.R. acknowledges the Research Award from the Alexander 31. Wu, A. et al. Integrating the active OER and HER components as the von Humboldt Foundation and Prof. Kornelius Nielsch at IFW Dresden Germany. heterostructures for the efficient overall water splitting. Nano Energy 44, 353–363 (2018). 32. Yan, H. et al. Anion‐modulated HER and OER activities of 3D Ni-V‐based Author contributions interstitial compound heterojunctions for high‐efficiency and stable overall Z.F.R. and Y.Y. led the project. L.Y. designed and performed most of the experiments and water splitting. Adv. Mater. 31, 1901174 (2019). analyzed most of the data including material synthesis, characterization, and electro- 33. Huang, Z.-F. et al. Chemical and structural origin of lattice oxygen oxidation chemical tests. Q.Z. carried out the thermoelectric measurements. S.W.S. performed the in Co-Zn oxyhydroxide oxygen evolution electrocatalysts. Nat. Energy 4, XPS characterization. B.M. and D.Z.W. took the TEM images. Z.J.Q. and J.M.B. per- 329–338 (2019). formed the in situ Raman tests. C.Z.W. and J.M.B. performed the GC tests. L.Y., S.C., Y. 34. Niu, S. et al. Se-doping activates FeOOH for cost-effective and efficient Y., and Z.F.R. wrote the paper. All of the authors have discussed the results and revised electrochemical water oxidation. J. Am. Chem. Soc. 141, 7005–7013 (2019). the paper together. 35. Qian, Q., Li, Y., Liu, Y., Yu, L. & Zhang, G. Ambient fast synthesis and active sites deciphering of hierarchical foam‐like trimetal-organic framework nanostructures as a platform for highly efficient oxygen evolution Competing interests The authors declare no competing interests. electrocatalysis. Adv. Mater. 31, 1901139 (2019). 36. Wei, Y.-S. et al. A single-crystal open-capsule metal-organic framework. J. Am. Chem. Soc. 141, 7906–7916 (2019). Additional information 37. Ye, S. H., Shi, Z. X., Feng, J. X., Tong, Y. X. & Li, G. R. Activating CoOOH Supplementary information is available for this paper at https://doi.org/10.1038/s41467- porous nanosheet arrays by partial iron substitution for efficient oxygen 019-13092-7. evolution reaction. Angew. Chem. Int. Ed. 57, 2672–2676 (2018). 38. Liang, H. et al. Amorphous NiFe-OH/NiFeP electrocatalyst fabricated at low Correspondence and requests for materials should be addressed to Y.Y., S.C. or Z.R. temperature for water oxidation applications. ACS Energy Lett. 2, 1035–1042 (2017). Peer review information Nature Communications thanks the anonymous reviewers for 39. Gao, S. et al. Partially oxidized atomic cobalt layers for carbon dioxide their contributions to the peer review of this work. electroreduction to liquid fuel. Nature 529, 68 (2016). 40. Zhang, J., Zhao, Z., Xia, Z. & Dai, L. A metal-free bifunctional electrocatalyst Reprints and permission information is available at http://www.nature.com/reprints for oxygen reduction and oxygen evolution reactions. Nat. Nanotech. 10, 444 (2015). Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in 41. Yu, F. et al. Recent developments in earth-abundant and non-noble published maps and institutional affiliations. electrocatalysts for water electrolysis. Mater. Today Phys. 7, 121–138 (2018). 42. Chang, B. et al. Bimetallic NiMoN nanowires with a preferential reactive facet: An ultraefficient bifunctional electrocatalyst for overall water splitting. ChemSusChem 11, 3198–3207 (2018). Open Access This article is licensed under a Creative Commons 43. Jia, J. et al. Nickel molybdenum nitride nanorods grown on Ni foam as Attribution 4.0 International License, which permits use, sharing, efficient and stable bifunctional electrocatalysts for overall water splitting. ACS adaptation, distribution and reproduction in any medium or format, as long as you give Appl. Mater. Interfaces 10, 30400–30408 (2018). appropriate credit to the original author(s) and the source, provide a link to the Creative 44. Dresp, S. et al. Direct electrolytic splitting of seawater: activity, selectivity, Commons license, and indicate if changes were made. The images or other third party degradation, and recovery studied from the molecular catalyst structure to the material in this article are included in the article’s Creative Commons license, unless electrolyzer cell level. Adv. Energy Mater. 8, 1800338 (2018). indicated otherwise in a credit line to the material. If material is not included in the 45. He, J. & Tritt, T. M. Advances in thermoelectric materials research: looking article’s Creative Commons license and your intended use is not permitted by statutory back and moving forward. Science 357, eaak9997 (2017). regulation or exceeds the permitted use, you will need to obtain permission directly from 46. Xu, X., Song, F. & Hu, X. A nickel iron diselenide-derived efficient oxygen- the copyright holder. To view a copy of this license, visit http://creativecommons.org/ evolution catalyst. Nat. Commun. 7, 12324 (2016). licenses/by/4.0/. 47. He, Q. et al. Highly defective Fe-based oxyhydroxides from electrochemical reconstruction for efficient oxygen evolution catalysis. ACS Energy Lett. 3, © The Author(s) 2019 861–868 (2018). 10 NATURE COMMUNICATIONS | (2019) 10:5106 | https://doi.org/10.1038/s41467-019-13092-7 | www.nature.com/naturecommunications

Journal

Nature CommunicationsSpringer Journals

Published: Nov 8, 2019

There are no references for this article.