Fast site-to-site electron transfer of high-entropy alloy nanocatalyst driving redox electrocatalysis — TEI-tagged view

Hongdong LipersName, Yi HanpersName, Huan ZhaopersName, Wenjing QipersName, Dan ZhangpersName, Yaodong YupersName, Wenwen CaipersName, Shaoxiang LipersName, Jianping LaipersName, Bolong HuangpersName, Lei WangpersName (2020date). “Fast site-to-site electron transfer of high-entropy alloy nanocatalyst driving redox electrocatalysis.” Nature Communicationstitle·j.

Object of study — ultrasmall PtNiFeCoCu high-entropy alloyterm·dom nanoparticlesterm·key as bifunctional electrocatalysts

Research question — how to design electrocatalysts that perform well for both reduction and oxidation

Contribution — a low-temperature synthesis of HEAterm·dom nanoparticlesterm·key with strong HERterm·dom and MORterm·dom performance and fast site-to-site electron transferterm·key

Full text — tagged

The reference list is not part of the text we hold for this paper. Our copy was extracted from the publisher's page and the bibliography did not survive that extraction, so the article below ends before its references. Nothing has been removed by us.

Fast site-to-site electron transferterm·key of high-entropy alloyterm·dom nanocatalyst driving redox electrocatalysisterm·key Abstract Designing electrocatalysts with high-performance for both reduction and oxidation reactions faces severe challenges. Here, the uniform and ultrasmall (~3.4 nm) high-entropy alloysterm·dom (HEAsterm·dom) Pt 18 Ni 26 Fe 15 Co 14 Cu 27 nanoparticlesterm·key are synthesized by a simple low-temperature oil phase strategy at atmospheric pressure. The Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst exhibits excellent electrocatalytic performance for hydrogen evolution reactionterm·dom (HERterm·dom) and methanol oxidation reactionterm·dom (MORterm·dom). The catalyst shows ultrasmall overpotentialterm·key of 11 mV at the current density of 10 mA cm −2 , excellent activity (10.96 A mg −1 Pt at −0.07 V vs. reversible hydrogen electrode) and stability in the alkaline medium. Furthermore, it is also the efficient catalyst (15.04 A mg −1 Pt ) ever reported for MORterm·dom in alkaline solution. Periodic DFTterm·key calculations confirm the multi-active sites for both HERterm·dom and MORterm·dom on the HEAterm·dom surface as the key factor for both proton and intermediate transformation. Meanwhile, the construction of HEAterm·dom surfaces supplies the fast site-to-site electron transferterm·key for both reduction and oxidation processes.

Introduction Electrocatalytic processes play a vital role in energy conversion and reducing the environmental pollution 1 – 5 . To improve the activity, selectivity, and stability of the catalytic reaction, it is necessary to develop high-performance advanced catalysts that can meet the needs of rapid development 6 – 12 . High-entropy alloysterm·dom (HEAsterm·dom) have attracted wide interest as catalytic materials in the past few years 13 – 16 . The alloy contains five or more elements which have similar atomic ratios 16 , 17 . The atomic sizeterm·dom of each component is different, which can cause lattice distortionterm·dom 18 . Besides, the presence of multiple components is conducive to promoting the formation of the solid solution phase and inhibiting the movement of dislocations. These characteristics of HEAterm·dom lead to some unique characteristics, such as strong fracture toughness, corrosion resistance, and high mechanical strength 19 – 21 . The ultimate goal of HEAterm·dom is to adapt these characteristics to any desired response by using almost infinite possible combinations of elements and modifying their composition. In catalysis field, the adsorption of molecules and intermediates species on the surface of the catalyst affects the catalytic activity 13 , 22 . These adsorption energies can be adjusted by alloying as compared to pure elements to increase catalytic activity 23 – 25 . Recently, some HEAsterm·dom had used as catalysts for electrocatalytic reactions, which display superior stability and catalytic selectivity and activity compared with traditional alloysterm·dom 13 , 26 – 30 . However, the traditional method mainly produce bulk HEAsterm·dom rather than nanostructures 26 , 31 – 34 . Moreover, the preparation of uniform nanostructured HEAsterm·dom with small size (<10 nm) currently requires specific equipment (fast heating/cooling rate, ~10 5 K per second), high temperature (~2000 kelvin), and high temperature resistant and conductive substrate (carbon nanofiber), such as carbon-thermal shock method 35 .

In this work, taking into account the high abundance of Ni, Fe, Co, and Cu and they can easily form solid solutions with Pt. These transition metal elements are selected because their similar atomic radius and lower heat of formation make them likely to form stable HEAsterm·dom 20 , 28 . In addition, the DFTterm·key calculation provides a means to better understand the catalytic process and direct catalytic design 13 . Based on this, we synthesize the small size (~3.4 nm) and uniform HEAterm·dom Pt 18 Ni 26 Fe 15 Co 14 Cu 27 nanoparticlesterm·key (NPs) by a simple low-temperature oil phase synthesis method at atmospheric pressure. The Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst displays a low overpotentialterm·key (11 mV at 10 mA cm −2 ), the activity of 10.96 A mg −1 Pt at −0.07 V vs. reversible hydrogen electrode (RHE) for HERterm·dom (reduction reaction) in 1 M KOH solution, and through the chronoamperometric method and 10,000th CV tests, it indicates that the notable stability of the HEAterm·dom catalyst. We further also measure the electrocatalytic activity of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C for MORterm·dom (oxidation reaction) at 1 M KOH + 1 M CH 3 OH electrolyte, it also exhibited high mass activity (15.04 A mg −1 Pt ). Theoretical calculations reveal that each element in HEAterm·dom displays different contributions for the electrocatalysisterm·key process, which promotes the site-to-site electron transferterm·key and the stabilization of the intermediates. HEAterm·dom showed the multi-active sites for both HERterm·dom and MORterm·dom for achieving superior performance and stability.

Results Synthesis and characterizations of PtNiFeCoCu HEAterm·dom NPs PtNiFeCoCu HEAterm·dom NPs were prepared through a simple one-pot oil phase synthesis method at 220 °C for 2 h. The inductively coupled plasma atomic emission spectroscopy (ICP-AES) result shows that the atomic ratio of Pt, Ni, Fe, Co, and Cu was 18:26:15:14:27 (Supplementary Table 1 ), the prepared NPs-HEA was named Pt 18 Ni 26 Fe 15 Co 14 Cu 27 . As shown in Fig. 1a, b , the transmission electron microscopy (TEM) images of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 NPs display the morphology is uniform, and the diameter of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 NPs is about 3.4 ± 0.6 nm (the inset in Fig. 1a ). The high-resolution TEM (HRTEM) (Fig. 1c ) image of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 NPs exhibits that the lattice spacing is 0.218 nm, corresponding to the (111) facet. The powder X-ray diffraction (PXRD) pattern (Supplementary Fig. 1a ) suggests the fcc structure of the NPs-HEAs, the two peaks at around 41.1 o and 47.8 o , which can be assigned to the (111) and (200) facets. The position of the broad diffraction peak is significant shift compared with the diffraction peaks of pure Pt, Ni, Fe, Co, and Cu (Supplementary Fig. 1b ), it shows that these elements have been introduced into the nanostructure to form HEAsterm·dom structure. Figure 1d shows the elemental maps of NPs-HEAs, Pt, Ni, Fe, Co, and Cu elements are uniformly distributed in the HEAterm·dom nanostructure. Fig. 1 TEM images and elemental mapping analysis. a , b TEM images of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 nanoparticlesterm·key. c HRTEM image of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 nanoparticlesterm·key. d The corresponding elemental mapping of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 nanoparticlesterm·key (scale bar, 5 nm).

The X-ray photoelectron spectroscopy (XPS) of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 NPs was tested, it also shows the presence of these Pt, Ni, Fe, Co, and Cu elements (Supplementary Fig. 2a ), the atomic ratio of Pt, Ni, Fe, Co, and Cu was 21.2:27.4:15.2:13.7:22.5, further proving that the HEAterm·dom was successfully prepared. As shown in Supplementary Fig. 2b , the Pt 4f 7/2 and Pt 4f 5/2 peaks are located at 71.3 and 74.7 eV, respectively. Pt mainly exists in the form of zero valences and a small amount of 2+ valence 29 . The Ni 2p spectra display the coexistence of Ni 0 (852.8 eV) and Ni 2+ (855.6 eV) and a satellite peak locate at 861.6 eV (Supplementary Fig. 2c ), the content of Ni 0 is lower than the content of Ni 2+ because of the high chemical activity of Ni 28 , 29 . The Fe 2p spectra in Supplementary Fig. 2d shows two peaks at 711.6 eV and 724.8 eV, which can be attributed to Fe 2p 3/2 and Fe 2p 1/2 , respectively. The two Co 2p peaks can be assigned to Co 2p 3/2 (780.8 eV) and Co 2p 1/2 (797.0 eV) (Supplementary Fig. 2e ). The Cu 2p spectra (Supplementary Fig. 2f ) also exhibits Cu 0 (931.7 eV) and Cu 2+ (933.5 eV) peaks coexist in the sample.

Electrocatalytic performance tests toward HERterm·dom In order to study the electrocatalytic performance of PtNiFeCoCu/C, we further treated with acetic acid to remove residual surfactants. From the FTIR spectrum (Supplementary Fig. 3 ), the acetic acid-treated sample (the purified PtNiFeCoCu/C), no obvious characteristic peaks corresponding to oleylamine and CTAC were observed 36 . The HERterm·dom (reduction reaction) activity and stability of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst were investigated by a series of electrochemical tests and compared with commercial Pt/Cterm·key. The Pt size (about 3.0 nm, Supplementary Fig. 4 ) on the Pt/Cterm·key catalyst is similar to the size of HEAterm·dom particles, so the electrocatalytic activity can be reasonably compared based on surface Pt sites. Figure 2a shows the CV curves of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C and commercial Pt/Cterm·key catalysts in N 2 -saturated 1 M KOH at a scan rate of 20 mV s −1 . After activation, the linear sweep voltammetry (LSV) curve of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst displays a low overpotentialterm·key of 11 mV at the current density of 10 mA cm −2 (normalized to the electrode area), which is far superior to commercial Pt/Cterm·key catalyst (84 mV) (Fig. 2b ). As shown in Fig. 2b, d , the area activity (normalized to the geometric area) of NPs-HEAs Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst reaches 83.78 mA cm −2 at −0.07 V vs. RHE, far more than area activity of commercial Pt/Cterm·key catalyst (8.42 mA cm −2 at −0.07 V vs. RHE). From the Fig. 2c (LSV curves, normalized to the Pt mass), the mass activity for commercial Pt/Cterm·key catalyst and Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst are 0.83 A mg −1 Pt and 10.96 A mg −1 Pt at −0.07 V vs. RHE (Fig. 2d ), respectively. And the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst also exhibits higher HERterm·dom performance in alkaline medium among the reported Pt-based catalysts and HEAsterm·dom catalysts (Supplementary Table 7 ). As shown in Fig. 2e , the Tafel slopes of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C and Pt/Cterm·key are 30 mV dec −1 and 98 mV dec −1 , respectively, which demonstrate that the HEAsterm·dom catalyst greatly boosts HERterm·dom kinetics. Fig. 2 Electrocatalytic performance of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C and Pt/Cterm·key for HERterm·dom in 1 M KOH electrolyte. a CV curves. b HERterm·dom polarization curves (geometrical area). c Pt mass loading normalized (mass activity) LSV curves. d Comparison of area activity and mass activity values for HERterm·dom at −70 mV vs. RHE. e Tafel slope. f HERterm·dom polarization curves (geometrical area) for Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C with different CV cycle.

To better understand the HERterm·dom performance of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst, we tested the electrochemical double-layer capacitance, electrochemical impedance spectra, and turnover frequencyterm·mth (TOFterm·mth). We found that the electrochemical double-layer capacitance of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst is higher than that of commercial Pt/Cterm·key catalyst (Supplementary Fig. 5 ), indicating that the NPs-HEAs can expose more active sites. Also, electrochemical impedance spectra (Supplementary Fig. 6 ) exhibits Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst has a smaller semicircular diameter than commercial Pt/Cterm·key catalyst, the transfer resistance of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C is much lower than that of commercial Pt/Cterm·key catalysts, indicating higher interfacial charge transfer rate and faster HERterm·dom kinetics. The TOFterm·mth value is used to characterize the activity of each site in the catalyst 37 . It is found that the TOFterm·mth value of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst is higher than Pt/Cterm·key catalyst under various potentials (Supplementary Fig. 7 ), which shows the faster HERterm·dom kinetics of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst. These results further indicate that the NPs-HEAs can improve the catalytic activity effectively.

The stability of the catalysts was checked using the chronoamperometric method and 10,000th CV in 1.0 M KOH solution. The stability of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C and Pt/Cterm·key catalysts were measured at an overpotentialterm·key of 11 mV and 84 mV (under current density 10 mA cm −2 condition, Supplementary Fig. 8 ), respectively. After 10 h test, 99% of the current density is maintained for the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst, while the current density of the Pt/Cterm·key catalyst is only kept about 53%. To further evaluate the stability of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C and Pt/Cterm·key catalysts, the 1000th and 10,000th CV cycles in 1 M KOH solution were performed. Figure 2f displays the LSV curves of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst before and after 1000th and 10,000th CV cycles. There is no obvious negative shift at the current density of 10 mA cm −2 , exhibiting higher stability than Pt/Cterm·key catalyst (Supplementary Fig. 9 , a negative shift of ∼33 mV at 10 mA cm −2 after 10,000th CV cycles). In addition, the morphology (Supplementary Fig. 10 ), the atomic ratio (Supplementary Table 1 ), lattice spacing and the XRD peaks (Supplementary Fig. 11 ) of catalyst do not change significantly after the stability test. In addition, the slight variations in the stoichiometry (Supplementary Table 2 ) after the electrocatalytic test was caused by the leaching of non-noble metals (Ni, Fe, Co, and Cu) in the catalyst 38 , 39 . XPS analysis (Supplementary Fig. 12 ) shows that the Pt 4 f, Ni 2p, Fe 2p, Co 2p, and Cu 2p spectra of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C were very similar before and after electrocatalysisterm·key, while the oxide species increased slightly (Supplementary Table 3 ). However, the Pt/Cterm·key catalyst shows obvious aggregation after the stability test (Supplementary Fig. 13 ). These results further suggest the excellent stability of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst.

Furthermore, taking the change of Cu element content as an example, the effect of stoichiometry variations on the electrocatalytic performance is discussed. Specifically, this synthesis method is used to prepare PtNiFeCoCu NPs with other Cu content (PtNiFeCoCu 26 with slight stoichiometry variations, Pt 21 Ni 27 Fe 19 Co 17 Cu 16 and Pt 15 Ni 24 Fe 13 Co 13 Cu 35 ) by changing the amount of Cu precursor. From Supplementary Fig. 14 , the morphology and size of the PtNiFeCoCu 26 , Pt 21 Ni 27 Fe 19 Co 17 Cu 16 and Pt 15 Ni 24 Fe 13 Co 13 Cu 35 are similar to Pt 18 Ni 26 Fe 15 Co 14 Cu 27 NPs. The PXRD pattern (Supplementary Fig. 15 ) suggests the fcc structure of the PtNiFeCoCu 26 , Pt 21 Ni 27 Fe 19 Co 17 Cu 16 and Pt 15 Ni 24 Fe 13 Co 13 Cu 35 NPs. The HERterm·dom performance of PtNiFeCoCu 26 /C catalyst with slight stoichiometry variations (Supplementary Figs. 16 – 24 and Supplementary Table 5 ) has no significant change compared with Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C (PtNiFeCoCu 27 /C). After activation (Supplementary Fig. 16 ), the PtNiFeCoCu 26 /C catalyst has close overpotentialterm·key (11 mV, Supplementary Fig. 17a ), area activity and mass activity (82.66 mA cm −2 and 10.81 A mg −1 Pt at −0.07 V vs. RHE, Supplementary Fig. 17a–c ), Tafel slope (30 mV dec −1 , Supplementary Fig. 18a ), electrochemical double-layer capacitance (7.01 mF, Supplementary Fig. 19a, d ), electrochemical impedance (Supplementary Fig. 20a ), TOFterm·mth value (Supplementary Fig. 21a, c ) and stability (Supplementary Figs. 22 – 24 ) compared with PtNiFeCoCu 27 /C catalyst. In addition, the HERterm·dom performance of Pt 21 Ni 27 Fe 19 Co 17 Cu 16 /C and Pt 15 Ni 24 Fe 13 Co 13 Cu 35 /C catalysts (Supplementary Figs. 16 – 24 and Supplementary Table 5 ) is slightly lower than the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst. The order of activity is as follows: Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C (83.78 mA cm −2 and 10.96 A mg −1 Pt at −0.07 V vs. RHE) ≈ PtNiFeCoCu 26 /C (82.66 mA cm −2 and 10.81 A mg −1 Pt at −0.07 V vs. RHE) > Pt 21 Ni 27 Fe 19 Co 17 Cu 16 /C (74.21 mA cm −2 and 9.65 A mg −1 Pt at −0.07 V vs. RHE) > Pt 15 Ni 24 Fe 13 Co 13 Cu 35 /C (59.76 mA cm −2 and 7.88 A mg −1 Pt at −0.07 V vs. RHE).

Electrocatalytic performance tests toward MORterm·dom To further explore the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst as redox bi-function electrocatalysts, we tested the MORterm·dom (oxidation reaction) activity. As shown in Fig. 3a and Supplementary Fig. 25 , the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst shows higher activity compared to Pt/Cterm·key catalyst for MORterm·dom in 1 M KOH + 1 M CH 3 OH electrolyte at a sweep rate of 20 mV s −1 . And an onset potential (the mass activity of 0.1 A mg −1 pt ) was observed to decrease by 133 mV in the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst compared to the Pt/Cterm·key catalyst (the inset of Fig. 3a ), indicating that the activation barrier of methanol oxidation is lower. The Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst achieves 10 times (4 times) higher in mass activity and area activity (15.04 A mg −1 pt , 114.93 mA cm −2 ) than that of Pt/Cterm·key catalyst (1.45 A mg −1 pt , 27.48 mA cm −2 ) at peak potential for MORterm·dom (Fig. 3b ). In addition, the MORterm·dom performance of PtNiFeCoCu 26 /C catalyst is similar to that of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst, the MORterm·dom performance of Pt 21 Ni 27 Fe 19 Co 17 Cu 16 /C and Pt 15 Ni 24 Fe 13 Co 13 Cu 35 /C catalysts is slightly lower than the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst. (Supplementary Fig. 26 and Supplementary Table 6 ). Moreover, among recently reported the Pt-based materials for MORterm·dom in alkaline medium, the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst exhibits higher mass activity (Supplementary Table 8 ). Fig. 3 Methanol electro-oxidation performance of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C and Pt/Cterm·key in 1 M KOH + 1 M CH 3 OH electrolyte. a CV curves (the inset is the onset potential, the mass activity of 0.1 A mg −1 pt ) of Pt/Cterm·key and Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C. b Peak values of mass activity and area activity. c Chronoamperometric tests for MORterm·dom at 0.65 V vs. RHE. d CV curves of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C before and after 1000 cycles.

To study the stability of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C and Pt/Cterm·key catalysts in MORterm·dom, the chronoamperometry test at 0.65 V vs. RHE and 1000 CV cycles were executed. After 5000 s chronoamperometry test (Fig. 3c ), the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C (3.79 A mg −1 pt ) catalyst shows higher stability than that of commercial Pt/Cterm·key (0.20 A mg −1 pt ). After 1000 CV cycles, the mass activity of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C and Pt/Cterm·key catalysts decay by about 6.4% and 26.9% (Fig. 3d and Supplementary Fig. 28a ), further confirmed the excellent stability of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst. From Supplementary Figs. 27 , 28 , the order of catalysts stability is as follows: Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C ≈ PtNiFeCoCu 26 /C > Pt 21 Ni 27 Fe 19 Co 17 Cu 16 /C > Pt 15 Ni 24 Fe 13 Co 13 Cu 35 /C. And the morphology and element ratio of PtNiFeCoCu/C catalyst with different rations did not obvious change after stability test (Supplementary Figs. 29 , 30 , Supplementary Table 1 ). However, the Pt/Cterm·key catalyst shows obvious aggregation after the stability test (Supplementary Fig. 30d ). XPS spectrum (Supplementary Fig. 31 ) shows that the Pt 4 f, Ni 2p, Fe 2p, Co 2p, and Cu 2p spectra of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C were very similar before and after MORterm·dom electrocatalysisterm·key. These test results indicate that the NPs-HEAs catalyst achieves double enhancement of catalytic activity and stability.

In MORterm·dom, the main route for catalyst deactivation is the poisoning effect of CO intermediates 40 – 42 . The I f / I b ( I f : forward current density, I b : backward current density) ratio of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst (3.26) is larger than that of Pt/Cterm·key (2.31), which shows the strong CO anti-poisoning performance of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst. And from CO stripping curves (Supplementary Fig. 32 ), the onset potentials of the Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C (0.391 V vs. RHE) display a 181 mV decrease compared with Pt/Cterm·key (0.572 V vs. RHE) catalyst. PtNiFeCoCu 26 /C (0.398 V vs. RHE), Pt 21 Ni 27 Fe 19 Co 17 Cu 16 /C (0.395 V vs. RHE), and Pt 15 Ni 24 Fe 13 Co 13 Cu 35 /C (0.413 V vs. RHE) catalysts have analogous onset potentials to Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C. It further shows that the PtNiFeCoCu/C catalyst has better CO anti-poisoning performance.

DFTterm·key studies We applied periodic DFTterm·key calculations to explore the HERterm·dom and MORterm·dom performances in HEAterm·dom. It is well known that the surface composition of the catalyst affects the catalytic performance. We use Ar + sputtering to study the change in surface element content of the PtNiFeCoCu NPs 43 . After Ar + sputtering for 10 s, the relative content of the elements has been reduced, but the reduced contents of Ni, Cu, and Fe were slightly higher than that of Pt and Co (Supplementary Table 4 ). We have also compared the PDOS of the HEAterm·dom with slightly different stoichiometry. As shown in Supplementary Fig. 33 , the comparison models show the highly similar electronic structure with the models applied in the work, in which the peak positions and patterns of d orbitals in each element display very limited change. Therefore, the slight stoichiometry variations of the HEAterm·dom will not significantly affect the electronic structure of the PDOS results. Based on this, after a detailed comparison of different random atomic arrangements, the HEAterm·dom structure with slightly Ni and Cu enriched surface has been applied as the lattice model due to the highest stability (Fig. 4a ). From the side view, the lattice has shown a highly stable structure, in which the lattice shows subtle distortion after relaxation, which indicates good durability for electrocatalysisterm·key. Meanwhile, the surface Ni and Co dominate the electroactive region near the Fermi level ( E F ) (Fig. 4b ). To further understand the electronic structures, the partial projected density of states (PDOSs) of each element in HEAterm·dom has been illustrated. Notably, Pt-5d occupies the deepest position near E V −4.5 eV (EV = 0 eV), playing as the electron reservoir for the reduction process such as HERterm·dom. Both Co-3d and Ni-3d orbitals dominate the bands near E F , which locate at E V −1.0 eV, contributing to the electron depletion center for HERterm·dom and MORterm·dom. Moreover, the 3d orbitals of Cu, Co, and Fe not only alleviate the energy barrierterm·dom of dual-way electron transferterm·key for the oxidation and reduction process but also facilitate the stabilization of intermediates for MORterm·dom (Fig. 4c ). A detailed study of the site-dependent PDOSs of each element was illustrated (Fig. 4d ). Notably, only the surface Pt demonstrates an evident upshift towards the E F , which promotes the electron transferterm·key for the HEAterm·dom surfaces. Fe shows a site-independent electronic structure within HEAterm·dom, which preserves the stable adsorption of intermediates with stronger anti-poisoning capability for the MORterm·dom process. From the bulk structure to the surface site, Co sites display an alleviation of the e g -t 2g splitting effect, supporting an enhanced electron transferterm·key efficiency for electrocatalysisterm·key. Both Ni exhibits the relatively stable d-band-center to maintain the electroactive electron boosting center. For Cu sites closer to the surface, the 3d orbitals show a slight upshift trend to support the electroactivity of the HEAterm·dom. Thus, the synergistic effect of multi-active sites on the HEAterm·dom surface determines the remarkable performance of HERterm·dom and MORterm·dom (Fig. 4d ). For the HERterm·dom process, the initial adsorption of water determines the efficiency of water-splitting and the following proton transfer. For the water adsorption process, the evident downshift of s,p orbitals in H 2 O has been noticed, which confirms the active electron transferterm·key from the HEAterm·dom to the water to achieve the stable adsorption and lays a good foundation for the following water dissociation (Fig. 4e ). Similarly, we notice the adsorption of CH 3 OH on HEAterm·dom indicates the evident downshifting of s,p bands, and overlapping with electroactive d orbitals of the surface (Fig. 4f ). For the multi-electron involved MORterm·dom process, the linear correlation of intermediates transformation is the key to guarantee the proton and electron transferterm·key. Such a linear correlation is noticed for the s,p orbitals of key intermediates along the MORterm·dom process, which not only supports the efficient oxidation of the intermediates but also leads to the optimal binding strength during the intermediate transformation. Thus, superior MORterm·dom performance is guaranteed in the HEAterm·dom (Fig. 4g ). Fig. 4 Density functional theory calculations for the structural configuration and PDOSs. a The side view of the structural configuration of HEAterm·dom. b The side view of HEAterm·dom structural configuration and the real spatial contour plots for bonding and anti-bonding orbitals near E F . And the top view of the real spatial contour plots for bonding and anti-bonding orbitals near E F for the HEAterm·dom. Dark green balls = Pt; Gray balls= Fe; Blue balls = Co; Pink balls = Ni and Orange balls = Cu. c The PDOSs of the HEAterm·dom. d The site-dependent PDOSs of Pt, Fe, Co, Ni, and Cu in HEAterm·dom. e The PDOSs for the water adsorption. f The PDOSs for the CH 3 OH adsorption. g The PDOSs for the key intermediates of MORterm·dom.

Then, we further interpret the reaction trend for both HERterm·dom and MORterm·dom from both the structural configuration and energetic reaction pathways. The most stable structural configurations of four key initial reactants and intermediates have been displayed. The most stable adsorption of CH 3 OH and H 2 O locates near Ni and Fe sites, respectively. The OH is stabilized in the neighboring hollow sites, which avoids the active site blocking during both HERterm·dom and MORterm·dom. Meanwhile, the H adsorption prefers the hollow site near Ni and Co, which is distinct from the OH. Thus, the multi-site adsorption for HERterm·dom and MORterm·dom process on HEAterm·dom guarantees superior performances (Fig. 5a ). Moreover, we have supplied the adsorption sites mapping of the surface to support the HERterm·dom mechanism (Fig. 5b ). Notably, the different sites in HEAterm·dom surfaces demonstrated very varied adsorption preference for the intermediates. For HERterm·dom, the initial adsorption of H 2 O locates on the Fe sites, which activates the dissociation of water molecules and facilitates the stabilization of *OH in the neighboring hollow sites. Meanwhile, the nearby Ni, Co show the relatively preferred H* adsorption after the water dissociation, leading to the stabilization of H in the hollow sites surrounded by Co and Ni. The generated H 2 shows overall weak binding to the surface, indicating the quick desorption process to guarantee the efficient HERterm·dom process. Therefore, our additional information has supplied the binding stability of the intermediates for the HERterm·dom process, which supports the Volmer–Heyrovsky mechanism. The HERterm·dom process supports a continuous downhill trend, confirming the efficient proton and electron transferterm·key (Fig. 5c ). Owing to the multi-active sites for OH and H, the water dissociation demonstrates a low activation barrier of 0.11 eV for the transition state (Fig. 5d ). For the MORterm·dom process, the rate-determining step occurs at [CHO* + 3*OH + 3H 2 O] to [HCOOH + 2*OH + 4H 2 O] with the largest energy barrierterm·dom of 0.45 eV. The transition state displays the activation energy of 0.64 eV. The overall MORterm·dom process is exothermic, which releases 2.34 eV energy (Fig. 5e ). The anti-poisoning capability is another essential requirement for the long-term application of MORterm·dom electrocatalysts. Compared to MORterm·dom, the formation of CO shows a much larger energy barrierterm·dom of 0.81 eV and activation energy (0.94 eV), resulting in the suppression of the CO poisoningterm·dom. The holistic reaction trend of the CO poisoningterm·dom mechanism is also much weaker than the MORterm·dom process, which explains both the superior electroactivity and durability of the HEAterm·dom (Fig. 5f ). Fig. 5 Density functional theory calculations for the structural configuration and energetic reaction pathways. a The structural configuration of stable adsorption of key intermediates in HERterm·dom and MORterm·dom. From left to right: CH 3 OH, H 2 O, CO 2 , H. Dark green balls = Pt; Gray balls = Fe; Blue balls = Co; Pink balls = Ni; Orange balls = Cu; Red balls = O and White balls = H. b The binding energy mapping of HERterm·dom. c The energetic pathway of the alkaline HERterm·dom. d The activation energies of water dissociation. e The energetic pathway of the alkaline MORterm·dom. f The energetic pathway of CO poisoningterm·dom.

Discussion In summary, we have synthesized the uniform and small size Pt 18 Ni 26 Fe 15 Co 14 Cu 27 NPs HEAterm·dom nanoparticlesterm·key by a simple low-temperature synthesis method at atmospheric pressure. Electrocatalytic test results showed that the obtained Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst has excellent bi-functional electrocatalytic properties for reduction reaction (HERterm·dom) and oxidation reaction (MORterm·dom). Pt 18 Ni 26 Fe 15 Co 14 Cu 27 /C catalyst showed an ultrasmall overpotentialterm·key (11 mV at 10 mA cm −2 ) and superior activity (10.96 A mg −1 Pt at −0.07 V vs. RHE) and stability for HERterm·dom, which is one of the best HERterm·dom activity in alkaline medium. And it is also the effective catalysts for MORterm·dom and displayed the excellent activity (15.04 A mg −1 Pt ) and better CO anti-poisoning in alkaline solution, which is the best alkaline MORterm·dom activity among the ever reported. Through DFTterm·key calculations, the origin of remarkable electroactivity and durability of HEAterm·dom in HERterm·dom and MORterm·dom is attributed to the synergistic effect of each element for efficient electron transferterm·key. The suitable electronic environment of the HEAterm·dom realizes the multi-active sites for appropriate adsorption of key intermediates and efficient electron transferterm·key during the electrocatalysisterm·key, which maximizes the utilization of surface electroactivity. The simple oil phase synthesis strategy proposed in this work, as well as the multi-active sites and fast site-to-site electron transferterm·key mechanism, which is expected to lay the foundation for the preparation of other HEAsterm·dom and their application in related electrocatalysisterm·key.

Methods Materials Platinum (II) acetylacetonate (Pt(acac) 2 , 97%), nickel (II) acetylacetonate (Ni(acac) 2 , 95%), Molybdenumhexacarbonyl (Mo(CO) 6 , 98%), oleylamine (OAm, >70%) and glucose were bought from Sigma-Aldrich. Cupric acetylacetonate (Cu(acac) 2 , 98%) was purchased from Energy Chemical. (1-Hexadecyl) trimethylammonium chloride (CTAC, 96%), Tris(2,4-pentanediaonato) Cobalt (III) (Co(acac) 3 , 98%), Iron (III) 2,4-pentanedionate (Fe(acac) 3 ), Nafion solution (5 wt.%) were supplied by Alfa Aesar. Methanol, cyclohexane, ethanol, and isopropanol were bought from Beijing Tongguang Fine Chemicals Company. Potassium hydroxide (KOH, 90%) was purchased from Aladdin.

Preparation of HEAterm·dom Pt 18 Ni 26 Fe 15 Co 14 Cu 27 NPs CTAC (50 mg) was added into oleylamine (5 mL) in a 15 mL vial. After sonication for about 15 min, Pt(acac) 2 (10 mg), Ni(acac) 2 (6.4 mg), Fe(acac) 3 (8.8 mg), Co(acac) 3 (8.9 mg), Cu(acac) 2 (6.5 mg), glucose (60 mg), and Mo(CO) 6 (33 mg) were added into the vial. In order to obtain a homogeneous solution, the mixture was sonicated for 1 h. The vial was heated to 220 ° C at 5 ° C min −1 and then kept 2 h under magnetic stirring at 400 rpm. The black colloidal products were collected by centrifugation and washed two times with an ethanol/cyclohexane mixture. Finally, the black colloidal products were kept in cyclohexane for further use.

Preparation of alloy PtNiFeCoCu with different element ratios NPs CTAC (50 mg) was added into oleylamine (5 mL) in a 15 mL vial. After sonication for about 15 min, Pt(acac) 2 (10 mg), Ni(acac) 2 (6.4 mg), Fe(acac) 3 (8.8 mg), Co(acac) 3 (8.9 mg), Cu(acac) 2 (6.2 mg), glucose (60 mg) and Mo(CO) 6 (33 mg) were added into the vial. In order to obtain a homogeneous solution, the mixture was sonicated for 1 h. The vial was heated to 220 o C at 5 o C min −1 and then kept 2 h under magnetic stirring at 400 rpm. The black colloidal products (PtNiFeCoCu 26 ) were collected by centrifugation and washed two times with an ethanol/cyclohexane mixture. Finally, the black colloidal products were kept in cyclohexane for further use. The synthesis of NPs with other element ratios only changes the amount of Cu(acac) 2 precursor (4.0 mg and 8.5 mg).

Preparation of PtNiFeCoCu/C The obtained 1 mg HEAterm·dom NPs dispersed in 10 mL cyclohexane was mixed with 4 mg of carbon (Ketjen Black-300) in 10 mL ethanol under sonication for 1 h, and then the product was collected via centrifugation with ethanol. The PtNiFeCoCu/C catalysts were further cleaned (remove organic species) with 0.5 M acetic acid (ethanol solution) under N 2 atmosphere. After being sonicating for 2 h, the products were collected by centrifugation and washed with ethanol for three times.

Characterization The TEM and high-resolution TEM (HRTEM) images of the samples were characterized by an FEI Tecnai-G2 F30 at an accelerating voltage of 300 KV. Powder X-ray diffraction (XRD) spectra were recorded on X’Pert-PRO MPD diffractometer operating at 40 kV and 40 mA with Cu Kα radiation. The compositions of the HEAterm·dom NPs were determined by the inductively coupled plasma atomic emission spectrometer (ICP-AES, Varian 710-ES). The catalysts after the durability tests were scratched off the glassy carbon electrode with the aid of sonication in ethanol and then collected for further TEM, XRD, and ICP characterization. The XPS analyses were carried out with Axis Supra spectrometer using a monochromatic Al Kɑ source (15 mA, 14 kV). Survey scan analyses were carried out with an analysis area of 300 × 700 μm 2 and a pass energy of 100 eV. High-resolution analyses were carried out with an analysis area of 300 × 700 μm 2 and a pass energy of 30 eV. Spectra have been corrected to the main line of the carbon 1s spectrum (adventitious carbon) set to 284.8 eV. Spectra were analyzed using CasaXPS software (version 2.3.14).

Electrochemical measurements The 1 mg PtNiFeCoCu/C catalysts were dispersed in a mixture of 495 µL ultrapure water, 495 µL isopropanol, and 10 µL Nafion solution, after sonication for 1 h, PtNiFeCoCu/C catalyst with the concentration of 1 mg mL −1 was obtained.

Electrochemical measurements were conducted on a CHI 760E Electrochemical Workstation (Shanghai Chenhua Instrument Corporation, China) in a conventional three-electrode cell. The graphite rod electrode as the counter electrode and a saturated calomel electrode (SCE) as the reference electrode. The working electrode was a glassy carbon electrode (GCE, diameter: 3 mm, area: 0.07065 cm 2 ). Ten microliters of the catalyst were dropped onto the GCE surface for further electrochemical tests. All the potentials reported in this work were converted to the reversible hydrogen electrode (RHE). Cyclic voltammograms (CVs) were performed in N 2 -saturated 1 M KOH solution from 0 to 1.2 V vs. RHE at a scan rate of 20 mV s −1 . Electrochemical impedance spectroscopy (EIS) measurements were measured at −50 mV vs. RHE in the frequency range from 10 kHz to 0.01 Hz in N 2 -saturated 1 M KOH solution.

Hydrogen evolution reactionterm·dom (HERterm·dom) measurements The HERterm·dom performance of the catalysts was evaluated by linear sweep voltammetry (LSV) with a scan rate of 5 mV s −1 in N 2 -saturated 1 M KOH solution, and all polarization curves were 95% iR-corrected. The durability tests were performed in 1.0 M KOH solution using the chronoamperometric method. And 1000th/10000th CVs were also measured to evaluate the stability of catalysts. 20% Pt/Cterm·key catalyst was also dropped on the GCE as a reference catalyst for electrochemical tests under the same conditions.

TOFterm·mth calculation The CVs curve was collected from 0 to 1.2 V vs. RHE in 0.5 M H 2 SO 4 solution with a scan rate of 20 mV s −1 : 1 n = Q H 2 * F , 2 TOFterm·mth S − 1 = I 2 * F * n , where n is the number of active sites, number 2 represents two electrons (produce one hydrogen molecule), Q H represents electron transferterm·key quantity (the hydrogen desorption peak), F is Faraday’s constant (96485.3 C mol −1 ), and I (A) is the current measured at a specific potential during LSV measurement.

Methanol oxidation reactionterm·dom (MORterm·dom) measurements The CVs for MORterm·dom were conducted in N 2 -saturated 1 M KOH + 1 M CH 3 OH solution between 0.2–1.2 V vs. RHE with a scan rate of 20 mV s −1 . For the MORterm·dom stability tests, chronoamperometric tests were performed at a fixed potential of 0.65 V vs. RHE, and 1000th CVs were also performed to evaluate the stability of catalysts.

CO stripping curves were carried out in 1 M KOH solution Before the tests, 1 M KOH solution was first deaerated with high-purity N 2 . Then, CO was bubbled into the cell for 15 min while the potential of the working electrode was held at a constant potential of 0.1 V vs. RHE. Then N 2 was bubbled into the system for 15 min to remove CO gas. After that, CO stripping curves were recorded between 0 and 1.2 V vs. RHE at a scan rate of 20 mV s −1 .

Calculation setup For all the calculations within this work, we have applied the DFTterm·key calculations within CASTEP code 44 . The GGA and PBE exchange-correlation functionals are selected for all the calculations 45 , 46 . The cutoff energy of plane-wave basis sets based on the ultrasoft pseudopotential has been set to 440 eV with the selection of the algorithm Broyden–Fletcher–Goldfarb-Shannon (BFGS) for all the geometry optimizations 47 . To specifically discuss the electrocatalysisterm·key on the surface, we have applied the HEAterm·dom model with similar stoichiometry as experimental characterizations. The HEAterm·dom model has been built, which consists of 126 atoms in total. The atomic arrangements of different elements are constructed randomly by following the same ratio as the experimentsterm·mth of Pt 18 Ni 26 Fe 15 Co 14 Cu 27 . Based on the components of the HEAterm·dom by experimental characterizations, Ni and Cu have the highest concentration, which is 26% and 27%, respectively. To determine the most possible preferred model, we have compared the total energy of the HEAterm·dom model with different surface arrangements, in which the present applied model with Cu and Ni slight rich feature has been the most stable one. Therefore, we have applied the HEAterm·dom surface model with Cu and Ni rich feature. The reaction energy has been considered based on the intermediate adsorptions on the surface 34 . The Monkhost–Pack reciprocal space integration was performed using coarse k-points with a mesh of 2 × 2 × 1 48 , which was guided by the initial convergence test. With these settings, the overall total energy for each step is converged to less than 5.0 × 10 −5 eV per atom. The Hellmann-Feynman forces on the atom were converged to less than 0.001 eV/Å.

Supplementary information Supporting Information Peer Review File