H2 In Situ Inducing Strategy on Pt Surface Segregation Over Low Pt Doped PtNi5 Nanoalloy with Superhigh Alkaline HER Activity

Surface segregation constitutes an efficient approach to enhance the alkaline hydrogen evolution reaction (HER) activity of bimetallic PtxNiy nanoalloys. Herein, a new strategy is proposed by utilizing the small gas molecule of H2 as the structure directing agent (SDA) to in situ induce Pt surface segregations over a series of PtNi5‐n samples with extremely low Pt doping (Pt/Ni = 0.2). Impressively, the sample of PtNi5‐0.3 synthesized under 0.3 MPa H2 delivers an extremely low overpotential of 26.8 mV (−10 mA cm−2) and Tafel slope of 19.2 mV dec−1, which is superior to most of the previously reported PtxNiy electrocatalysts. This is substantially related to the strong H2 in situ inducing effect to generate Pt‐rich@Ni‐rich core‐shell nanostructure of PtNi5‐0.3 with an ultrahigh Pt surface content of 46%. The specific mechanistic effects of H2 during the PtNi5‐n synthesis process are well illustrated based on the combined experimental and theoretical studies. The density functional theory mechanism simulations further unravel that the evolved active site of PtNi5‐n can efficiently reduce the reaction Gibbs free energies; especially for the scenario of PtNi5‐0.3, the downward‐shifted d band center of the Pt active site significantly reduces the PtH bond strength, eventually resulting in the lowest absolute value of ΔGH.

the negligible greenhouse gas emissions of the reaction. [1][2][3] However, the development of energy production system based on hydrogen evolution reaction (HER) is currently limited by either high cost or low efficiency of the electrode catalyst. [4] For example, although the platinum (Pt) is widely accepted as the best HER catalyst in acid medium, the high cost greatly hinders its wide applications; additionally, as for the alkaline hydrogen evolution reaction (alkaline-HER), the Pt however shows low reaction efficiency due to the sluggish activity in the Volmer step (H 2 O + * + e -→ H* + OH -). [5,6] To overcome these problems, considerable efforts have been devoted to making the Pt electrocatalysts less costly but more efficient, especially for the scenario of the alkaline-HER due to its relatively mild reaction condition which allows the non-Pt group metals being incorporated into the Pt electrocatalyst to lower its cost and simultaneously enhance its alkaline-HER activity. [7][8][9] Typically, the surface segregation has been reported to be an efficient approach to synthesize highly active Pt-based bimetallic nanoalloy electrocatalysts for the alkaline-HER. [10][11][12] Large amounts of work have been recently focused on the synthesis of the surface segregated Pt x Ni y bimetallic nanoalloys, due to the excellent water splitting activity of the dopant Ni in Volmer step. [13][14][15][16] For example, Cao et al. [14] fabricated one type of PtNi nanoalloy with a Pt-rich@Ni-rich core-shell nanostructure, which achieved much higher activity than that of the commercial Pt/C by displaying an overpotential of 104 mV (10 mA cm −2 ) and Tafel slope of 73 mV dec −1 . Zhang et al. [15] reported one type lotus-thalamus-shaped PtNi nanoalloy with a Pt-rich surface, which exhibited an even lower overpotential of 27.7 mV (10 mA cm −2 ) and Tafel slope of 27 mV dec −1 .

Introduction
Electrochemical water splitting to produce hydrogen has drawn great attention to reduce the reliance on fossil fuels for human beings, owing to the massive resources of the water as well as Pt-rich surface can be evolved after the CO post-treatment due to the stronger binding energy of CO to Pt. Similarly, taking advantage of the strong interaction of H 2 with Pt, the Pt segregated PtM (M = Au, [21] Ni, [22] Co [23] ) bimetallic nanoalloys were successfully synthesized after the post treatment by H 2 .
Inspired by the above literature reports as well as the basic design principle of enhancing the Pt utilization efficiency, herein we propose a simple one-pot synthesis strategy by simultaneously introducing the H 2 into the synthesis system of PtNi 5 , which possesses an extremely low Pt/Ni mole ratio of 0.2, to in situ induce Pt surface segregation for the alkaline HER. Following this idea, a range of PtNi 5 -n (n = 0, 0.1, 0.3, 0.5, 0.7 MPa, representing the H 2 pressure) nanoalloys were successfully synthesized. Among them, the PtNi 5 -0.3 delivers an extremely low overpotential of 26.8 mV (10 mA cm −2 , 1 m KOH) and Tafel slope of 19.2 mV dec −1 , which surpasses most of the previously reported Pt-based alkaline-HER electrocatalysts (See Figure S1 and Tables S1, Supporting Information). Deep mechanistic insights into the H 2 in situ induced Pt segregation as well as the structure-activity relationship of the PtNi 5 -n were well illustrated based on the combined experimental (H 2 -assisted metal precursor reduction kinetics) and theoretical [density function theory (DFT) simulations] approaches. Finally, we would like to emphasize that the developed H 2 in situ inducing surface segregation strategy is simple (one-pot synthesis without tedious post treatment), environmentally friendly (H 2 as the SDA), and what is most important is efficient to synthesize highly active and economical PtNi 5 nanoalloy as a promising candidate for the industrial application. Figure 1a displays the X-ray diffraction (XRD) patterns of the as-synthesized samples PtNi 5 -n. As can be seen, all the samples exhibit the characteristic diffraction patterns of the face centered cubic (fcc) structure type; and the Pt-Ni nano alloys can be verified to be formed according to the diffraction patterns of the pure Pt fcc (ICSD 04-0802) and Ni fcc (ICSD 04-0850).

Physiochemical Characterizations
Specifically, only the diffraction patterns belonging to the Nirich phase Pt-Ni nanoalloy (2θ = 43.7, 50.7, and 74.5 o ) can be observed for the sample of PtNi 5 -0; while an additional diffraction pattern belonging to the Pt-rich phase Pt-Ni nanoalloy (2θ = 40.5-42.1 o ) is evolved for the samples of PtNi 5 -n (n = 0.1, 0.3, 0.5 and 0.7 MPa), [14] which verifies the strong in situ inducing effect of H 2 on Pt surface segregations, especially for the scenario of P H2 of 0.3 MPa displaying the strongest peak intensity at 2θ of ≈40.8 o . The specific ratios of the Pt and Ni in Pt-rich and Ni-rich phases of the PtNi 5 -n were further predicted based on the Vegard's rule, [24] as shown in Figure 1b (or see Table S2, Supporting Information). Obviously, the PtNi 5 -0.3 possesses the highest Pt ratio of 79% over the Pt-rich phase nanoalloys, quantitatively confirming the strongest Pt segregation effect of H 2 of 0.3 MPa. Figure 2a-c show the Cs-corrected high-angle annular dark-field scanning transmission electron microscopy (Cscorrected HAADF-STEM) images of the PtNi 5 -n (n = 0, 0.3, and 0.5 MPa), which clearly display the well-alloyed nanoparticles (NP) in the truncated octahedral shapes and with the average NP sizes of 6.7, 27.5, and 32.6 nm, respectively. The gradually increased NP sizes along with increasing of H 2 pressures (0-0.5 MPa) can be related to the accelerated metal precursor reduction rate by higher H 2 pressure, which will be further discussed later in 'Kinetic Effect of H 2 on Metal Precursor Reduction' (Section 2.3.1). To further specify the surface elemental distributions, the energy dispersive X-ray spectroscopy (EDS) mapping was thereby conducted, as shown in Figure 2d Figure 1b). The Pt-rich@Ni-rich nanostructure of PtNi 5 -0.5 was less obvious in Figure 2f, which can be related to the comparable Pt (53%) and Ni (47%) ratios of the Pt-rich phase of PtNi 5 -0.5 (see Figure 1b).
To further characterize the generated Pt-rich@Ni-rich core-shell nanostructures, the HRTEM and Cs-corrected  Table S2, Supporting Information). This finding quantitatively validates the Pt-rich@Ni-rich coreshell nanostructure of PtNi 5 -0.3. Similar findings can also be observed for the PtNi 5 -0.5 (see Figure S2f, Supporting Information), whereas only one lattice spacing of 2.10 Å [ fcc-(111)] can be found for PtNi 5 -0 ( Figure S2b, Supporting Information) indicating the uniform growth of PtNi 5 -0. In addition to that, the Cs-corrected HAADF-STEM ( Figure 3c) and combined elemental intensity analysis (Figure 3d) can also support the Pt-rich@Ni-rich nanostructure of PtNi 5 -0.3: i) the bright and dark spot regions respectively representing the Pt-rich shell and Ni-rich inner core can be clearly observed in Figure 3c; ii) and the element intensity gradually grew up along the direction of Line 1 (inner-core → edge sites) due to the Pt surface segregations to generate the Pt-rich shell (Figure 3d).
Due to the fact of that the alkaline-HER majorly occurs on the surface of the catalyst, it is thereby significantly important to specify the surface elemental chemical states as well as contents of the synthesized samples. In light of that, the X-ray photo electron spectroscopy (XPS) was conducted over the PtNi 5n samples (see Figure 4a-d). As shown in Figure 4b, the peaks centered at around ≈71.3 and ≈74.7 eV can be respectively assigned to be Pt 0 4f 7/2 and Pt 0 4f 5/2 , which indicates that the metallic Pt constitutes the dominant Pt species on the surface of PtNi 5 -n, being greatly favorable for the alkaline-HER. As noted, the slight peak shifts of the Pt 4f toward the lower binding energies can be related to the gradually increased Pt surface content of the PtNi 5 -n (n = 0.1, 0.3, 0.5, and 0.7 MPa) samples relative to that of PtNi 5 -0. [25] As for the surface Ni, two type Ni species of the metallic Ni (852.6 eV) and Ni(OH) 2 (855.6 eV) can be found (Figure 4b), wherein the formation of Ni(OH) 2 can be related to the strong water splitting effect of the surface Ni 0 during the synthesis process. [26] The Pt surface contents derived by XPS were further profiled in Figure 4d, being associated with the bulk contents detected by ICP-OES (see Table S3, Supporting Information) as a reference. As can be seen, the Pt surface content of PtNi 5 -0 is comparable to the bulk Pt, verifying the uniform growth of PtNi 5 -0. However, much higher surface Pt contents can be found for the PtNi 5 -n (n = 0.1, 0.3, 0.5, and 0.7 MPa), which decrease following the order of 46% (PtNi 5 -0.3) >42% (PtNi 5 -0.1) >38.4% (PtNi 5 -0.5) >36.3% (PtNi 5 -0.7). This finding quantitatively validates the availability of the H 2 in situ inducing strategy of present work to synthesize highly surface segregated PtNi 5 -n bimetallic nanoalloy, especially for the

Alkaline-HER Activity Measurement
In this section, the alkaline-HER activities of the PtNi 5 -n and commercial Pt/C were evaluated in a H 2 -saturated 1 m KOH solution (PH = 13.7) and with the catalyst sample being directly spread onto the glassy carbon electrode (diameter of 5 mm) under a loading of 10 µg cm −2 (see Figure 5a-f and Table S4, Supporting Information). The polarization curves were obtained based on the linear voltametry sweeping approach under a scan rate of 5 mV s −1 and a rotation rate of 1600 rpm. The potentials presented were versus the reversible hydrogen electrode (RHE) and all polarization curves were current-resistance (iR) corrected.
As shown in Figure 5a, the obvious peak shifts of H upd (underpotentially detected H) toward the lower potentials can be clearly observed for the PtNi 5 -n samples in comparison to that of the commercial Pt/C (0.28 and 0.38 V), which implies the significantly improved alkaline-HER activities of PtNi 5 -n owing to the extensively weakened H + adsorption energy, [27] especially for the scenario of PtNi 5 -0.3 displaying the lowest H upd value of ≈0.1 eV. This finding can be further verified by the derived linear sweep voltammetry curves of Figure 5b displaying the obvious decreases of the overpotentials (η) at -10 mA cm −2 for the PtNi 5 -n samples relative to the commercial Pt/C, especially for the PtNi 5 -0.3 delivering an unprecedentedly low overpotential of 26.8 mV (-10 mA cm −2 ). In addition to that, the volcano shapes can be clearly observed for the derived mass activity (MA) and specific activity (SA) (see Figure 5c , see Table S4, Supporting Information). The PtNi 5 -0.3 also exhibits the highest TOF ( Figure 5d) and lowest Tafel plot (19.2 mV dec −1 ,  5e), which further confirms its highest activity. The long-term test was further conducted to evaluate the reaction stability of PtNi 5 -0.3 (Figure 5f), wherein only a slight current loss of 3% can be found after 24 h's test, reflecting the excellent reaction durability of the best performed sample of PtNi 5 -0.3.

Mechanistic Insight into Specific Effects of H 2
As stated above, the proposed H 2 in situ inducing Pt surface segregation strategy can successfully synthesize highly active PtNi 5 -n nanoalloys for the alkaline-HER being associated with the excellent reaction stability. In this section, the specific in situ inducing effects of H 2 (kinetically and thermodynamically) during PtNi 5 -n synthesis were systematically investigated based on the combined experimental and theoretical approaches.

Kinetic Effect of H2 on Metal Precursor Reduction
Inspired by the literature report [28] suggesting that the metal precursor reduction rate differences determine the final nanostructures of the bimetallic nanoalloy as well as our previous study [29] finding that the H 2 behaving as a reductant greatly favored the metal precursor reductions, herein we investigated the H 2 -mediated metal precursor reduction kinetics based on the Finke-Watzky model [30,31] (see details in Supporting Information) to illustrate the kinetic effect of H 2 on Pt surface segregations to generate the Pt-rich@Ni-rich nanostructures (see Figure 6a-f). As can be seen, the H 2 can efficiently promote the metal precursor (both Ni 2+ and Pt 2+ ) reductions (Figure 6a,b) and the experimental data points can be well-fitted by the constructed kinetic models (represented by the curve lines). Furthermore, the metal precursor reduction rate ratio (R Ni / R Pt ) of PtNi 5 -0 ( Figure 6c) was obviously lower than the stoichiometric Ni/Pt ratio of PtNi 5 (Ni/Pt = 5); whereas much higher reduction rate ratios (R Ni / R Pt > 5) can be clearly observed for the scenarios of PtNi 5 -0.3 and 0.5 (Figure 6d,e) at t = (0.5-1.0) and (0.25-0.5) h, respectively. This would be greatly favorable for the Pt-rich@Ni-rich core-shell nanostructure formations due to the fact of that the quickly reduced metallic Ni (R Ni / R Pt > 5) can be preferentially nucleated to generate the Ni-rich core, which thereby favors subsequent nucleation of the metallic Pt to eventually generate the Pt-rich shell. This finding can also be well supported by the nanostructure characterizations of the PtNi 5 -0.3 samples obtained at different time intervals (t = 15, 30, 45, 60, 120 min) during its synthesis process (see Figure S3, Supporting Information). Additionally, the diverse reduction behaviors of Pt 2+ and Ni 2+ ions under different H 2 pressures (0, 0.3, and 0.5 MPa, Figure 6c-e) can be closely correlated with the related reduction In addition to that, as observed in Figure 2b,c, the accelerated reduction rates of Ni 2+ and Pt 2+ precursors substantially favor growth of the PtNi 5 -0.3 and 0.5 nanostructures, which leads to the larger NP sizes than that of PtNi 5 -0. As noted, although the NP sizes of PtNi 5 -0.3 and 0.5 were larger than that of PtNi 5 -0, the evolved active site (Pt-rich phase shell) behaves much more actively than that of PtNi 5 -0, which can be verified based on the DFT simulations as discussed in 'Alkaline-HER Simulations by DFT and In-Depth Insight' (Section 2.4). Finally, based on the data points profiled in Figure 6c-e, a R Ni-R Pt distribution diagram is further depicted in Figure 6f to distinguish the uniform growth and core-shell growth regions according to the criteria of R Ni /R Pt = 5. Obviously, all the data points were located in the uniform growth region (R Ni /R Pt < 5) for the PtNi 5 -0; while three and two points were located in the core-shell growth regions (R Ni /R Pt > 5) for PtNi 5 -0.3 and 0.5. More data points being located in core-shell growth regions being associated with higher R Ni /R Pt values leads to much more obvious coreshell nanostructure of PtNi 5 -0.3 related to that of PtNi 5 -0.5.

Thermodynamic Effect of H 2 on Pt Surface Content Adjustment by DFT
In this part, the thermodynamic effect of H 2 during PtNi 5 -n (n = 0.3 and 0.5 MPa) synthesis (T = 453 K) was further investigated based on the ab initio thermodynamic analysis. Initially, according to the XPS characterization results of PtNi 5 -0.3 (Pt = 46%) and 0.5 (Pt = 38.4%), two types of PtNi 5 models respectively with the surface Pt content of 50% (being close to © 2021 Wiley-VCH GmbH 46%) and 38% were constructed, as shown in Figure 7a. Based on that, the Gibbs surface free energies (γ) were calculated under the diverse surface hydrogen coverages (θ H = 0-1 ML, see Figure S4, Supporting Information) and H 2 pressures (0.3 and 0.5 MPa), as profiled in Figure 7b-e. As can be seen, the θ H of 1/2 ML leads to PtNi 5 -38% achieving the most thermodynamically stable state (Figure 7b,d); while the θ H of 1/16 ML results in PtNi 5 -50% being of the most stable state (Figure 7c,e) under the PtNi 5 synthesis conditions (P H2 = 0.3 and 0.5 MPa, T = 453 K). Further comparing the γ of PtNi 5 -38% and PtNi 5 -50% related to the most thermodynamically stable states at T = 453 K (see Figure 7f), one can find that the PtNi 5 -50% possesses the lower γ than that of PtNi 5 -38% under the H 2 pressure of 0.3 MPa, which however became converse under the higher H 2 pressure of 0.5 MPa. This finding unravels that the H 2 prefers adjusting the Pt surface content of PtNi 5 -0.3 and PtNi 5 -0.5 to be respectively 50 and 38% to achieve the thermodynamically stable state, which agrees well with the experimental results.

Short Summary
In light of the above investigations, the H 2 in situ inducing effect on Pt surface segregations to generate Pt-rich@Ni-rich nanostructure can be summarized in two aspects. Kinetically, the H 2 working as the reductant results in the metal precursor reduction rate difference being much higher than the stoichiometric ratio of PtNi 5 (R Ni / R Pt >5), especially for the scenario of 0.3 MPa H 2 , eventually leading to the generation of Pt-rich@Ni-rich core-shell nanostructure. Thermodynamically, the H 2 working as the SDA can adjust the surface Pt content through influencing the surface free energy of the exposed facet in the manner of dissociative adsorption mode: the H 2 of 0.3 MPa leading to the surface Pt of 50% being under the stable state, while further increasing the H 2 to 0.5 MPa leading to decreasing of the surface Pt (38%) to achieve its thermodynamically stable state.

Alkaline-HER Simulations by DFT and In-Depth Insight
In this part, the alkaline-HER activity of PtNi 5 -n (n = 0, 0.3, and 0.5 MPa) and commercial Pt/C were simulated by DFT based on the constructed models of PtNi 5 -18% ( Figure S6a, Supporting Information), 38%, 50% ( Figure 7a) and Pt(111) ( Figure S6b, Supporting Information). According to the literature reports, [32] the real reaction route can be justified by the Tafel slope. For the scenario of present work, the reaction would prefer to follow the Volmer-Tafel mechanism due to the low Tafel slopes (<30 mV dec −1 , Figure 5e). ) and H ad (ΔG H ), being of two types of key activity indicators for Volmer and Tafel steps, respectively. [33] Obviously, the PtNi 5 -n samples exhibit much lower absolute values of G H O 2 ∆ and ΔG H than those of Pt/C (see Figure 8b), which indicates much higher HER activity. [34] Moreover, the derived activity decreasing order (PtNi 5 -0.3 > PtNi 5 -0.5 > PtNi 5 -0 > Pt/C) is in good agreement Actually, it is not surprising that the PtNi 5 -n samples exhibit the superior activities in Volmer step relative to that of the Pt/C, being majorly related to the higher intrinsic H 2 O-splitting activity of the dopant Ni. To gain an in-depth insight into the much higher electrochemical activity of PtNi 5 -n with respect to Pt/C in the subsequent Tafel step, the d band center relative to the Fermi level (ε d− ε F ) of the Pt active sites of PtNi 5 -n and Pt/C (see Figure S8d, Supporting Information) was further calculated based on projected density of state (PDOS, see Figure 8c), due to the fact that the Tafel-step activity is largely determined by the specific electronic structure and coordination geometry of the exposed active site. [35] The d band center is efficient to evaluate the metaladsorbate interaction strength; and the higher value of the d band center indicates more emptiness of the antibonding state (being less filled) which thereby leads to the stronger bonding strength (vice versa). [36,37,38] As can be seen (Figure 8c), the (ε d− ε F ) of PtNi 5 -n varying with the surface Pt content due to the ligand effect is obviously lower than that of Pt(111) (−2.44 eV). Moreover, a linear correlation between the (ε d −ε F ) and the ΔG H of PtNi 5 -n and Pt/C can be clearly observed in Figure 8d, which is consistent with the literature report [39] demonstrating that the d band center can be utilized as a key activity descriptor for the alkaline-HER. In light of that, we can get an in-depth insight that the significantly improved Tafel-step activity of the PtNi 5 -n originates from the downward-shifted d band center of the Pt active site, which can significantly weaken the PtH bond strength and thereby pronouncedly enhance the related electrocatalytic activity, especially for the best performed sample of PtNi 5 -0.3 possessing the lowest (ε d −ε F ) value of -3.15 eV (PtNi 5 -50%, Figure 8c), being corresponding to the highest Tafel-step activity.

Correlations Between Physiochemical Property and Alkaline-HER Activity
Combining the activity measurement results with the characterizations (XRD, Cs-corrected HAADF-STEM, EDS mapping, Figure 8. a) DFT simulated Gibbs free energy diagram for alkaline-HER over PtNi 5 -n (n = 0, 0.3, and 0.5 MPa) samples represented by the constructed models of PtNi 5 -18%, PtNi 5 -38%, and PtNi 5 -50%, respectively; as noted, the PtNi 5 -18% is also selected from five different PtNi 5 -18% models with a minimum optimization energy (see Figure S7 HRTEM and XPS) and DFT calculation results together, we can find that the significantly improved alkaline-HER activity of the synthesized PtNi 5 -n (n = 0.1, 0.3, 0.5, and 0.7 MPa) can be closely correlated with the strong in situ inducing effect of H 2 on the Pt surface segregations. As shown in Figure 5c,d, the activity decreasing order unraveled by the MA, SA, and TOF is coincident with the Pt surface content decreasing order, following the trend of PtNi 5 -0.3 (46.0%) > PtNi 5 -0.1 (42.0%) > PtNi 5 -0.5 (38.4%) > PtNi 5 -0.7 (36.3%) > PtNi 5 -0 (18.9%), wherein the H 2 of 0.3 MPa possesses the strongest Pt-surfacesegregation effect due to the resulting highest R Ni / R Pt ratio (see Figure 6d), being greatly favorable for Pt segregation to generate the Pt-rich@Ni-rich core-shell nanostructure; meanwhile, the strong thermodynamic effect of H 2 (0.3 MPa) can also lead to an ultrahigh Pt surface content (46%) of PtNi 5 -0.3 (see Figure 7f), which contributes to its highest alkaline-HER activity. In addition to Pt, the Ni also plays an important role in the alkaline-HER. As depicted in Figure 5a, the obvious negative peak shifts in characteristic OH − adsorption/desorption region of 0.4-1.0 V can be clearly observed for the PtNi 5 -n samples (especially for the best performed PtNi 5 -0.3) relative to that of the commercial Pt/C, which is greatly favorable for the OH − adsorption and thereby benefiting alkaline-HER reaction. Meanwhile, as further unraveled by DFT, due to the synergistic effect between the surface Ni and Pt, the synthesized PtNi 5 -n samples exhibit much lower absolute values of ΔG H2O and ΔG H relative to the commercial Pt/C, which constitutes the major reason of the much higher alkaline-HER activity of PtNi 5 -n, especially for the best performed sample of PtNi 5 -0.3 displaying the lowest values of 0.48 and 0.04 eV, respectively. The PDOSbased d band center simulation results (Figure 8c,d) shed an indepth insight that the downward-shifted d band centers of the Pt active sites of the PtNi 5 -n samples can efficiently reduce the PtH bond strength, thereby leading to the significantly promoted Tafel-step activity, wherein the best performed PtNi 5 -0.3 exhibiting the lowest d band center value of -3.15 eV is corresponding well with its highest Tafel-step activity.

Conclusion
We have developed a simple one-pot synthesis strategy by utilizing H 2 as the SDA to in situ synthesize a series of highly active and economical PtNi 5 -n samples (with extremely low Pt/Ni of 0. 2) for the alkaline-HER. Among them, the samples of PtNi 5 -0.3 delivers a record-high activity (η = 26.8 mV at 10mA cm −2 , Tafel slope of 19.2 mV dec −1 ), due to the strong H 2 in situ inducing effect to generate a Pt-rich@Ni-rich core-shell nanostructure with an ultra-high Pt surface content of 46%. The H 2 majorly functions in two aspects during the synthesis process: i) kinetically, it significantly affects the metal precursor reduction kinetics, wherein the H 2 of 0.3 MPa resulting in the largest Ni 2+ and Pt 2+ reduction rate difference (R Ni /R Pt > 5) greatly favors Pt surface segregation to generate Pt-rich@Ni-rich nanostructure; ii) thermodynamically, it can adjust the Pt surface composition through alerting the surface free energy, for example, 0.

Experimental Section
Synthesis of the PtNi 5 -n Catalysts: A mixing solution of Pt(acac) 2 (0.16 mmol), Ni(acac) 2 (0.80 mmol), PVP (200 mg), aniline (0.5 mL) and benzyl alcohol (20 mL) was prepared in a 50-mL Teflon-lined autoclave and under stirring for 30 min. Thereafter, the autoclave was pressurized by H 2 under diverse pressures of 0, 0.1, 0.3, 0.5, and 0.7 MPa. After that, the sealed autoclave was heated to 180 °C and maintained at this temperature for 6 h with a stirring rate of 200 rpm. After being precipitated, washed by a mixing solution of deionized water and ethanol and dried at 60 °C under vacuum for 24 h, the final products were obtained.
Characterization: The crystalline structure of the catalyst samples were evaluated by X-ray diffraction (XRD) patterns with a Bruker D8-Advance diffractometer using Cu Kα radiation. Based on the XRD, the specific ratios of Pt and Ni in Pt-rich and Ni-rich phases of the PtNi 5 -n nanoalloy were predicted by Vegard's rule; and the lattice spacings were calculated based on the software of JADE 6.5. TEM, HRTEM, and EDS mapping were conducted over a JEM-2100F microscope with an accelerating voltage of 200 kV to characterize the structure morphologies and surface elemental distributions of synthesized samples. The Cs-corrected HAADF-STEM images were collected on the Titan 80-300 microscope (FEI, USA). The XPS was performed over an ESCALAB 250 XPS photoelectron spectrometer with an Mg Kα X-ray source (Thermo, USA) to determine the surface elemental contents. The C 1s at 284.5 eV was utilized as a criterion. The ICP-OES (inductively coupled plasma optical emission spectrometry) was utilized to determine the metal contents in the bulk of catalyst samples over Thermo Fisher Scientific iCAP 6000 Series.
Electrochemical Measurements: An electrochemical workstation (CHI760E, CH Instruments) coupled with a three-electrode cell system was used to evaluate the alkaline-HER activities of the synthesized samples, wherein a saturated Hg/HgO electrode, carbon rod, and glassy carbon were respectively utilized as the reference, counter, and working electrodes. The catalyst ink was prepared by mixing 2 mg of catalyst, 2 mg of carbon black, 0.25 mL of deionized water, 50 µL of Nafion (5 wt%), and 0.7 mL of ethanol together, being followed by ultra-sonication in an ice bath for 0.5 h. The alkaline-HER activity measurements were carried out in a H 2 -saturated 1 m KOH solution (PH = 13.7) with the catalyst samples being directly spread onto the glassy carbon electrode (diameter of 5 mm) with a loading of 10 µg cm −2 . The polarization curves were obtained based on the linear sweep voltametry (LSV) approach under a scan rate of 5 mV s −1 and a rotation rate of 1600 rpm from 0.1 to-0.3 V. Tafel slope was obtained by fitting the linear portion to the Tafel equation: η = blog(j)+ a. The chronoamperometry was measured under a constant potential of 50 mV versus RHE during the reaction stability test. The specific calculation approaches for the turnover frequency (TOF), electrochemically active surface area (ECSA), and mass/ specific activity (MA and SA) are stated in detail in the Supporting Information.
Metal Precursor Kinetics Experiment: The metal precursor reduction kinetics were conducted according to the following procedures: i) initially, the reactor was filled with 10 mL Ni(acac) 2 (0.80 mmol) solution and heated up to 180 °C; ii) after that, the Pt(acac) 2 (0.16 mmol, 20 mL) was quickly pumped into the reactor under a flow rate of 50 mL min −1 ; iii) then the H 2 (0.3 and 0.5 MPa) was fed into the system to start the reduction reaction; iv) the Ni 2+ and Pt 2+ concentrations remaining in the solution was thereby measured by ICP-OES at a time interval of 15 min. The kinetic models were constructed based on the Finke-Watzky model [38,39] which is stated in detail in the Supporting Information.
DFT Calculations: All DFT calculations were performed on the Vienna ab initio simulation package (VASP) using the projector augmented wave www.afm-journal.de www.advancedsciencenews.com

(11 of 11)
© 2021 Wiley-VCH GmbH (PAW) method and Perdew-Burke-Ernzerhof (PBE) exchange correlation functional with an energy cutoff of 400 eV. The convergence threshold was set to be 10 −4 eV in energy and 0.05 eV Å −1 in force. The Brillouin zone was sampled with a 2 × 2 × 1 Monkhorst-Pack k-point grid. The p(4 × 4) unit cell with four layers was employed to calculate the Gibbs surface free energy of the Pt(111) and mixed-PtNi 5 (111) surfaces. Upon slab optimization, the bottom two layers were fully fixed while the other layers were allowed to relax. To avoid non-physical influence, a 10 Å vacuum layer was added along the z direction normal to the surface. Diverse PtNi 5 -n models with surface Pt content of 18, 38, and 50% were constructed; see details in Figures S4, S6, and S7, Supporting Information. Based on the DFT of ab initio thermodynamic (AIT), the thermodynamic effect of H 2 (Pt surface content adjustment) during the PtNi 5 -n synthesis process was evaluated, wherein the surface free energies under diverse H surface coverages (θ H = 0-1 ML) were thereby calculated and with the specific calculation method being described in detail in the Supporting Information. Additionally, the alkaline-HER activity of the PtNi 5 -n samples were also simulated by DFT including the Volmer and Tafel steps. The adsorption Gibbs free energies of H 2 O and H + were thereby calculated as also described in detail in the Supporting Information. Finally, the projected density of state (PDOS) of the active site of Pt over construed models of PtNi 5 -18, 38, and 50% (see Figure S8d, Supporting Information) as well as the Ni of Ni(111) (as a reference) was accordingly derived, based on which the d band centers (ε d ) were obtained.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.