Controlled Aggregation of Cobalt and Platinum Atoms via Plasma Treatment for Exceptional Hydrogen Evolution Reaction Activity

Abstract

Designing and developing highly active, stable, and cost-effective hydrogen evolution reaction (HER) catalysts is crucial in the field of water electrolysis. In this study, we utilize N-doped porous carbon (CoNC) derived from zeolite imidazole metal–organic frameworks (ZIF-67) as support and prepare CoNC-Pt-IM-P via chemical impregnation (CoNC-Pt-IM) and plasma treatment. Systematic analyses reveal that calcined CoNC with pyridinic nitrogen could serve as a robust support to strongly anchor PtCo nanoclusters, while argon plasma treatment could lead to a noticeable aggregation of Co and Pt atoms so as to alter the electronic environment and enhance intrinsic HER catalytic activity. CoNC-Pt-IM-P could exhibit outstanding catalytic activity toward HER, achieving an exceptionally low overpotential of 31 mV at the current density of −10 mA cm⁻² and a Tafel slope of 36 mV dec⁻¹. At an overpotential of 50 mV, its mass activity reaches 4.90 A mg_Pt⁻¹, representing enhancements of 1.5 times compared to CoNC-Pt-IM and 12.3 times compared to commercial 20 wt% Pt/C. Furthermore, it could operate stably for over 110 h at a current density of −10 mA cm⁻², demonstrating its exceptional durability. This work uses plasma treatment to achieve the controllable aggregation of Co and Pt atoms to enhance their catalytic activity, which has the advantage of avoiding excessive particle aggregation compared to the commonly used method of high-temperature calcination.

1. Introduction

Hydrogen, as a clean and sustainable energy carrier, possesses the highest energy density among known fuels [1,2,3,4]. When utilized in fuel cells, it not only offers extremely high energy conversion efficiency but also produces water as the sole product, rendering it a zero-pollution energy source. Therefore, the development of water electrolysis technologies to convert renewable energy into hydrogen is crucial for achieving sustainable energy in the future [5,6]. Significant progress has been made in hydrogen production via water electrolysis and its conversion to electricity through fuel cells. Proton exchange membrane water electrolysis (PEMWE) technology, in particular, stands out due to its superior current density compared to other water electrolysis techniques [7,8,9]. To reduce the energy required for hydrogen production through electrolysis, efficient and stable catalysts are essential to lower the activation energy of the reaction. To date, noble metal catalysts, mainly Pt-based catalysts, are the most effective for the hydrogen evolution reaction (HER) in acidic environments [10,11]. However, state-of-the-art Pt/C catalysts, which typically contain 20 wt% Pt, face limitations in application due to issues including the high cost of Pt, the poor utilization efficiency of Pt, and unsatisfactory stability. Specifically, poor stability mainly stems from Pt particle aggregation, carbon support corrosion, and Pt dissolution in the electrolyte. Therefore, designing and synthesizing Pt-based catalysts with a low noble metal content, high activity, and enhanced stability is crucial. Current optimization strategies to address these challenges include single-atom catalysts, alloying, and combining Pt with carbon-based supports [12,13].

In recent years, metal–organic frameworks (MOFs) have garnered significant attention due to their simple synthesis methods, large surface area, high porosity, and strong loading capacity [14,15,16]. MOFs have been widely used to prepare heteroatom-doped porous carbon as sacrificial template precursors to enhance the catalytic performance of alloy catalysts supported on carbon [17,18]. For instance, Li et al. encapsulated Co@Ir core–shell nanoparticles (NPs) in N-doped porous carbon derived from zeolite imidazole metal–organic frameworks (ZIF-67) and tested its HER activity in 1.0 M KOH. They observed an overpotential of 121 mV at a current density of 10 mA cm⁻². In their Co@Ir/NC-10% catalyst, the Ir shell provided abundant active sites for the oxygen evolution reaction (OER), while the Co-NC framework offered highly active sites for HER [19]. Similarly, Yang et al. developed Co@Pd nanoclusters (NC) through the controlled pyrolysis of ZIF-67, surpassing Pd/C as an HER catalyst and delivering a current density of 10 mA cm⁻² at an overpotential of 98 mV with a Tafel slope of 55 mV dec⁻¹. Furthermore, they fabricated a CoPt-platinum atomic site (PtSA)/nitrogen-doped porous carbon framework (NDPCF) electrocatalyst, which exhibited ultralow overpotentials under both alkaline and acidic conditions at a high current density of −200 mA cm⁻² (over potential: 110 mV in acidic conditions), demonstrating promising long-term durability for up to 100 h or 10,000 cycles. This performance was attributed to the synergistic effects of the PtSA and CoPt alloy [20]. However, comprehensive insights into tuning the microenvironment of catalysts, such as adjusting neighboring Pt atomic sites to enhance the electrocatalytic activity and stability of PtCo nanoclusters, are still lacking.

This study demonstrates that plasma bombardment can effectively modulate the electronic structure of Pt and Co atoms encapsulated within NC nanoframeworks derived from high-temperature (800 °C) pyrolysis. This results in a stable framework with nitrogen vacancies that could securely anchor Pt and Co atoms. After plasma treatment, the Pt and Co atoms are more tightly bonded, while the abundant nitrogen sites provide ample electrons, significantly enhancing the catalyst’s activity and corrosion resistance under acidic conditions. In a 0.5 M H₂SO₄ solution, the CoNC-Pt-IM-P catalyst with a Pt loading of just 2.3 wt% exhibited excellent catalytic performance with overpotentials of only 31 mV at −10 mA cm⁻² and 98 mV at −200 mA cm⁻². Consequently, its mass activity at an overpotential of 50 mV reached a value as high as 4.9 A mgPt⁻¹. Additionally, CoNC-Pt-IM-P can stably operate for over 110 h at −10 mA cm⁻², demonstrating exceptional catalytic stability. This work presents a simple and practical strategy for synthesizing multi-metallic catalysts, which can be extended to the preparation of other bi-metallic or multi-metallic catalysts, offering valuable insights for future research.

2. Experimental Procedure

2.1. Chemicals

2-methylimidazole (2-Mel, Aladdin, AR), cobalt(II) nitrate hexahydrate (Co(NO₃)₂·6H₂O, Aladdin, Shanghai, China, AR), cobalt(II) phthalocyanine (CoPC, J&K Scientific, Beijing, China, 92%), dicyandiamide (DCDA, Energy Chemical, Shanghai, China, 99%), chloroplatinic acid hexahydrate (H₂PtCl₆·6H₂O, Aladdin, AR, Pt ≥ 37.5%), isopropyl alcohol (C₃H₈O, Macklin, Shanghai, China, 99.5%), Nafion (LIGE SCIENCE, Tianjin, China, 5%), carbon paper (TGP-H-060, Sinero, Suzhou, China), Pt/C (HWRK CHEM, Beijing, China, 20 wt%), deionized water, and sulfuric acid (H₂SO₄, Sinopharm Chemical Reagent Co., Ltd., Shanghai, China, 98%) were utilized in this study.

2.2. Synthesis of Electrocatalysts

First, 2.46 g of 2-methylimidazole (2-Mel) and 2.18 g of Co(NO₃)₂·6H₂O were dissolved in 30 mL of methanol. After complete dissolution, the methanol solution of Co(NO₃)₂ was slowly added to the 2-Mel solution while stirring. Once the addition was completed, the mixture was sealed in a beaker and stirred for 12 h. The resulting mixture was then filtered, washed three times, and dried in a vacuum oven at room temperature for 6 h, yielding a uniform purple ZIF-67 powder.

To prepare CoNC, ZIF-67 was placed in a tubular furnace that was filled with argon as a protective atmosphere. It was heated to 800 °C at a rate of 5 °C min⁻¹ and maintained at this temperature for two hours. After cooling, the product was ground to obtain a fine black powder, resulting in the N-doped porous carbon material, which was denoted as CoNC.

The preparation of CoNC-Pt-IM involved adding 100 mg of CoNC support powder to 20 mL of deionized water, followed by the addition of 5 mL of H₂PtCl₆·6H₂O solution (0.83 mg Pt mL⁻¹). The mixture was stirred for 5 h and then filtered, washed three times, and dried in a vacuum oven at room temperature for 6 h. The resulting product was ground to obtain a uniform CoNC-Pt-IM.

The preparation of the control sample, ZIF-67-Pt, followed the same protocol used for CoNC-Pt-IM, but the ZIF-67 was used as the support instead of CoNC, while all other steps remained unchanged.

For the preparation of CoNC-Pt-IM-P, ZIF-67 was first calcined to obtain the support, followed by impregnation with the Pt precursor solution. Subsequently, plasma treatment was applied using argon plasma at a power setting of 100 W for a duration of 10 min. This process yielded the final product, CoNC-Pt-IM-P.

2.3. Characterizations

X-ray Powder Diffraction (XRD) data were collected on a Bruker D8 Advance X-ray Polycrystalline Diffractometer at 40 kV/40 mA with Cu Ka radiation (k = 1.541874 Å) in the angular range of 5–80° for the 2θ angle. X-ray photoemission spectroscopy (XPS, Al-K-alpha, Thermo Scientific, Waltham, MA, USA) and scanning electron microscopy (SEM, Jeol, Tokyo, Japan) were carried out on carbon tape. Transmission electron microscopy (TEM, Jeol) measurements were completed on a carbon-coated copper TEM grid support.

2.4. Electrochemical Measurements

All tests were conducted using an Autolab (Metrohm PGSTAT302N, Herisau, Switzerland) electrochemical workstation in a three-electrode configuration. A carbon rod and Hg/Hg₂SO₄/saturated K₂SO₄ electrodes were used as the counter and reference electrodes, respectively, and the working electrode was prepared by mixing the samples (including CoNC, CoNC-Pt-IM, CoNC-Pt-IM-P, and commercial 20 wt%Pt/C) with Nafion and drop-casting them onto carbon paper for testing at 25 °C. The catalyst preparation procedure was as follows: 10 mg of catalyst powder, 100 μL of Nafion solution (5 wt%), and 900 μL of isopropanol were mixed and sonicated for 30 min until a homogeneous suspension was obtained; then, 100 μL of the prepared slurry was drop-cast onto a piece of carbon paper (1 × 1.5 cm², with a coated area of 1 × 1 cm²) and left to dry for 30 min until the solvent was evaporated. This resulted in a uniform layer of catalyst powder that adhered to the carbon paper with a catalyst loading of 1.0 mg cm⁻². All electrochemical results were reported with respect to the reversible hydrogen electrode (RHE). Linear sweep voltammetry (LSV) was performed by sweeping the potential at a rate of 5 mV s⁻¹ over a range of −0.7 V to 0.0 V in 0.5 M H₂SO₄. Data were collected after stable cyclic voltammetry (CV) curves were obtained. The measured potentials against Hg/Hg₂SO₄ were converted to RHE using the following equation:

E_(RHE) = E_Hg/Hg2SO4 + 0.059 × pH + 0.65

where E_Hg/Hg2SO4 is the working potential. To evaluate the charge transfer properties of the electrode material, electrochemical impedance spectroscopy (EIS) measurements were performed. The parameters were set as follows: a frequency range from 0.01 Hz to 100 kHz with a voltage of −0.7 V and an amplitude of 5 mV.

3. Results and Discussion

3.1. Catalyst Morphology

The synthesis process of the samples is illustrated in Figure 1a. The porous N-doped carbon (CoNC) was obtained by calcining ZIF-67 and then impregnating it in the Pt precursor solution to yield CoNC-Pt-IM. The plasma treatment was applied to achieve the stable anchoring of CoNC-Pt-IM-P. The morphology of the synthesized materials was examined using scanning electron microscopy (SEM). As shown in Figure S1a,b, the synthesized ZIF-67 exhibited a smooth surface and a dodecahedral structure with an average size of approximately 400 nm, consistent with literature reports [21,22]. After calcination, the obtained CoNC exhibited a rough and wrinkled surface while retaining its dodecahedral shape, indicating that the carbon framework remained intact (Figure 1b,c and Figure S1c,d). Moreover, the CoNC-Pt-IM maintained its dodecahedral morphology after impregnation, showing no significant changes to the rough surface compared to CoNC (Figure 1d,e) and no observable particle loading. In contrast, after directly immersing ZIF-67 in H₂PtCl₆ solution, ZIF-67-Pt experienced noticeable hydrolysis and structural collapse, and thus, its morphology became irregular (Figure S1e,f). This demonstrated the superior structural stability of the CoNC support obtained through calcination [23]. Additionally, CoNC exhibited distinct porous structures, which was expected to facilitate electrolyte penetration and allow ions to interact more effectively at the solid–liquid interface. Similarly, CoNC-Pt-IM-P retained the same configuration after the plasma treatment [24].

As shown in the transmission electron microscope (TEM) images (Figure 2a,b), the CoNC-Pt-IM support features uniformly distributed metal nanoparticles with an average particle size of 8.86 ± 5.26 nm. Subsequent measurements of the lattice spacings of the supported metal particles indicate values of 0.21 nm and 0.23 nm (Figure 2c–e), corresponding to the Co (111) and Pt (111) crystal planes, respectively [16,25]. This suggests the independent presence of both Co nanoparticles and Pt nanoparticles on the support, with a notable distance between them. High-Angle Annular Dark Field Scanning Transmission Electron Microscopy (HAADF-STEM) images display the elemental distribution (Figure S2a–d), further confirming a relatively uniform distribution of Pt atoms, which had no significant bonding to Co atoms. In contrast, CoNC-Pt-IM-P exhibits a distinct distribution of metal nanoparticles at a larger scale (Figure 2f), and the average particle size increases to 9.26 ± 4.82 nm (Figure 2g). Figure 2h,i reveal two neighboring metal nanoparticles with lattice spacings of 0.21 nm and 0.23 nm, corresponding to the Co (111) and Pt (111) crystal planes. Unlike the spatial separation observed in CoNC-Pt-IM, the CoNC-Pt-IM-P samples clearly show a closer proximity of the metal nanoparticles, indicating partial alloying. Additionally, as displayed in Figure 2j–m, compared to the uniform distribution of Pt in CoNC-Pt-IM, the Pt atoms in CoNC-Pt-IM-P show tighter integration with Co atoms. However, CoNC-Pt-IM and CoNC-Pt-IM-P both exhibit significant nitrogen doping with a relatively uniform distribution and no noticeable differences (Figure S3).

3.2. Chemical Structure and Surface Composition

X-ray diffraction (XRD) data of the prepared samples are presented in Figure 3a. The XRD characteristic peaks of the synthesized MOF material matched well with the standard pattern of ZIF-67, confirming its successful synthesis (Figure S4a). However, after impregnation, the hydrolysis occurring during the impregnation process could have led to structure degradation so that the intensity of the characteristic peaks in the ZIF-67-Pt sample decreased significantly. This is consistent with the abovementioned SEM images (Figure S1e,f). As depicted in Figure 3a, after calcination, the characteristic peaks of ZIF-67 disappeared in the CoNC sample, with distinct diffraction peaks newly appearing at 44°, 51°, and 76°, which could be attributed to the metallic Co (PDF#15-0806), confirming the formation of Co nanoparticles [26]. The diffraction peaks of the CoNC-Pt-IM sample were nearly identical to those of CoNC, indicating that impregnation did not significantly alter the properties of the pristine support. This demonstrates the excellent stability of the carbon framework. After the plasma treatment, the intensity of the metallic Co diffraction peaks showed a significant increase in the XRD pattern of CoNC-Pt-IM-P, suggesting the growth of Co nanoparticles. Additionally, the peak at 52° should be attributed to the Co (200) plane. The absence of distinct crystalline peaks for Pt is likely due to the ultrasmall size and low content of Pt particles. Meanwhile, some of the original Co diffraction peaks had a slight shift, implying the partial alloying of PtCo.

To further elucidate the structural differences among the samples, X-ray photoelectron spectroscopy (XPS) was employed to investigate the surface chemical composition of these catalysts. The survey-scan XPS spectra of the catalysts (Figure 3b and Figure S5) reveal peaks corresponding to Co 2p, Pt 4f, C 1s, and N 1s, with an additional peak around 520 eV attributed to O 1s from the adsorbed air on the catalyst surface. The Pt 4f spectrum of CoNC-Pt-IM-P shows two main peaks, Pt 4f_7/2 (71.23 eV) and Pt 4f_5/2 (72.45 eV), attributed to the Pt⁰ and Pt²⁺ oxidation states (Figure 3c) [27,28]. Compared with CoNC-Pt-IM, the proportion of Pt⁰ in CoNC-Pt-IM-P decreased from 55.3% to 52.3%, while Pt²⁺ increased from 44.6% to 47.7%, suggesting electronic transfer due to the partial alloying of PtCo. For comparison, in the Pt 4f spectrum of ZIF-67-Pt (Figure S6, Table S1), the peaks were mainly identified as Pt²⁺ (72.39 eV) and Pt⁴⁺ (75.74 eV), indicating that despite using the same impregnation method, the unique nature of the CoNC support was crucial for facilitating the reduction and anchoring of Pt during the process. The lower ratio of metallic Pt may have resulted in the decreased HER activity.

Furthermore, the high-resolution Co 2p spectra of CoNC-Pt-IM-P (Figure 3d, Table S2) displayed peaks corresponding to Co⁰ (778.5 eV), Co²⁺ (780.0 eV), Co³⁺ (782.5 eV), and Co-N_x (796.1 eV) [29,30]. This indicated that cobalt species in CoNC-Pt-IM-P predominantly existed as cobalt nanoclusters featuring Co²⁺-N coordination states. The C 1s spectra of CoNC-Pt-IM and CoNC-Pt-IM-P (Figure 3e and Figure S6) both exhibited three peaks that were associated with graphitic C-C (284.8 eV), C-N (286.0 eV), and C-O [31,32]. In the N 1s spectra (Figure 3f), four peaks appeared at 398.2, 399.4, 401.0, and 403.5 eV, corresponding to pyridinic N, pyrrolic N, graphitic N, and oxidized N states, respectively [33]. Notably, the content of pyridinic nitrogen in CoNC-Pt-IM-P was higher than that in CoNC-Pt-IM (Figure S7d), suggesting that the PtCo alloy may be embedded within the CoNC framework.

3.3. Electrochemical Activity Evaluation in Acidic Media

To evaluate the hydrogen evolution reaction (HER) behavior, electrochemical tests were conducted on CoNC, CoNC-Pt-IM, CoNC-Pt-IM-P, and commercial Pt/C (20 wt%) in a 0.5 M H₂SO₄ aqueous solution. All potentials were referenced to the reversible hydrogen electrode (RHE) and corrected for 90% iR to eliminate the ohmic drop in the electrolyte. As shown in Figure 4a,b and Figure S8, the HER activity of the pristine CoNC sample was poor, exhibiting a Tafel slope of 189 mV dec⁻¹ and an overpotential of 161 mV at −10 mA cm⁻². After impregnation with Pt, CoNC-Pt-IM demonstrated significant improvement, with the Tafel slope and overpotential decreasing to 66 mV dec⁻¹ and 33 mV, respectively, indicating the effective adhesion of Pt to the CoNC nanoframe and enhanced catalytic activity. For comparison, ZIF-67-Pt exhibited an overpotential of 109 mV at −10 mA cm⁻² (Figure S7), both significantly inferior to CoNC-Pt-IM. The results highlight that the stable CoNC nanoframe could facilitate electrolyte penetration and enhance ion transfer at the solid–liquid interface. After subsequent plasma treatment, the HER catalytic performance of CoNC-Pt-IM-P further improved, achieving an ultralow overpotential of 31 mV at 10 mA cm⁻² and a Tafel slope of 36 mV dec⁻¹, surpassing the commercial 20wt% Pt/C catalyst, which displayed an overpotential of 50 mV and a Tafel slope of 39 mV. Given that the CoNC-Pt-IM-P catalyst possessed a Pt loading of just 2.3 wt% (obtained from ICP-AES), the mass activity of CoNC-Pt-IM-P was calculated to be 4.9 A mg_Pt⁻¹, significantly higher than the 0.4 A mg_Pt⁻¹ for Pt/C.

The plasma treatment did not significantly alter the CoNC support or Pt loading amount, suggesting that it was the alloy formation that led to the significantly enhanced intrinsic catalytic activity. The electrochemical active surface area (ECSA) was characterized by measuring the double-layer capacitance (C_dl) of the catalysts within a non-Faradaic potential range. As illustrated in Figure S9, the C_dl of CoNC-Pt-IM-P (63.42 mF cm⁻²) was greater than that of CoNC-Pt-IM (35.15 mF cm⁻²) and CoNC (24.45 mF cm⁻²), indicating that CoNC-Pt-IM-P provided more HER active sites. The electrochemical impedance spectroscopy (EIS) results (Figure 3d,e) reveal that the charge transfer resistance (R_ct) of CoNC-Pt-IM-P (0.47 Ω) was significantly lower than that of CoNC-Pt-IM (237.9 Ω) and CoNC (73.7 Ω), demonstrating its superior charge transfer capability. At −0.07 V, compared to the pristine CoNC (73.7 Ω), CoNC-Pt-IM-P displayed a much lower R_ct (0.47 Ω), indicating minimal interfacial resistance and a faster charge transfer process. Moreover, to meet industrial application requirements, HER catalysts should not only possess outstanding catalytic performance but also exhibit good stability. As shown in Figure 4e, the chronoamperometric tests indicated that CoNC-Pt-IM-P could stably operate for over 110 h at a current density of −10 mA cm⁻², demonstrating its excellent electrochemical stability [34,35].

4. Conclusions

In this study, we successfully synthesized a novel CoNC-Pt-IM-P electrocatalyst through a combination of MOF template calcination and plasma treatment. Both CoNC-Pt-IM and CoNC-Pt-IM-P could retain their dodecahedral morphology, and elemental mapping confirmed the uniform surface loading of Pt. Moreover, the argon plasma treatment led to the noticeable aggregation of Co and Pt atoms so as to alter the electronic environment and enhance the intrinsic HER catalytic activity. CoNC-Pt-IM-P exhibited excellent HER activity, with an overpotential of only 31 mV at a current density of −10 mA cm⁻², which was lower than that of CoNC-Pt-IM (33 mV) and commercial Pt/C catalysts (50 mV). Meanwhile, its mass activity reached 4.9 A mg⁻¹ Pt, which was 1.5 times that of CoNC-Pt-IM and 12.2 times that of the commercial Pt/C catalyst. Furthermore, CoNC-Pt-IM-P demonstrated remarkable stability, maintaining performance with negligible voltage changes over 110 h at a current density of −10 mA cm⁻². This work successfully used plasma treatment to achieve the controllable aggregation of Co and Pt atoms to enhance their catalytic activity, providing significant insights for the modification of multi-metal catalysts.