Conductive Metal–Organic Frameworks with Extra Metallic Sites as an Efficient Electrocatalyst for the Hydrogen Evolution Reaction

Abstract

The 2D conductive metal–organic frameworks (MOFs) are expected to be an ideal electrocatalyst due to their high utilization of metal atoms. Exploring a new conjugated ligand with extra active metallic center can further boost the structural advantages of conductive MOFs. In this work, hexaiminohexaazatrinaphthalene (HAHATN) is employed as a conjugated ligand to construct bimetallic sited conductive MOFs (M2₃(M1₃∙HAHATN)₂) with an extra M–N₂ moiety. Density functional theory (DFT) calculations demonstrate that the 2D conjugated framework renders M2₃(M1₃∙HAHATN)₂ a high electric conductivity with narrow bandgap (0.19 eV) for electron transfer and a favorable in‐plane porous structure (2.7 nm) for mass transfer. Moreover, the metal atom at the extra M–N₂ moiety has a higher unsaturation degree than that at M–N₄ linkage, resulting in a stronger ability to donate electrons for enhancing electroactivity. These characteristics endow the new conductive MOFs with an enhanced electroactivity for hydrogen evolution reaction (HER) electrocatalysis. Among the series of M2₃(M1₃∙HAHATN)₂ MOF, Ni₃(Ni₃∙HAHATN)₂ nanosheets with the optimal structure exhibit a small overpotential of 115 mV at 10 mA cm⁻², low Tafel slope of (45.6 mV dec⁻¹), and promising electrocatalytic stability for HER. This work provides an effective strategy for designing conductive MOFs with a favorable structure for electrocatalysis.

Hexaiminohexaazatrinaphthalene (HAHATN) is employed as a conjugated ligand to introduce an extra M–N₂ moiety for constructing M2₃(M1₃∙HAHATN)₂ conductive metal–organic frameworks (MOFs) (M1 = Ni or Co or Cu; M2 = Ni or Cu). The extra metal atom in the M1–N₂ moiety exhibits a higher unsaturated degree than that in traditional M2–N₄ linkage, which remarkably enhances the electroactivity of conductive MOFs.

1. Introduction

Water electrolysis is a pivotal strategy to generate clean hydrogen, a promising alternative to fossil fuels.^{[ 1 ]} To date, the efficient electrocatalysts for hydrogen evolution reaction (HER) are mainly inorganic nanostructures.^{[ 2 ]} The active sites of these nanostructures for HER only exist on their surface and/or edge.^{[ 3 ]} The major unexposed metal atoms in the bulk phase are inert for HER, which strictly limits the metal atom utilization.^{[ 4} , ⁵ , ^{6 ]}

Due to the exposed metal site, large surface area and structural controllability,^{[ 7} , ^{8 ]} metal–organic frameworks (MOFs) are considered as preferred heterogeneous catalysts for various catalytic reactions.^{[ 9} , ¹⁰ , ^{11 ]} Particularly, the conductive subgroup of MOFs possesses a high ability of electron transfer beyond traditional MOFs. Consequently, the application of MOFs in electrocatalysis eventually becomes a reality.^{[ 12} , ^{13 ]} Conductive MOFs consist of transition metal atoms and conjugated organic ligands by M–N₄ linkages, which provide a fully π‐conjugated structure with exceptional electrical conductivity.^{[ 14 ]} Up to now, the conductive MOFs have been employed as efficient electrocatalysts for HER, oxygen evolution reaction, and oxygen reduction reaction.^{[ 15} , ¹⁶ , ^{17 ]} Although these conductive MOFs seem to possess the potential application for electrocatalysis due to the extraordinary structure, their electroactivity remains inferior to the practical application. The theoretical and experimental investigations have demonstrated that the metal atoms in the M–N₄ linkages still remain the original oxidation state during the electrocatalytic process, which hardly shows effective activity for electrocatalysis.^{[ 18 ]} At present, hexaiminotriphenylene (HITP, Scheme S1, Supporting Information) and/or its analogs are usually employed as the conjugated organic ligands for conductive MOFs,^{[ 19} , ²⁰ , ²¹ , ^{22 ]} which barely provide other coordinated sites to be high active metal center for electrocatalysis. Thus, designing new conjugated organic ligand with extra coordinated sites to incorporate effective active centers should be the primary priority for improving electrocatalytic performance of conductive MOFs.

Hexaazatriphenylene (HATN, Scheme S2, Supporting Information) is a N‐containing tris(bidentate) polyheterocyclic ligand with electron‐deficient conjugated planar structure. The bidentate tertamine of HATN can coordinate various metal ions to form M₃∙HATN with a two‐coordinated (M–N₂) moiety.^{[ 23 ]} This structure can expose unoccupied positions of metal atom in a high content (≈14 at%), which are regarded as an efficient single‐atom center for catalysis.^{[ 24} , ^{25 ]} HATN structure is also propitious to stabilize metallic ion with valence variation, which can serve as a key active intermediate for the electrocatalytic reaction.^{[ 26 ]} Since the structure of HATN are extreme controllable, various analogs have been developed as fundamental moieties to fabricate supramolecular systems, especially metallic‐supramolecular frameworks.^{[ 27} , ²⁸ , ^{29 ]} As a result, HATN may be the optimal structure as a conjugated ligand to fabricate conductive MOFs. In this novel framework, the extra metal atoms, coordinated by bidentate tertamine, can serve as efficient active centers for electrocatalysis, similar to those of single‐atom catalysts.^{[ 30 ]} The fully conjugated structure and M–N₄ linkages of conductive MOFs can offer strong guarantees for electron transfer, further improving electroactivity.^{[ 31} , ^{32 ]} Additionally, the conductive MOFs, consisting of HATN‐based six‐member ring, have a larger structure with an expanded in‐plane pore (>2 nm), which can enhance the mass diffusion of small molecules in the framework.^{[ 33 ]} Thus, HATN‐based conductive MOFs may be ideal electrocatalysts with high activity. Unfortunately, HATN‐based conductive MOFs and their applications are rarely investigated so far.

In this work, hexaiminohexaazatrinaphthalene (HAHATN, Scheme S3, Supporting Information), an analog of HATN, is designed as the organic ligand to fabricate various bimetallic sited conductive MOFs with in‐plane mesoporous structures (2.7 nm). We present computational and experimental evidences for investigating the active sites in these conductive MOFs. Both theoretical and experimental results indicate the bidentate‐based metal site (M–N₂) possesses higher activity than that of M–N₄ linkage for HER. Ni₃(Ni₃∙HAHAT)₂ nanosheets, the bimetallic sited conductive MOFs with optimal chemical compositions in this work, exhibit outstanding HER performances in alkaline solution, such as low overpotential of 115 mV at 10 mA cm⁻², small Tafel plot of 45.6 mV dec⁻¹, and excellent electrocatalytic stability. Therefore, employing HATN analog as conjugated organic ligand is an effective strategy to resolve the deficiency of traditional conductive MOFs in the field of electrocatalysis, which has extraordinary significance for designing and synthesizing of next‐generation conductive MOF‐based electrocatalysts.

2. Results and Discussion

Various nickel precursors including Ni‐based MOFs are usually used to construct efficient HER electrocatalysts due to their low cost and high activity, including Ni‐based MOFs.^{[ 34} , ^{35 ]} In this work, we employ Ni₃∙HAHATN as a conjugated ligand to construct conductive Ni₃(Ni₃∙HAHATN)₂ MOFs via a synthetic step and two consecutive coordination reactions (Scheme 1 ). The products in each synthetic process were preliminarily confirmed through nuclear magnetic resonance (NMR) and Fourier transform infrared (FTIR) technology (Figures S1–S5, Supporting Information). Similar to the traditional Ni₃(HITP)₂, Ni–N₄ structure in Ni₃(Ni₃∙HAHATN)₂ is used as linkage to construct the conjugated framework. Unfortunately, Ni atom at Ni–N₄ linkage is low‐active for the electrocatalytic reactions due to its invariant oxidation state during electrochemical process.^{[ 18 ]} Beyond the Ni–N₄ linkage, Ni₃(Ni₃∙HAHATN)₂ also possesses extra Ni–N₂ moieties with higher unsaturation degree, which can exhibit better electroactivity for the electrocatalytic reactions.

Scheme 1.

Synthetic diagram of conductive Ni₃(Ni₃∙HAHATN)₂ MOFs.

Powder X‐ray diffraction (PXRD) measurement reveals Ni₃(Ni₃∙HAHATN)₂ has high crystallinity (Figure 1 ). For clear clarification, we simulate the structure of Ni₃(Ni₃∙HAHATN)₂. After geometry optimization, Ni₃(Ni₃∙HAHATN)₂ shows a 2D metal–organic coordinated framework, constituted by a periodic hexagonal ring of six Ni₃∙HAHATN moieties connected through Ni–N₄ linkages (inset in Figure 1). An expanded mesoporous structure (2.7 nm) uniformly exists in the plane of Ni₃(Ni₃∙HAHATN)₂ due to the larger ligand of Ni₃∙HAHATN. As with P6/mmm symmetry, this type of MOFs has equivalent in‐plane lattice lengths approaching to 29.5 Å (a = b; α: 90°, β: 90°, γ  : 120°) with interlayer separation of ≈3.43 Å. The simulated PXRD result matches well with the experimental pattern, confirming the successful synthesis of Ni₃(Ni₃∙HAHATN)₂. Consequently, the prominent peaks at 3.34° and 5.98° in PXRD can be separately assigned to (100) and (200) plane, respectively. A weak diffraction peak at 25.94° is corresponding to the (001) plane (3.43 Å), and inferring Ni₃(Ni₃∙HAHATN)₂ can exist as few‐layered structure.

Figure 1.

Experimental and simulated PXRD patterns of Ni₃(Ni₃∙HAHATN)₂. The inset shows the optimization structure of Ni₃(Ni₃∙HAHATN)₂ slab.

Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were carried out to analyze the morphology and structure of Ni₃(Ni₃∙HAHATN)₂. SEM images show Ni₃(Ni₃∙HAHATN)₂ exhibits an obviously petaloid morphology in large‐scale, consisting of continuous thin‐layered nanosheets with abundant wrinkles (Figure 2a). The interaction of these nanosheets brings an extensively hierarchical porous structure, and the size of pores is in range of dozens of nanometers. The anomalous porous morphology can efficiently increase electroactivity due to exposure of active centers and rapid mass diffusion for Ni₃(Ni₃∙HAHATN)₂ nanosheets. N₂ adsorption/desorption isotherm was carried out to investigate the porous feature of Ni₃(Ni₃∙HAHATN)₂ nanosheets (Figure S6, Supporting Information). Pore size distribution curve shows the existence of two type mesoporous structures, and the wide peak at 46.3 nm is originated from hierarchical porous structure of the interlayer interaction. An obvious sharp peak at ≈2.6 nm is also observed, which is attributed to the in‐plane pores in keeping with the crystal structure of geometric optimization. Moreover, in this framework structure, the unsaturated nickel atom in Ni–N₂ moiety is suspended in the in‐plane mesoporous structure, which is benefit to capture the proton. Similar to the SEM analysis, TEM image illustrates the Ni₃(Ni₃∙HAHATN)₂ is made up of smooth silk‐like layered structures (Figure 2b). Generally, black‐line region in high‐resolution transmission electron microscopy (HRTEM) image is originated from the perpendicular nanosheets to support substrate, which can be used to estimate the thickness of nanosheets (Figure S7, Supporting Information).^{[ 36 ]} The width of the black‐line regions in this work is less than 1.6 nm, corresponding to the thickness of Ni₃(Ni₃∙HAHATN)₂ nanosheets. Additionally, the analysis of atomic force microscopy (AFM) indicates that the thickness of Ni₃(Ni₃∙HAHATN)₂ nanosheets is 1.6 nm (Figure S8, Supporting Information), in consistent with TEM characterization.

Figure 2.

a) SEM, b) TEM and c) EDX Mapping images of Ni₃(Ni₃∙HAHATN)₂ nanosheets.

The element composition and surface chemical state of Ni₃(Ni₃∙HAHATN)₂ nanosheets were identified by energy dispersive spectroscopy (EDS) elemental mapping and X‐ray photoelectron spectroscopy (XPS) technologies. Mapping analysis reveals the uniform distribution of nickel and nitrogen element on Ni₃(Ni₃∙HAHATN)₂ nanosheets (Figure 2c). XPS spectrum also proves the Ni₃(Ni₃∙HAHATN)₂ nanosheets are composed of Ni, N, and C elements; and the Ni content on the surface is about 12 at%, similar to the theoretical value. After subtracting Shirley background, high‐resolution XPS spectra are fitted by mixture function of Lorentzian and Gaussian (Figure 3a–c). The existence of Ni^ǁ can be testified in high‐resolution Ni 2p XPS spectrum (Figure 3a). The two typical peaks at 853.7 and 871.5 eV can be assigned to Ni 2p3/2 and Ni 2p1/2 of Ni–N₂ moiety, while another two peaks at 859.9 and 877.5 eV can be assigned to the satellite peaks.^{[ 37} , ^{38 ]} High‐resolution N 1s spectrum is deconvoluted into two peaks of C—N bond at 396.3 eV and Ni—N at 397.5 eV (Figure 3b).^{[ 39 ]} According to C spectrum, three subpeaks at 282.8, 283.9, and 286.6 eV can be assigned to the C 1s orbital of C=C, C—N, C—O, respectively (Figure 3c). All the analytic results indicate the Ni₃(Ni₃∙HAHATN)₂ nanosheets are achieved in this work.

Figure 3.

High‐resolution XPS spectra of a) Ni 2p, b) N 1s, and c) C 1s of Ni₃(Ni₃∙HAHATN)₂. d) Calculated partial density of state (PDOS) of mono‐layered Ni₃(Ni₃∙HAHATN)₂ slab.

In theoretical, the conjugated ligand and Ni–N₄ linkage can impart Ni₃(Ni₃∙HAHATN)₂ with a fully π‐conjugated framework with excellent conductivity. Four‐point probe test manifests that the compressed pellet of Ni₃(Ni₃∙HAHATN)₂ nanosheets possesses a high electrical conductivity with 2 S cm⁻². Additionally, the density functional theory (DFT) calculation was carried out to estimate the electronic band structure of Ni₃(Ni₃∙HAHATN)₂ by using the gradient‐corrected functional (GGA‐PBE) function (Figure 3d). The result of partial density of states (PDOS) indicates that single‐layered slab of Ni₃(Ni₃∙HAHATN)₂ exhibits a very narrow band‐gap of ≈0.19 eV, suggesting that the Ni₃(Ni₃∙HAHATN)₂ nanosheets should possess excellent conductivity for rapid electron transfer during the electrocatalytic process.

By regulating metal ionic species, a series of fully conjugated conductive MOFs M2₃(M1₃∙HAHATN)₂ (M1:Co or Cu, M2:Ni or Cu) was obtained in this work (Figure S9, Supporting Information). The various products of M2₃(M1₃∙HAHATN)₂ were confirmed by FTIR spectra (Figures S10 and S11, Supporting Information). After geometry optimization, all M2₃(M1₃∙HAHATN)₂ exhibit 2D fully conjugated structures, similar to Ni₃(Ni₃∙HAHATN)₂. All conductive MOFs belong to P6/mmm symmetry with analogous cell parameters. According to coordinated metal ionic species, M2₃(M1₃∙HAHATN)₂ presents different M1–N₂ moiety in in‐plane mesopores and M2–N₄ linkage. DFT calculation indicates single‐layered M2₃(M1₃∙HAHATN)₂ slabs all have very narrow band‐gap with neglectable differences (Figures S12–S14, Supporting Information). Owing to the analogous crystal structures, XRD patterns of three M2₃(M1₃∙HAHATN)₂ exhibit obviously diffraction peaks in low angle (Figure S15, Supporting Information), similar to Ni₃(Ni₃∙HAHATN)₂. TEM images show these M2₃(M1₃∙HAHATN)₂ possess wrinkled thin‐layered morphologies with the thickness of 1–2 nm (Figure S16, Supporting Information). Meantime, homogeneous layered structures are also observed in SEM images of M2₃(M1₃∙HAHATN)₂ (Figure S17, Supporting Information). More importantly, EDS mapping analysis shows two species of metal atoms uniformly distribute in the nanosheets of M2₃(M1₃∙HAHATN)₂, which is direct evidence for the existence of two type metallic sites (M1–N₂ and M2–N₄) (Figures S18–S20, Supporting Information). The chemical composition and states of M2₃(M1₃∙HAHATN)₂ were also determined by XPS measurements (Figures S21–S23, Supporting Information). Obviously, employing HAT structure as a conjugated ligand is an effective strategy to construct novel conductive MOFs with bimetallic sites, which is a significant experience for designing and synthesizing of next‐generation conductive MOFs with multiple functions.

The electrochemical property of Ni₃(Ni₃∙HAHATN)₂ and traditional conductive MOFs Ni₃(hexaiminotriphenylene)₂ (Ni₃(HITP)₂) were investigated by various electrochemical technologies. In this work, all reported potentials were converted to reversible hydrogen electrode (RHE), and the current densities obtained were normalized by the geometric area of the working electrode. The linear sweep voltammetry (LSV) tests were carried out in N₂‐saturated 0.1 m KOH solution at a scan rate of 5 mV s⁻¹ to evaluate the HER activities of Ni₃(Ni₃∙HAHATN)₂ and Ni₃(HITP)₂ (Figure 4a). The LSV curves after iR correction (95%) display that Ni₃(Ni₃∙HAHATN)₂ has an HER onset potential of 12 mV, which is much smaller than that (51 mV) of Ni₃(HITP)₂. The η₁₀, defined as the overpotential at a current density of 10 mA cm⁻², is 115 mV for Ni₃(Ni₃∙HAHATN)₂, dramatically superior to the η₁₀ of 176 mV for Ni₃(HITP)₂. The result explicitly indicates the Ni–N₂ moiety plays a significant role in the HER activity of Ni₃(Ni₃∙HAHATN)₂. Compared to the similar reports of MOF‐ or MOF‐derived electrocatalysts,^{[ 15} , ³⁵ , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ , ⁴⁷ , ⁴⁸ , ^{49 ]} the electroactivity of Ni₃(Ni₃∙HAHATN)₂ exhibits obviously superiority in this field (Table 1 ). To investigate the effect of M1–N₂ site, the HER activities of various Ni₃(M1₃∙HAHATN)₂ samples were also evaluated by LSV measurement (Figure 4b). The η₁₀ values of HER on Ni₃(Cu₃∙HAHATN)₂, Ni₃(Co₃∙HAHATN)₂, and Ni₃(Ni₃∙HAHATN)₂ are 207, 162, and 115 mV, respectively. Obviously, the gradually enhanced HER activity indicates the active discrepancy of metal ionic species: Cu–N₂ < Co–N₂ < Ni–N₂, which is another strong evidence for M1–N₂‐based active center in M2₃(M1₃∙HAHATN)₂. To further verify the active sites of M2₃(M1₃∙HAHATN)₂, the electrocatalytic activity of Cu₃(Cu₃∙HAHATN)₂ sample was tested. By comparison, the η₁₀ value (230 mV) of Cu₃(Cu₃∙HAHATN)₂ is just slightly larger than that of Ni₃(Cu₃∙HAHATN)₂. Thus, this result reveals that the M1–N₂ moiety plays a major role in the electroactivity of M2₃(M1₃∙HAHATN)₂ toward HER. Additionally, cathodic current density is also an important factor to evaluate the HER activity of electrocatalyst. The Ni₃(Ni₃∙HAHATN)₂ sample exhibits a high current density of 21.2 mA cm⁻² at potentials of −0.15 V, about four times as high as that of Ni₃(HITP)₂. The HER current densities for Ni₃(Co₃∙HAHATN)₂, Ni₃(Cu₃∙HAHATN)₂, and Cu₃(Cu₃∙HAHATN)₂ are 7.6, 3.2, and 1.9 mA cm⁻², which are inferior to Ni₃(Ni₃∙HAHATN)₂. Thus, the above experimental data further confirm the unsaturated M1–N₂ site can serve as highly active center to remedy the shortcoming of traditional conductive MOFs toward electrocatalysis.

Figure 4.

a) HER polarization curves of Ni₃(HITP)₂ and Ni₃(Ni₃∙HAHATN)₂ samples. b) Polarization curves of the various M2₃(M1₃∙HAHATN)₂ samples and c) the corresponding Tafel plots. d) Electrocatalytic diagram of Ni₃(Ni₃∙HAHATN)₂ nanosheets toward HER.

Table 1.

Comparison of the HER activity for Ni₃(Ni₃∙HAHATN)₂ with other MOF‐ or MOF‐derived electrocatalysts

Catalyst	Electrolyte	Overpotential a) [mV]	Tafel plot [mV dec−1]	Ref.
Ni3(Ni3∙HAHATN)2	0.1 m KOH	115	45	This work
NiFe‐MOF/NF	1 m KOH	160	96	[40]
Ni‐ZIF@NF	1 m KOH	218	233	[41]
ZIF‐8 derived MoC	1 m KOH	182	60	[42]
MIL‐88A derived FeP	1 m KOH	95	72	[43]
HUST‐200@C	–	131	79	[44]
CoN2S2 MOFs	0.5 m H2SO4	283	71	[15]
Co‐BTSe	0.1 m KClO4	343	97	[45]
2D NiFe‐MOF/NF	1 m KOH	134	–	[35]
MOS 1	0.1 m KClO4	340	149	[46]
MOS 2	0.1 m KClO4	530	189	[46]
NENU‐500	0.5 m H2SO4	237	96	[47]
HKUST‐1	0.5 m H2SO4	691	127	[47]
GO/Cu‐MOF	0.5 m H2SO4	209 b)	95	[48]

^a)The overpotential at 10 mA cm⁻²

^b)The overpotential at 30 mA cm⁻².

Tafel slope is a key parameter to evaluate the HER kinetics. Tafel plots of M2₃(M1₃∙HAHATN)₂ and Ni₃(HITP)₂ samples were obtained by the Tafel equation according to their LSV curves (Figure 4c). Tafel slope of Ni₃(Ni₃∙HAHATN)₂ reaches up to 45.6 mV dec⁻¹, which is far less than that of Ni₃(HITP)₂ (94.2 mV dec⁻¹). Moreover, the Tafel slope values of Ni₃(Co₃∙HAHATN)₂, Ni₃(Cu₃∙HAHATN)₂, and Cu₃(Cu₃∙HAHATN)₂ are 98.2, 101.5, and 112.4 mV dec⁻¹, respectively. Among these conductive MOFs, Ni₃(Ni₃∙HAHATN)₂ exhibits the smallest Tafel slope, indicating the fastest kinetics toward HER electrocatalysis.

In general, the electrocatalytic activity relates to the conductivity of catalysts for effective electron transfer. PDOS calculation and four‐probe measurement demonstrate the rigid conjugated structure endows M2₃(M1₃∙HAHATN)₂ with an excellent electrical conductivity, which is beneficial for electron transfer during HER. Moreover, the SEM image exhibits the Ni₃(Ni₃∙HAHATN)₂ membrane on electrode still maintain layered structure with hierarchical porosity after fabrication process of the modified electrode (Figure S24, Supporting Information). Benefiting from the advantage of the unique porous morphology, the Ni₃(Ni₃∙HAHATN)₂ facilitate mass transfer during the electrocatalytic process (Figure 4d). Theoretically, only an unsaturated site can provide binding affinity to proton to achieve electron transfer. In this work, the M1–N₂ moiety of M2₃(M1₃∙HAHATN)₂ has a higher unsaturation degree in comparison with that of M–N₄ linkage. For instance, the PDOS results show the electronic orbits (s and p‐orbits) of Ni, coordinated by bidentate tertamine (Ni–N₂), are located beside the Fermi level in Ni₃(Ni₃∙HAHATN)₂ slab, but the location of electronic orbits of Ni in linkages (Ni–N₄) is far away (Figure 5a). Moreover, the d‐orbit of Ni atoms in Ni–N₂ moieties is also closer to Fimi level than that of the Ni atoms in Ni–N₄ linkages, which indicates the Ni–N₂ moieties have stronger absorption and bonding capacity for proton.^{[ 50 ]} Similar phenomenon occurs on the other M2₃(M1₃∙HAHATN)₂ slabs as well (Figures S25 and S26, Supporting Information). To further investigate the active site of M2₃(M1₃∙HAHATN)₂, the free energies of hydrogen adsorption (ΔG _H*) were calculated via DFT. According to thermodynamics and kinetics, the optimal ΔG _H* should be approached to thermoneutral 0 eV.^{[ 51 ]} The DFT simulation was also performed to calculate the values of ΔG _H* of various M2₃(M1₃∙HAHATN)₂ slabs (Figure 5b). In Ni₃(Ni₃∙HAHATN)₂, the Ni site of Ni–N₄ linkage expresses an exothermic ΔG _H* of +0.49 eV, which indicates that the Ni–N₄ linkage is hard to combine with protons. The adsorption energy of H* of Ni atom in Ni–N₂ moiety decreases to −0.12 eV, which is close to the optimal zero overpotential (ΔG _{H *} = 0 eV). This phenomenon illustrates the Ni–N₂ site need lesser energy to break the bond of Ni‐H to complete the catalytic process for HER. Similarly, the length of N—H bond in Ni–N₂ site is 1.603 Å, smaller than that in Ni–N₄ linkage (1.628 Å), suggesting the proton is easier to adsorb on the Ni site of Ni–N₂ moiety (Figure 5c). And hydrogen absorption in Ni–N₄ linkage makes the framework of Ni₃(Ni₃∙HAHATN)₂ bending, which breaks the rigid 2D structure. These results show the Ni–N₂ moieties in Ni₃(Ni₃∙HAHATN)₂ is more active for HER than Ni–N₄ linkages in conductive MOFs. Meantime, the ΔG _H* of M1–N₂ center in M2₃(M1₃∙HAHATN)₂ slabs also exhibits a similar regularity as the electrochemical results: Ni–N₂ in Ni₃(Ni₃∙HAHATN)₂ (−0.12 eV) > Co–N₂ in Ni₃(Co₃∙HAHATN)₂ (−0.29 eV) > Cu–N₂ in Ni₃(Cu₃∙HAHATN)₂ (−0.61 eV) > Cu–N₂ in Cu₃(Cu₃∙HAHATN)₂ (−0.64 eV). According to these results, M2₃(M1₃∙HAHATN)₂ is expected to be an ideal coordinated structure toward electrocatalysis, which proves our design concept of new conductive MOFs for electrocatalysis is feasible.

Figure 5.

a) PDOS of Ni atom in Ni–N₂ and Ni–N₄ of Ni₃(Ni₃∙HAHATN)₂ slab. b) Free‐energy profiles toward HER at various electrocatalytic sites. c) The hydrogen adsorption slabs of Ni₃(Ni₃∙HAHATN)₂ at Ni–N₂ and Ni–N₄ sites.

The durability is another parameter to evaluate the practicability of electrocatalysts. The HER stability of Ni₃(Ni₃∙HAHATN)₂ was primarily investigated by repeating CV tests in a 0.1 m KOH solution. After 1000 cycles, the overpotential of Ni₃(Ni₃∙HAHATN)₂ only displays a tiny deformation, relative to that of the initial scan (Figure S27, Supporting Information). The result indicates the Ni₃(Ni₃∙HAHATN)₂ sample possesses promising electrocatalytic stability for HER. Sequentially, the excellent durability of Ni₃(Ni₃∙HAHATN)₂ was further confirmed by chronoamperometry measurement at 10 mA cm⁻². After 10 h test, the HER current reveals a negligible attenuation, which maintains 83.4% of initial activity (Figure 6a). Meantime, the attenuation of Ni₃(Ni₃∙HAHATN)₂ still keep at a low level (78.6%) at a high current density of 50 mA cm⁻². On the other hand, the XRD pattern of Ni₃(Ni₃∙HAHATN)₂ maintains the particular crystal structure after electrochemical test (Figure S28a, Supporting Information). Similarly, the TEM and SEM characterizations show the electrocatalytic process cannot break the 2D structure of Ni₃(Ni₃∙HAHATN)₂ nanosheets (Figure 6b and Figure S28b, Supporting Information). The results reveal that the rigid conjugated coordinated structure makes Ni₃(Ni₃∙HAHATN)₂ possess excellent durability during HER process.

Figure 6.

a) Time‐dependent HER current density curves for Ni₃(Ni₃∙HAHATN)₂ sample at 10 and 50 mA cm⁻². b) TEM image of Ni₃(Ni₃∙HAHATN)₂ sample after chronoamperometry test.

3. Conclusion

To better exert the structural advantages of conductive MOFs, a type of new conductive M2₃(M1₃∙HAHATN)₂ MOFs with extra metallic sites (M1–N₂) was designed and prepared by employing a novel conjugated ligand. The larger ligand molecule endows M2₃(M1₃∙HAHATN)₂ with an expanded in‐plane porous structure (2.7 nm) for mass transfer. DFT calculations and four‐probe measurement clearly demonstrate the M2₃(M1₃∙HAHATN)₂ nanosheets possess excellent conductive ability for electron transfer. Importantly, the introduced metal atom in M1–N₂ moiety exhibits a higher unsaturated degree than that in traditional M2–N₄ linkage. The DFT calculations (PDOS and free energies of hydrogen adsorption) predict the major active center for HER in M2₃(M1₃∙HAHATN)₂ is M1–N₂ moiety rather than M2–N₄ linkage. Subsequently, a series of HER electrocatalytic tests confirm the prophecy of theory in experimental. The optimal Ni₃(Ni₃∙HAHATN)₂ nanosheets exhibit a remarkably activity enhancement for HER compared to traditional Ni₃(HITP)₂ conductive MOF. The rigid layered structure imparts Ni₃(Ni₃∙HAHATN)₂ nanosheets with an excellent durability. These characteristics make Ni₃(Ni₃∙HAHATN)₂ as a promising electrocatalyst for HER, and confirm our design concept of new conductive MOFs is meaningful for electrocatalysis.

4. Experimental Section

Reagents and Chemicals

1,2,4,5‐benzenetetramine tetrahydrochloride (Scheme S4a, Supporting Information), and hexaketocyclohexane octahydrate (Scheme S4b, Supporting Information) were purchased from Sigma‐Aldrich, USA. HITP ligand was obtained from TCI (Shanghai), China. Nickel(II) chloride, cobalt(II) chloride, cupric(II) chloride, and ammonium hydroxide were obtained from Macklin reagent company, China. Ether, ethanol and N,N‐dimethylformamide (DMF) were purchased from Beijing Chemical Reagent Company, China. All commercial chemicals were directly used without further purification.

Preparation of HAHATN

HAHATN was prepared by a microwave synthetic strategy, according to the procedure available in previous works.^{[ 48 ]} Briefly, 702 mg of 1,2,4,5‐benzenetetramine tetrahydrochloride (2.47 mmol) and 224 mg of hexaketocyclohexane octahydrate (0.718 mmol) were dispersed in 10 mL ethanol with concentrated hydrochloric acid (38%, 2 mL). The suspension was transferred to a microwave reaction vessel and reacted at 160 °C for 45 min. Then, the resulting solution was filtered to remove black byproduct and further concentrated under vacuum to dryness. The black HAHATN was obtained by recrystallizing in ethanol and washed twice with ether (225 mg, 45% yield).

Preparation of Ni₃∙HAHATN Ligand

Ni₃∙HAHATN ligand was prepared by a selective coordinated process. 162 mg of nickel chloride was dissolved in 30 mL ethanol, and the salt solution was adjusted pH by hydrochloric acid to 4. Then 100 mg of HAHATN was added into the acidic salt solution and refluxed for 4 h. The black Ni₃∙HAHATN ligand was obtained by recrystallizing in ethanol to remove unreacted nickel chloride.

Preparation of Co₃∙HAHATN Ligand

Co₃∙HAHATN ligand was prepared by a similar method. In this process, 162 mg of nickel chloride was replaced by 159 mg of cobalt chloride to coordinate with HAHATN. The yield of Co₃∙HAHATN ligand was 102 mg.

Preparation of Cu₃∙HAHATN Ligand

Cu₃∙HAHATN ligand was also prepared by a similar method. 162 mg of nickel chloride was replaced by 164 mg of cupric chloride to coordinate with HAHATN. The yield of Co₃∙HAHATN ligand was 95 mg.

Preparation of Ni₃(M1₃∙HAHATN)₂ Conductive MOFs

Ni₃(M1₃∙HAHATN)₂ was obtained by a coordinated reaction in alkaline. 51 mg of nickel chloride was dissolved in 42 mL deionized water with ammonium hydroxide (2.5 mL), and the mixed solution was stirred at 65 °C under air atmosphere. Then, 100 mg of M1₃∙HAHATN ligand was added into the solution with continuous stirring for 2 h. The product was purified by washing with ethanol and deionized water.

Preparation of Cu₃(Cu₃∙HAHATN)₂ Conductive MOFs

Cu₃(Cu₃∙HAHATN)₂ was obtained by a similar method, 51 mg of nickel chloride was replaced by 53 mg of cupric chloride to coordinate with Cu₃∙HAHATN ligand.

Preparation of Ni₃(HITP)₂ Conductive MOFs

Ni₃(HITP)₂ conductive MOFs was synthesized according to the previous report.^{[ 26 ]} 6.6 mg of nickel chloride was dissolved in 5 mL of water and 0.3 mL of concentrated aqueous ammonia. 10 mg of HITP solution (5 mL of water) was added into the above‐mentioned salt solution. This mixture was stirred at 65 °C for 3 h. The resulting black powder was centrifuged, filtered, and then washed by ethanol and deionized water. Then, the solid was dried under vacuum at 150 °C.

Characterization

NMR spectra were obtained by 400 MHz Bruker ASCEND NMR Spectrometer. D/Max‐3cX′Pert was employed to record the XRD patterns. The SEM images and EDX mapping were taken by a HITACHI SU8020 field‐emission electron microscope. TEM images were made on a TECNAI F20 transmission electron microscope. PHI‐5000 X‐ray photoelectron spectrometer was used for XPS measurement. Nitrogen sorption experiments were performed with a nitrogen physical adsorption instrument (ASAP 2400). Before testing, Ni₃(Ni₃∙HAHATN)₂ sample was firstly degassed at 150 °C for 10 h.

Preparation of Electrode

In this work, rotating disk glassy carbon electrode (RDE) with 3 mm in diameter was employed as the working electrode. Before preparing the modified electrode, the working electrode was firstly treated by pre‐polishing to remove contaminant on the surface of the electrode. On the other side, 8 mg of various M2₃(M1₃∙HAHATN)₂ samples were dispersed in 2 mL mixed solvent (V _water:V _DMF = 1:1) and ultrasonicated for 10 min to achieve a uniform ink. 10 µL of well‐dispersed sample ink was dropped onto the pre‐treated RDE and dried in vacuum. Then, 5 µL of Nafion ethanol solution was used to prevent electrocatalyst from falling off.

Electrochemical Measurement

The HER electrochemical measurements were performed on in CHI 760E electrochemical workstation with a three‐electrode system. The counter electrode and reference electrode were employed by graphite and saturated calomel electrode, respectively. The electrochemical measurements were carried out in N₂‐saturated 0.1 m KOH electrolyte and all potentials in this work were transformed to RHE via the following equation

$$E_{\text{RHE}}=E_{SCE}+0.243\text{V}+0.059\times \text{pH}$$ (1)

LSV measurements were performed at 1600 rpm in 5 mV s⁻¹ with 95% iR corrections to evaluate their activities.

DFT Calculation

DFT simulation of various M2₃(M1₃∙HAHATN)₂ slabs were calculated by CASTEP and DMol³ modules. In these slabs, a vacuum layer of 30 Å was established to avoid periodic interaction. And, PDOS calculation was performed by GGA+U functional with additional Coulomb potential (U _Ni = 3.1, U _Co = 3.1 and U _Cu = 3.0 eV) for 3d‐orbit. A plane‐wave energy cut off of 400 eV was used together with norm‐conserving pseudopotentials, and the Brillouin zone was sampled with a 2 × 2 × 1 Monkhorst‐Pack grid. The structure was fully optimized until the force on each atom was less than 10⁻³ eV Å⁻¹. On the other side, the free energy (∆G) was computed from

$$\mathrm{\Delta }G=\mathrm{\Delta }E+\text{ZPE}-T\mathrm{\Delta }S$$ (2)

where ∆E means the total energy, ZPE was the zero‐point energy, the entropy (∆S) of each adsorbed state were obtained from DFT calculation, whereas the thermodynamic corrections for gas molecules were from standard tables.

Conflict of Interest

The authors declare no conflict of interest.

Supporting information

Supporting Information

Huang H., Zhao Y., Bai Y. M., Li F. M., Zhang Y., Chen Y., Conductive Metal–Organic Frameworks with Extra Metallic Sites as an Efficient Electrocatalyst for the Hydrogen Evolution Reaction. Adv. Sci. 2020, 7, 2000012 10.1002/advs.202000012