The Dynamic Formation from Metal‐Organic Frameworks of High‐Density Platinum Single‐Atom Catalysts with Metal‐Metal Interactions

Abstract Single‐atom catalysts (SACs) hold great promise for highly efficient heterogeneous catalysis, yet the practical applications require the development of high‐density active sites with flexible geometric structures. The lack of understanding in the dynamic formation process of single atoms in the host framework has been plaguing the controllable synthesis of next generation SACs. Here using Co‐based metal‐organic frameworks (MOFs) as a starting substrate, we fully elucidated the formation of high‐density Pt single atoms with inter‐site interactions in derived Co3O4 host. The cation exchange process and dynamic evolution of Pt−Pt interactions, organic ligand cleavage and Pt‐oxygen coordination formation during the pyrolysis process have been unambiguously interpreted by a series of in situ/ex situ spectroscopic measurements and theoretical computation. These findings would direct the synthesis of high‐density SACs with metal‐metal interactions, which demonstrate significantly enhanced structural flexibility and catalytic properties.

reported previously. [1] In a typical procedure, 580 mg of cobalt (II) nitrate hexahydrate (Co(NO3)2•6H2O) was dissolved in 20 mL of deionized (DI) water containing 30 mg of hexadecyltrimethylammonium bromide (CTAB). Then, this solution was injected into 140 mL of aqueous solution with 9.08 g of 2-methylimidazole and stirred at room temperature for 60 min. The purple precipitate was collected by centrifugation and washed with ethanol five times.
Co-BDC was synthesized according to previous literature. [2] In a typical procedure, 124.6 mg of Benzenedicarboxylic acid (BDC) was dissolved into a mixture of 32 mL of N, Ndimethylformamide (DMF), 2 mL of ethanol and 2 mL of DI water. Next, 218.3 mg of Co(NO3)2•6H2O was dissolved in the solution. Then, 0.8 mL Triethylamine (TEA) was quickly added in the mixture, a uniform colloidal suspension can be obtained after stirring for 5 mins.
The colloidal solution was ultrasonicated overnight before centrifuged and washed with ethanol for 5 times.
In a cation exchange process, 100 mg of ZIF-67 nanocrystals were dispersed in 50 mL of DI water, then 9.5 mg (2 mmol L -1 ) of potassium hexachloroplatinate (IV) hydrate (K2PtCl6·xH2O) were dissolved in 10 mL of DI water and added into the ZIF-67 solution slowly under stirring conditions. The reaction was quenched by centrifuging the suspension after 3 hours, the Pt-ZIF-67 precipitate was collected and washed twice with DI water and three times with ethanol.
The Pt-ZIF-67-300 o C nanocrystals were obtained after drying the precipitate at 60 o C overnight in a vacuum oven and pyrolyzed at 300 o C in air for 4 hours. Au-ZIF-67 was achieved by a similar procedure with applying 2 mmol L -1 of gold (III) chloride hydrate (HAuCl4·xH2O) aqueous solution in cation exchange process. Pt-Co-BDC-300 o C was prepared by similar cation exchange process with 10 mL of K2PtCl6·xH2O (2 mmol L -1 ) aqueous solution and 50 mL of Co-BDC solution (2 mg mL -1 ), followed by centrifuged and washed with ethanol and pyrolyzed in air at 300 o C for 4 hours.

Materials characterization. XRD data was collected on a Rigaku MiniFlex 600 X-Ray
Diffractometer. The soft XAS measurements were performed at the SXR beamline in Australian Synchrotron Radiation Facility, Melbourne. Ex-situ characterizations were conducted on samples pyrolyzed at different temperatures from 100 to 300 o C. The holding time was 15 mins at each temperature in the range of 100-275 o C and 10/60/90 mins for 300 o C. ICP-MS analysis was conducted using an Agilent 7500cx instrument with attached laser ablation system. HAADF-STEM images were recorded by using a FEI Titan G2 80-300 microscope at 300 kV equipped with a probe corrector. Projected Z 2 -map simulations were performed by using the qSTEM program. [3] DRIFTs measurements. DRIFTs measurements were performed using a Nicolet iS-50 In-situ XAS measurements. The Pt L3-edge XAS measurements were performed at the beamline of 4B9A beamline in Beijing Synchrotron Radiation Facility. The X-ray was monochromatized by a double-crystal Si (111) monochromator. The incident and transmitted X-ray intensities were monitored by using standard ion chambers, and the monochromator was detuned to reject higher harmonics. In-situ XAS measurements were conducted with a home-made furnace at a heating rate of 2 o C/min and held at 100, 150, 200, 250 o C for 15 mins and 300 o C for 90 mins. While the XAS raw data were background subtracted, normalized, and Fourier transformed by standard procedures within the ATHENA program, the least-squares curve fitting analysis of the EXAFS data was carried out using the ARTEMIS program. [4] The Pt L3-edge theoretical XANES calculations were carried out with the FDMNES code in the framework of real-space full multiple-scattering scheme. Muffin-tin approximation for the potential was used. [5] Satisfactory convergence for the cluster size had been achieved.

DFT calculations. DFT calculations were carried out using the Vienna ab-initio Simulation
Package (VASP). [6] The exchange-correlation interaction was described by generalized gradient approximation (GGA) with the Perdew-Burke-Ernzerhof (PBE) functional. [7] The projector augmented wave (PAW) pseudopotential scheme is used and the force and energy convergence tolerance values with respect to plane-wave cutoff and K-point density were set to be 0.01 eV Å -1 and 10 -5 eV, respectively. [8] Free energies and formation energies of all the compounds were obtained using the GGA-DFT plus Hubbard-U framework (GGA+U).
Ferromagnetic spin-polarized calculations were employed for all magnetic materials. The GGA+U calculations were performed using the model proposed based on the TEM images with the Ueff (Ueff = Coulomb U -exchange J) values of 4.4, 6.7 and 3.2 eV for Co 2+ , Co 3+ and Pt, respectively. [9] Co2(2- The reaction free energy of this reaction can be represented by ∆G reaction = G2-MeIm + GCo2(2- Among the reactants and products, the Co2(2-MeIm)7, the Co2(2-MeIm)4O and the 2-MeIm can be treated as solid state. Here, we assign the enthalpy and entropy as H = E DFT and S = 0 (at moderate temperatures). [10] Thus, the Gibbs free energy (Gi(s)) of a solid element i can be represented by DFT calculations as Gi(s) = E DFT i(s). The GO2(g) at the standard state can be presented as 2μO(g) = GO2(g) = EDFT O2(g) + EZPE O2(g) + RT -TSexp O2(g) ,where Sexp O2(g) = 205.152 J mol -1 K -1 is taken from the literature. [11] Here, EDFT O2(g) = -8.746 eV and EZPE O2(g) = 0.105 eV can be obtained by DFT calculations. KOH electrolyte with a CHI potentiostat (CHI 760E) at a rotating speed of 1600 rpm. The polarization curves of the catalysts were obtained with scan rates of 5 mV s -1 . The ECSAs of the catalysts were determined by CO stripping voltammograms. [12] The electrolyte was bubbled with 20 % CO in argon for 30 min with the working electrode held at 0.2 V vs. RHE, followed by argon purging for 20 min to remove the excess CO. Then the CO stripping voltammograms were collected in the potential range from around 0 V to 1.2 V vs. RHE during Ar bubbling.
The ECSAs values were calculated based on the following equation [13] : where QCO_strip is the integrated charge obtained from CO-stripping curves, the Q theo CO_strip is the theoretical value for a two-electron transfer assuming the oxidation of one CO to CO2 per Pt atom.