Open Access Article
Xinxin
Wang
,
Yucheng
Jin
,
Tianyu
Zheng
,
Ning
Li
,
Yuesheng
Han
,
Baoqiu
Yu
,
Kang
Wang
*,
Dongdong
Qi
*,
Tianyu
Wang
* and
Jianzhuang
Jiang
*
Beijing Advanced Innovation Center for Materials Genome Engineering, Beijing Key Laboratory for Science and Application of Functional Molecular and Crystalline Materials, Department of Chemistry and Chemical Engineering, School of Chemistry and Biological Engineering, University of Science and Technology Beijing, Beijing 100083, China. E-mail: kangwang@ustb.edu.cn; qdd@ustb.edu.cn; twang@ustb.edu.cn; jianzhuang@ustb.edu.cn
First published on 16th April 2024
Fabricating three dimensional (3D) supramolecular frameworks (SMFs) into stable crystalline nanosheets remains a great challenge due to the homogeneous and weak inter-building block interactions along 3D directions. Herein, crystalline nanosheets of a 3D SMF with a uniform thickness of 4.8 ± 0.1 nm immobilized with Pt nanocrystals on the surface (Q[8]/Pt NSs) were fabricated via the solid–liquid reaction between cucurbit[8]uril/H2PtCl6 single crystals and hydrazine hydrate with the help of gas and heat yielded during the reaction process. A series of experiments and theoretical calculations reveal the ultrahigh stability of Q[8]/Pt NSs due to the high density hydrogen bonding interaction among neighboring Q[8] molecules. This in turn endows Q[8]/Pt NSs with excellent photocatalytic and continuous thermocatalytic CO oxidation performance, representing the thus-far reported best Pt nano-material-based catalysts.
Herein, we describe the fabrication of crystalline nanosheets of a 3D cucurbit[8]uril (Q[8]) SMF via direct chemical synthesis. A solid–liquid reaction between the Q[8]/H2PtCl6 single crystals dispersed in water and liquid hydrazine hydrate affords ultrathin supramolecular nanosheets of Q[8] decorated with platinum particles (Q[8]/Pt NSs) with a uniform thickness of 4.8 ± 0.1 nm and ultrastability, according to the transmission electron microscopy (TEM) and atomic force microscopy (AFM) images. Powder X-ray diffraction (PXRD) analysis discloses the crystalline nature of the Q[8] nanosheets and Pt nanocrystals. Experimental investigations in combination with theoretical calculations reveal the important role of explosive gases (N2 and HCl) and heat yielded during the chemical reaction process in the formation of the nanosheets of 3D SMF and the high density hydrogen bonding interaction in stabilizing the Q[8]/Pt NSs. A series of spectroscopic technologies including Fourier transform infrared (FT-IR), X-ray absorption structure (XAS), and X-ray photoelectron spectroscopy (XPS) unveil the immobilization of Pt nanocrystals onto the surface of Q[8] nanosheets through O–Pt coordinating interactions, endowing the immobilized Pt nanocrystals with significantly enhanced stability. This in turn results in an excellent photocatalytic and, in particular, continuous thermocatalytic CO oxidation performance of Q[8]/Pt NSs with high CO-to-CO2 conversion efficiency and good durability, representing the thus-far reported best Pt nano-material-based catalysts.
C hydrogen bonds. The portal carbonyls of the central Q[8] are close to the outer surface of two neighbouring Q[8] molecules, while the outer surface of the central Q[8] is close to the portal carbonyls of the other two neighbouring Q[8] molecules. This reciprocal interaction leads to a quasi-Td symmetry for the unit formed by the five Q[8] molecules, which further extends into a 3D hydrogen-bonded network, Fig. 1. It is worth noting that the multiple hydrogen-bonding interactions existing between neighbouring Q[8] molecules endow this Q[8] SMF with comparable stability to 3D MOFs and 3D COFs, as proved by its unchanged crystal structure after immersing the Q[8] single crystal into various organic solvents (e.g., DMF and DMSO), boiling water, H2O2, strong acid (6.0 M HCl), and strong base (10.0 M NaOH) for 24 h, Fig. S1.† Such strong and homogeneous inter-building block interactions, however, also result in the failure in exfoliating the Q[8] single crystal into nanosheets via conventional strategies including sonication and mechanical delamination, Fig. S2.† Herein, the reaction between the block-shaped Q[8]/H2PtCl6 single crystals with a similar arrangement of Q[8] molecules dispersed in water with liquid N2H4·H2O at 50 °C for 1 hour yields ordered square nanosheets, Q[8]/Pt NSs, as indicated by the SEM images, Fig. 2a, S3 and S4.† In addition, the reaction time and concentration of N2H4·H2O have been varied to optimize the fabrication process of Q[8]/Pt NSs, Fig. S5–S7.† In the case of Q[8]/Pt NSs, the nanosheets appear after reaction for 10 minutes at 50 °C. Along with further increase in the reaction time, the amount of nanosheets increases, accompanied by the gradual disappearance of the block-shaped crystals of Q[8]/H2PtCl6. After reaction for 1 hour, only nanosheets can be observed from TEM and SEM images, revealing the complete conversion of Q[8]/H2PtCl6 to Q[8]/Pt NSs. Moreover, the nanosheets gradually become thinner with the increase in the N2H4·H2O concentration to 12 mmol L−1, Fig. S7.† Further increase in the N2H4·H2O concentration leads to the decrease in the amount of nanosheets accompanied by the gradual increase in the block-shaped crystals of Q[8]/Pt. As a result, Q[8]/Pt NSs prepared from Q[8]/H2PtCl6 with 12 mmol per L N2H4·H2O at 50 °C for 1 hour were therefore used for the detailed investigation in the present work. In particular, the optimized fabrication process of Q[8]/Pt NSs has been carried out for three times, giving almost the same PXRD patterns and almost the same morphology as revealed by the TEM and SEM images, Fig. S8 and S9,† demonstrating the excellent reproducibility of the preparation method. Moreover, the large-scale preparation of Q[8]/Pt NSs can be easily achieved by using the present method, Fig. S10,† fulfilling the requirement for large-scale production. AFM measurements prove the smooth surface of Q[8]/Pt NSs with a uniform thickness of ca. 4.8 ± 0.1 nm, corresponding to the thickness of six Q[8] layers along the b direction in the Q[8] single crystal, Fig. 2b and c. TEM image reveals the transparent square-like morphology of Q[8]/Pt NSs with the edge length of several micrometers, Fig. 3a, confirming their ultrathin nanosheet nature. Further evidence for this point comes from the clear observation of the Tyndall effect when a red laser is passed through the dispersed ethanol solution of Q[8]/Pt NSs, inset in Fig. 3a. The highly ordered periodic structure of Q[8]/Pt NSs was characterized by high-resolution TEM (HR-TEM) and fast Fourier transform (FFT) images, Fig. 3b and c. HR-TEM demonstrates a clear quadrilateral network structure with a pore size of ca. 0.38 nm, consistent with the calculated channel cavity size of ca. 0.37–0.41 nm (ca. 0.22–0.26 nm after deducting the van der Waals radii of the atoms on the channel surface) for Q[8]/Pt NSs along the b axis on the basis of the single crystal structure of Q[8], Fig. S11 and S12.† Moreover, the Q[8] units can be identified as bright spots in the HR-TEM image, while the channel of Q[8]/Pt NSs is identified as slightly darker spots. The rectangle-shaped dark areas adjacent to the bright spots and the dimmer spots correspond to the curved channel in the Q[8]/Pt NSs, Fig. 3d and e, revealing the same arrangement of Q[8] molecules in the nanosheets with that in Q[8] single crystals. In particular, the selected area FFT image shows a highly ordered arrangement with independent diffraction points, confirming the high crystallinity of Q[8]/Pt NSs, inset in Fig. 3c. In addition, HR-TEM images unveil the widely dispersed Pt nanoparticles with a size of 2–10 nm on Q[8]/Pt NSs, revealing the reduction of H2PtCl6 into Pt0, Fig. S13.† It is worth noting that the size of the Pt nanoparticles is significantly larger than the cavity size of the Q[8] molecule, indicating the location of Pt nanoparticles on the surface of Q[8]/Pt NSs. The pore size distribution of Q[8]/Pt NSs is calculated to be centred at ca. 0.3 nm on the basis of small-angle X-ray scattering (SAXS) analysis, consistent with the result according to the single crystal data of Q[8], Fig. S14.† Energy-dispersive X-ray spectroscopy (EDS) elemental mapping displays the homogeneous distribution of C and N atoms within Q[8]/Pt NSs, Fig. 3f. Inductively coupled plasma optical emission spectrometry analysis indicates 4.8% content of Pt in Q[8]/Pt NSs.
The crystalline structure of Q[8]/Pt NSs was further explored by PXRD. As shown in Fig. 3g and i, the PXRD pattern of Q[8]/Pt NSs displays a series of strong peaks at 8.1, 8.9, 9.2, 10.2, 12.0, 12.5, 13.5, 14.0, 15.3, 16.2, and 18.4°, which match well with the (211), (220), (112), (301), (321), (400), (411), (420), (303), (422), and (224) planes of the pattern of Q[8] single crystals, indicating the same arrangement of Q[8] molecules in the nanosheets as that in Q[8] single crystals, in line with the HR-TEM results. This is further confirmed by the Pawley refinement of the PXRD pattern with small agreement factors of Rp = 4.24% and Rwp = 7.64%, Fig. S15.† Nevertheless, the PXRD pattern of Q[8]/Pt NSs also shows four peaks at 38.4, 44.0, 69.4, and 78.4°, corresponding to the (111), (200), (220), and (311) planes of hcp Pt (PDF card no. 06-0663), disclosing the crystalline nature of the Pt nanoparticles, Fig. 3h and i. In addition, thermogravimetric analysis indicates that Q[8]/Pt NSs can be thermally stable up to 320 °C in air, Fig. S16.† Nevertheless, after being exposed to various organic solvents (e.g., DMF and DMSO), boiling water, H2O2, strong acid (6.0 M HCl), and strong base (10.0 M NaOH) for 24 h, Q[8]/Pt NSs still exhibit unchanged PXRD patterns, Fig. S17,† demonstrating their excellent chemical stability owing to the multiple hydrogen-bonding interactions between neighbouring Q[8] molecules.
In the Pt 4f XPS spectrum, Q[8]/Pt NSs display two major peaks at 70.3 and 73.7 eV attributed to Pt0 together with two weak peaks at 71.6 and 74.9 eV due to Pt2+,48–50 Fig. S18.† The Pt0/Pt2+ ratio amounts to 81
:
19 in Q[8]/Pt NSs, indicating the highly abundant zero-valence Pt in the nanocrystals. In the O 1s XPS spectrum of Q[8]/Pt NSs, in addition to a peak at 531.3 eV owing to C
O in Q[8],50 a sign ascribed to C–O–Pt is also observed at 533.0 eV,51 revealing the strong interaction between Pt nanocrystals and Q[8] in Q[8]/Pt NSs, Fig. S19.† Furthermore, a typical absorption peak due to the carbonyl band of Q[8] was observed at 1734 cm−1 in the FT-IR spectrum of pure Q[8], which shifts to 1724 cm−1 for Q[8]/Pt NSs, Fig. S20,† proving the effective bonding of Pt nanocrystals with the portal carbonyl groups of Q[8] in Q[8]/Pt NSs.52 In addition, in the Pt L3-edge X-ray absorption near-edge structure (XANES) spectrum, Fig. S21,† the energy position and spectral features of Q[8]/Pt NSs are similar to those for Pt foil, confirming the mainly zero valence metal state of Pt in Q[8]/Pt NSs. Moreover, the Pt L3 extended X-ray absorption fine structure (EXAFS) spectrum in the R space of Q[8]/Pt NSs shows the peaks at 2.4 and 1.8 Å, Fig. S22,† which are indexed as the Pt–Pt interaction and Pt–O coordination, respectively.53–55 These results further demonstrate the domination of metallic Pt species and the existence of coordination interaction between Pt nanocrystals and Q[8] molecules in Q[8]/Pt NSs.
To understand the formation mechanism of the Q[8]/Pt NSs, both experimental and theoretical investigations have been carried out. According to eqn (1), drastic reduction of H2PtCl6 revealing the strong penetration not only produces the Q[8] crystalline layer in the same depth, but simultaneously yields a large amount of N2 and HCl gases and releases a large amount of heat.
| H2PtCl6 + N2H4·H2O → Pt + N2↑ + H2O + 6HCl↑, ΔrHm = −127 kJ mol−1 | (1) |
Combination of the gas-releasing explosive effect with the simultaneously yielding heat in turn leads to the successful exfoliation of the Q[8] crystalline nanosheets. This process proceeds in a repeated manner until the complete reduction of Q[8]/H2PtCl6 single crystals. Additional support for this point comes from the failed fabrication of the crystalline nanosheets starting from the same Q[8]/H2[PtCl6] single crystals using weak reducing agents like methanol and glycol without yielding gas during the reduction reaction process, which however induces the isolation of Pt nanocrystal-immobilized Q[8] single crystals adopting the unchanged bulk morphology to the starting Q[8]/H2PtCl6 single crystals, Fig. S23–S26.† Actually, N2 gas and heat release measurement result gives further support for the above-mentioned gas-releasing explosive effect. As shown in Fig. S27,† N2 gas was detected as the sole gas product by GC spectroscopy and bright-field microscopy images since HCl yielded would dissolve in the aqueous reaction system once it leaves from the single crystals. In addition, as can be found, at the very beginning step of the reaction, the amount of N2 gas increases in a gradual manner. However, after the reaction proceeds for 10 min, the amount of N2 gas suddenly increases, suggesting the yielding process of the gas in a microscopic explosive manner at this time scale with a large amount of nanosheets formed, Fig. S28.† The temperature change in the reaction system was revealed by the thermal imager. As shown in Fig. S29,† the temperature increases from 50.1 to 52.6 °C after the reaction between 10.0 mg Q[8]/H2PtCl6 single crystals and 0.02 mL N2H4·H2O for 10 min. On the basis of the above analysis, a two-step fabrication process for the nanosheets, Q[8]/Pt NSs, was proposed, Fig. S30 and S31:† (1) the reduction process from PtIV to Pt0 together with the formation of Pt nanocrystals and (2) the formation of the crystalline ultrathin nanosheets of 3D SMF together with the formation of more Pt nanocrystals.
CO oxidation was selected as the probe reaction to investigate the catalytic performance of the Q[8]/Pt NSs due to the critical role of this reaction in cleaning air and fuel-cell applications.55 Before the catalysis investigation, the free-standing flexible films of the Q[8]/Pt NSs, named Q[8]/Pt NS film-10/20/30/40, were fabricated through filtering 10/20/30/40 mg of Q[8]/Pt NS suspension in ethanol on the filter membrane, respectively, Fig. 4a and b. The cross-sectional SEM images of each membrane sample show a uniform thickness of 10, 20, 30, and 40 μm, respectively, Fig. 4c and S32.† For comparison, the film of 20 mg Q[8]/Pt material isolated from reducing Q[8]/H2PtCl6 using CH3OH was also fabricated in the same manner, affording the CH3OH-reduced Q[8]/Pt film-20.
The visible-light-driven photocatalytic CO oxidation performance of these films was investigated using an idealized gas mixture (0.8% CO, 20% O2, and He balance) at 25 °C. GC spectroscopy was used to quantify the gaseous product. As shown in Fig. 4d and S33–S35,† CO could be oxidized to CO2 under the photocatalysis of the Q[8]/Pt NS films with a CO2 selectivity of 100%. The CO-to-CO2 conversion rates of the Q[8]/Pt NS films with different thicknesses amount to 25.9, 28.3, 27.1, and 24.6 mmolCO gPt−1 h−1, respectively, Fig. 4e, corresponding to a turnover frequency (TOF) of 132, 145, 139, and 126 h−1. In contrast, the CH3OH-reduced Q[8]/Pt film-20 shows a CO conversion rate of 0.13 mmolCO gPt−1 h−1 with a TOF of 6.5 h−1, Fig. S36,† much inferior to the Q[8]/Pt NS film-20. In particular, the result of the Q[8]/Pt NS films is superior to that of the commercial P25 TiO2, and actually represents one of the best visible-light-driven photocatalytic CO oxidation performances in terms of catalytic efficiency, Table S1,† revealing the significant advantage of the Q[8]/Pt NSs in the heterogeneous photocatalysis. It is worth noting that no CO2 was detected when N2 instead of CO was used as the starting material, Fig. S37,† unveiling the origin of CO2 from the CO molecules in the photosynthesis system of the Q[8]/Pt NS films. Besides, the Q[8] support itself was essentially inactive for the photocatalytic CO oxidation reaction under the same reaction conditions, disclosing the essential role of the Pt nanocrystals immobilized at the surface of Q[8] nanosheets in the high activity of the Q[8]/Pt NS films. In addition, the performance of the Q[8]/Pt NS film-20 shows only a negligible drop after five consecutive runs with each run lasting for 40 min under light irradiation, indicating the excellent photostability of the Q[8]/Pt NS films, Fig. 4f and Table S1.† This is further supported by the unchanged morphology, XPS spectrum, and XRD pattern of Q[8]/Pt NS film-20 after the stability tests, Fig. S38–S41.†
Towards further extending the application potential of the present Q[8]/Pt NSs, the continuous thermocatalytic CO oxidation performance of the Q[8]/Pt NS films was also evaluated by a dynamic test with the feed gas (0.8% CO, 20% O2, and He balance) passing through the Q[8]/Pt NS films in a fixed-bed reactor at a flow rate of 25 mL min−1. As displayed in Fig. 4g, the Q[8]/Pt NS film-10/20/30/40 can achieve the CO oxidation at 40 °C with a specific reaction rate of 396, 540, 576, and 504 mmolCO gPt−1 h−1, respectively, and a CO2 selectivity of 100%. Along with the increase in temperature, the CO conversion rate increases. At the typical working temperature for polymer electrolyte membrane fuel cells of 80 °C, the CO-to-CO2 conversion rate of the Q[8]/Pt NS film-10/20/30/40 reaches 9026, 9284, 9437, and 9330 mmolCO gPt−1 h−1, respectively, corresponding to a TOF of 1760, 1810, 1840, and 1819 h−1, Fig. 4h and S42.† In good contrast, no CO2 was detected for the CH3OH-reduced Q[8]/Pt film-40 under the same conditions, Fig. S43.† In particular, the catalytic efficiency of the Q[8]/Pt NS films for CO oxidation is 1–2 orders of magnitude higher than that of the thus-far reported best supported Pt catalysts and is also much superior to the standard gold nanocatalyst Au/Fe2O3, Table S2,† indicating the outstanding performance of the Q[8]/Pt NS films for continuous CO oxidation. Moreover, the CO conversion and the CO2 selectivity over the Q[8]/Pt NS film-40 remain at constant values over a 1000 minute run under a space velocity of 1.0 × 106 mL gPt−1 h−1 at 80 °C, Fig. 4i. Further studies reveal that Q[8]/Pt NS film-40 remains stable for 300 minutes even at a 5 times higher space velocity of 4.2 × 107 mL gPt−1 h−1. Nevertheless, no obvious change could be observed from the TEM, SEM, XRD, and XPS data of the Q[8]/Pt NS films after the stability tests, Fig. S44–S47,† demonstrating the excellent durability of the Q[8]/Pt NS films, compared to the reported state-of-the-art Pt nano-material-based catalysts, Table S2.†
To get an insight into the catalysis mechanism, the adsorption behavior of CO on the catalyst surface was investigated by in situ DRIFTS spectroscopy. Fig. 5a and b show the visible light-driven and thermo-driven in situ DRIFTS spectra, respectively, of Q[8]/Pt NS films under an atmosphere of 0.8% CO, 20% O2, and remaining He as a function of time. At the beginning of the visible light-driven reaction, the band due to the Pt–CO bond appeared at 2064 cm−1, Fig. 5a, indicating the strong CO adsorption of the Pt nanocrystals.56,57 In addition, the band due to the C–O stretching vibration of CO2 at 2349 cm−1 gradually increases, revealing the CO-to-CO2 conversion under the catalysis of Q[8]/Pt NSs.58,59 This is also true for the thermo-driven reaction, Fig. 5b and S48a,† in line with the DFT calculation results. To further clarify the CO → CO2 reaction mechanism, DFT calculation was carried out using the optimized Q[8]/Pt NS model, in which Pt nanoparticles are located in the shallow pits at the surface of Q[8] NSs and constructed by three Q[8] molecules, Fig. S48a.† According to the calculation results from this model, when one O2 molecule is absorbed onto two adjacent Pt atoms, the O–O bond is activated along with the formation of a transition state [*O⋯O⋯*C
O] with the potential barrier of 1.17 eV, Fig. S48b,† which could be easily driven by the visible light irradiation or thermal vibration at ca. 80 °C. This transition state then gets split into O* and CO2 with a significant Gibbs free energy decrease as high as 2.49 eV. Along with the leaving of this newly generated CO2, another CO molecule is attracted onto the same Pt catalytic site and oxidized to CO2 by the O* with an energy decrease of 1.02 eV, finally completing one catalytic cycle.
Footnote |
| † Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4sc00656a |
| This journal is © The Royal Society of Chemistry 2024 |