Emerson C.
Kohlrausch
ab,
Higor Andrade
Centurion
c,
Rhys W.
Lodge
a,
Xuanli
Luo
d,
Thomas
Slater
e,
Marcos J. L.
Santos
b,
Sanliang
Ling
d,
Valmor R.
Mastelaro
c,
Matthew J.
Cliffe
a,
Renato Vitalino
Goncalves
c and
Jesum
Alves Fernandes
*a
aSchool of Chemistry, University of Nottingham, Nottingham, NG7 2RD, UK. E-mail: jesum.alvesfernandes@nottingham.ac.uk
bPrograma de Pós-Graduação em Ciência dos Materiais, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS 91501-970, Brazil
cSão Carlos Institute of Physics, University of São Paulo, São Carlos, SP 13560-970, Brazil
dAdvanced Materials Research Group, Faculty of Engineering, University of Nottingham, Nottingham, NG7 2RD, UK
eElectron Physical Sciences Imaging Centre, Diamond Light Source Ltd., Oxfordshire OX11 0DE, UK
First published on 22nd November 2021
Atomically-dispersed metal catalysts (ADMCs) on surfaces have demonstrated high activity and selectivity in many catalytic reactions. However, dispersing and stabilising individual atoms in support materials in an atom/energy-efficient scalable way still presents a significant challenge. Currently, the synthesis of ADMCs involves many steps and further filtration procedures, creating a substantial hurdle to their production at industrial scale. In this work, we develop a new pathway for producing ADMCs in which Pt atoms are stabilised in the nitrogen-interstices of a graphitic carbon nitride (g-C3N4) framework using scalable, solvent-free, one-pot magnetron sputtering deposition. Our approach has the highest reported rate of ADMC production of 4.8 mg h−1 and generates no chemical waste. Deposition of only 0.5 weight percent of Pt onto g-C3N4 led to improved hydrogen production by factor of ca. 3333 ± 450 when compared to bare g-C3N4. PL analysis showed that the deposition of Pt atoms onto g-C3N4 suppressed the charge carrier recombination from the photogenerated electron–hole pairs of Pt/g-C3N4 thereby enhance hydrogen evolution. Scanning transmission electron microscope imaging before and after the hydrogen evolution reaction revealed that the Pt atoms stabilised in g-C3N4 have a high stability, with no agglomeration observed. Herein, it is shown that this scalable and clean approach can produce effective ADMCs with no further synthetic steps required, and that they can be readily used for catalytic reactions.
Magnetron sputtering has recently emerged as a promising technique for the production of metal nanoparticles (MNPs) on a wide variety of supports (e.g. powder, liquids).17–22 It is one of the select few ‘green’ and scalable top-down methods,17 and has already been applied on a large scale in the glass-coating and semiconductor industries. Crucially, this approach is carried out in ultraclean, high vacuum environments and can therefore generate extremely active metal species with clean surfaces not occluded by ligands or surfactants.17,18,23 In the magnetron sputtering process, accelerated argon (Ar) ions collide elastically with a highly pure metal target, which expels atoms onto a support material (Fig. 1a). Gohara & Yamazaki recently demonstrated that single Pt atoms can be dispersed and stabilised onto a single layer graphene deposited onto TEM grids using magnetron sputtering deposition.24 However, the deposition of ADMCs into bulk powder samples, suitable for use at scale and thus in catalysis, has not been demonstrated.
Herein we show, for the first time, the deposition of ADMCs into bulk powder using magnetron sputtering. We stabilised Pt atoms into the nitrogen-interstices of graphitic carbon nitride (g-C3N4). Furthermore, our approach showed the highest rate of ADMC production reported so far (at least 7.6 times higher than the previous benchmark3,6,25), and does so without generating chemical waste (Table S5†). The deposition and stabilisation of Pt atoms onto a g-C3N4 framework was confirmed by aberration corrected scanning transmission electron microscopy (AC-STEM) and extended X-ray absorption fine structure (EXAFS) measurements, and was further supported by density functional theory (DFT) calculations. Furthermore, the photocatalytic performance of Pt/g-C3N4 was tested for hydrogen evolution. Our Pt/g-C3N4 shows outstanding photocatalytic activity when compared with similar reported systems (Table S6†). AC-STEM and X-ray photoelectron spectroscopy (XPS) analysis before and after the hydrogen evolution reaction confirmed the stability of Pt atoms on a g-C3N4 framework. These results demonstrate that the scalable magnetron sputtering approach can generate effective ADMCs for catalytic processes such as hydrogen generation using only one clean synthetic step.
The deposition of Pt atoms on g-C3N4via magnetron sputtering was carried out using a bespoke AJA International system (Fig. S1, see the ESI† for details). g-C3N4 (1 g) was placed into a tailor-made stirring sample-holder, and then loaded into the magnetron sputtering chamber (reaching 3 × 10−8 torr background pressure in 40 min). After waiting 10 min for background pressure stabilisation, Ar gas (99.9999%) was introduced into the chamber, reaching 3 × 10−3 torr. Upon using an applied power (370 V and 16 mA), highly energetic ions were accelerated against the Pt target leading to a cascade of collisions at the surface of the Pt target, thus ejecting primarily individual neutral Pt atoms from the target22,27,28 which were then deposited onto the g-C3N4 framework (Fig. 1a). The g-C3N4 powder was stirred constantly during the Pt atoms deposition to allow for atomic dispersion throughout the whole powder (Fig. 1a). In this fashion, Pt atoms deposition was carried out for 12 min, yielding Pt/g-C3N4 (0.5 wt% of Pt onto 1 g of g-C3N4) measured by inductively coupled plasma optical emission spectrometry (ICP-OES). This gave an ADMC production-rate of 4.8 mg h−1, with a Pt loading on g-C3N4 of 5 mg and a total ‘feedstock-to-product’ magnetron sputtering process time of 63 min, which is the highest rate of production reported so far (Table S5†). Diverse characterisation techniques revealed no significant differences after deposition of the Pt atoms onto g-C3N4, including FTIR, PXRD, specific surface area and CHN elemental analysis (Fig. S2, S3 and Table S1†). Notably, the PXRD data show no evidence of either Pt or PtO2 in bulk or nanoparticulate form (Fig. S2b†). To demonstrate the generality of this method, Ni and Co atoms were atomically dispersed onto g-C3N4 framework using similar parameters (Fig. S5†).
AC-STEM measurements revealed highly dispersed Pt atoms in the g-C3N4 framework, with no Pt NPs observed (Fig. 1b and S3†). XAS measurements were performed to probe the local environment of Pt atoms (Fig. 1c and d). Fig. 1c shows the XANES spectrum of Pt/g-C3N4, and the standard spectra of Pt foil and PtO2. The white line intensities of these spectra revealed the oxidation state of the Pt atoms in each case, and the intensity for Pt/g-C3N4 was intermediate between that of Pt foil and PtO2. This suggested that Pt atoms were slightly positively charged due to coordination with the g-C3N4 framework, which agreed with XPS measurements (Fig. S3a†).29–31 The Pt K-edge EXAFS of Pt/g-C3N4 showed only one peak at ca. 1.5 Å (not phase-corrected), which can be associated to Pt–N/C coordination (Fig. 1d and Table S2†).10,30,31 No peaks associated with Pt–Pt and/or Pt–O coordination were observed (Fig. 1d), which is consistent with the AC-STEM and PXRD analysis. To obtain further atomistic insight into the interaction between Pt atoms and g-C3N4, DFT calculations were performed (see the ESI† for details). For these calculations, the nitrogen-interstice of s-triazine and tri-s-triazine units of g-C3N4 framework were considered as sites for stabilisation of single Pt atoms (Fig. S6 and Table S3†).32 Therefore, we suggest that Pt atoms were stabilised in the threefold nitrogen-interstice (Fig. 1e), as the first EXAFS peak appeared at a slightly shorter bond distance than would be expected for Pt–O, whereas the sixfold nitrogen Pt–N appeared at a slightly longer bond distance than Pt–O (Tables S2 and S3†). This corroborated with our EXAFS results and previous reports on Pt atoms stabilisation onto g-C3N4.10,31
Photocatalytic hydrogen production reactions were performed to investigate the catalytic photoactivity of Pt/g-C3N4 and g-C3N4 (Fig. 2a and Table S4†). As expected, pristine g-C3N4 showed negligible hydrogen evolution and hydrogen was only detected in large enough quantities to be reliably measured after 3 h. In contrast, Pt/g-C3N4 showed a much higher activity, improving the hydrogen production 3333 ± 450 times after 5 h of reaction. Moreover, our results compare favourably to similar systems previously reported, demonstrating the high photocatalytic activity of our Pt/g-C3N4 system (Table S6†).
To gain further insights into the high photocatalytic activity of Pt/g-C3N4, we carried out photoluminescence (PL) measurements to investigate the efficiency of charge transfer and separation (Fig. 2b). The PL spectrum of g-C3N4 exhibits an intense peak at ca. 460 nm, which is ascribed to electron–hole recombination of g-C3N4. The deposition of Pt onto the g-C3N4 surface led to a decrease of the PL intensity, implying suppressed charge carrier recombination (Fig. 2b) arising from the efficient dissociation of photogenerated electron–hole pairs of Pt/g-C3N4, in agreement with the photocatalytic results.33,34 AC-STEM images and XPS analysis of the Pt/g-C3N4 after 5 h under reactions conditions show no agglomeration and no significantly changes in Pt electronic properties (Fig. 2c and S7,† respectively), the major mechanism responsible for catalyst deactivation, which demonstrates the high stability of Pt atoms in the g-C3N4 framework.
Footnote |
† Electronic supplementary information (ESI) available: Sample preparation process, experimental techniques, additional figures, tables, and references. See DOI: 10.1039/d1ta08372d |
This journal is © The Royal Society of Chemistry 2021 |