Theoretical investigation on the interaction of subnano platinum clusters with graphene using DFT methods

doi:10.1016/j.commatsci.2014.09.033

Computational Materials Science

Volume 96, Part A, January 2015, Pages 268-276

https://doi.org/10.1016/j.commatsci.2014.09.033 Get rights and content

Highlights

•
The interaction between subnano platinum clusters and graphene has been investigated.
•
The bridge site adsorption is the most stable configuration for Pt_n/PG.
•
The adsorption energies are gradually decreased with the Pt_n clusters size increasing.
•
The PDOS for the d orbitals of Pt atoms and p orbitals of C atoms at the interface have been analyzed.
•
The electron transfer is relevant to the adsorption energy but not absolutely.

Abstract

The interaction between subnano platinum clusters which are composed of 4–27 atoms with pristine graphene (PG), monovacancy graphene (VG) and Stone–Wales defect graphene (5775) has been investigated using the methods of density functional theory (DFT). According to the similar structures of Pt_n clusters and eliminating the disruption of different adsorption styles, the subnano platinum clusters all interact with the substrates through one platinum atom. The calculation results show that the adsorption energy of Pt_n clusters interacted with three type substrates are all gradually decreased with the size increasing. The adsorption energies of Pt_n adsorbed on VG are increased significantly compared with those adsorbed on PG and 5775. The results have been further demonstrated by the partial density of states (PDOS), spin density distribution, Mulliken population analysis and electron density difference of interfacial Pt and C atoms. The analyses results indicate that the size of Pt_n clusters have effects on the interfacial interaction.

Graphical abstract

Introduction

Direct methanol fuel cells (DMFC) have received special attention as a novel device for converting energy owning to their higher energy density, less pollutant emission, and lower operating temperature [1], [2], [3]. The catalyst, as a momentous part determining the performance and cost of DMFC, has been extensively studied. Although the research of non-platinum catalysts has been made certain progress in recent years, the platinum is still the most effective and leading catalyst for DMFC [4], [5], [6]. However, the world reserve of platinum could not suffice the industry development. To resolve this problem, improving the utilization and efficiency of Pt catalyst is necessary. One of the practical methods is supported platinum on the carbon materials with large specific surface such as carbon-block [7], mesoporous carbon [8], carbon nanotube [9], [10] and graphene [11], [12]. Among the various carbon materials, garphene as catalyst support has received growing attention since it was discovered by Novoselov et al. in 2004 [13]. The graphene exhibits large specific surface, high electrical conductivity and good thermal stability [14], [15], [16]. Ideally, pristine graphene is stable hexagonal monolayer network consisted of sp²-hybridized carbon atoms, so the interaction between platinum and pristine graphene is relatively weak. Pt nanoparticles can migrate and diffuse when are supported on pristine graphene, and consequently its stability and activity will be reduced [17], [18], [19]. Therefore, pristine graphene cannot be a suitable support for Pt catalyst. However, experimentally, graphene is always imperfect which contains defects such as vacancies, reconstruction and functional groups [20], [21], [22]. Fortunately, these defects locally increase the reactivity of the structure and trap other atoms through changing mechanical electronic and magnetic properties of graphene [23].

It is a meaningful subject to clarify the interaction of Pt nanoparticles with graphene. Many people have researched it from theoretically and experimentally, but not comprehensively. Okamoto [24] examined the interface between graphene and a Pt₁₃ or Au₁₃ cluster. He found that introducing a carbon vacancy into a graphene sheet enhanced the interaction between graphene and the metal clusters. Five- or seven-member rings introduced into the graphene also increased the stability of the interface. Okazaki-Maeda et al. [25] investigated structures of various configuration Pt_n clusters on pristine graphene by first principles calculations. They concluded that the interfacial interaction between a Pt cluster and graphene seriously depends on the shape and size of a cluster and manner of contact on a graphene sheet. Fampiou et al. [26] researched the interaction of Pt_n (n = 1–4, 13) clusters to pristine graphene, vacancies, unreconstructed divacancies, pentagon–octagon–pentagon (5–8–5) reconstructed divacancies, and haeckelite (555–777) reconstructed divacancies in graphene by employing a combination of empirical-potential-based simulated annealing and DFT calculations. They found that point defects and their reconstructions in graphene acted as strong binding traps for Pt_n clusters and there was a clear tendency of charge transfer from Pt_n clusters to the graphene substrate according the electronic structure analysis. Their perspective about the substrates containing almost all of defect types was quite comprehensive containing, but the size of Pt_n clusters only limited to n ⩽ 13. Ramos-Sanchez et al. [27] analyzed Pt_n (n = 4, 6, 13, 19, 38) clusters with graphite by DFT calculation. Their results demonstrated that the interaction energy between Pt and graphite had been contributed from orbital hybridization and Van der Waals interactions. Julio et al. [23] created trapping centers in graphene with a highly focused electron beam and then succeeded localize single metal atoms or clusters on graphene. Siburian et al. [28], [29] expounded the formation mechanism of Pt subnano-clusters on graphene nanosheets by using XRD, XPS and HR-TEM methods and controlled the size of Pt clusters to subnanometer by adjusting pH and Pt loadings. Ding et al. [30] understood and modeled the metal–graphitic interfaces from various perspectives including microscopy, spectroscopy, electro-chemical techniques and electrical measurements. Yoo et al. [31] proved that electrocatalytic activity of Pt subnano-clusters on graphene nanosheet surface was improved signally through current–potential curves for methanol oxidation reaction and CO stripping voltammograms. They inferred that these phenomena were connected to the specific electronic structures of Pt from existing defects sites and forming 0.5 nm Pt particles on graphene nanosheets.

On the basis of the above analysis, there is a necessity to further study the Pt_n clusters size effects for the catalyst composite. The aim of our study is to investigate the interaction of different-sized Pt_n clusters on pristine (PG), monovacancy (VG) and Stone–Wales defect graphene (5775), and then clarify the effect about interface of them. From experimental works, it is known that Pt subnano-clusters on graphene reveal an unusually high activity. Siburian et al. [28] reported that subnano Pt clusters (0.8 nm) with an extremely large surface area (170 m²/g) showed a high CO tolerance as an anode catalyst in PEFCs. Yoo et al. [31] found that Pt particles below 0.5 nm in size were formed on graphene nanosheets (GNS) and improved the catalytic activities of Pt/GNS. Xu et al. [32] obtained the homogeneous distribution Pt particles with an average diameter of 1.4 ± 0.2 nm on graphene. The diameters of these clusters including Pt₄, Pt₆, Pt₁₈ and Pt₂₇ are very similar with those of experimental reports. We calculated the adsorption energies of Pt_n clusters on graphene, and also compared the alteration of adsorption energies, Pt–Pt bonds and Pt–C bonds. In addition, we analyzed the partial density of states (PDOS) and spin density distribution about interfacial Pt and C atoms. Then the detailed information about the electron density difference and the electrons transfer in the interface were discussed.

Section snippets

Computational details

In this work, all the geometry optimization calculations were performed using DMol³ package (versions 5.0, Accelrys. Inc.) [33], [34]. The Perdew–Burke–Ernzerhof (PBE) [35] function within the generalized gradient approximation (GGA) was adopted to describe the exchange–correlation interaction. The double numerical DNP was chosen as the basis set. The Brillouin zone was conducted using a 3 × 3 × 1 Monkhorst–pack grid. All structures were fully optimized without any symmetry constraints until the

Geometries of Pt_n (n = 4, 6, 18 and 27) clusters on the PG, VG and 5775

The optimized results of Pt₄, Pt₆, Pt₁₈ and Pt₂₇ clusters adsorbed on the graphene are depicted in Fig. 3, Fig. 4. And the specific data for adsorption energies, average bond lengths of Pt–Pt, interfacial Pt–Pt and Pt–C are displayed in Table 1. For Pt₄/PG and Pt₆/PG, two structures are considered for each composite in the initial configuration. One is a Pt atom adsorbed at the top site above carbon atom, the other is a Pt atom adsorbed at the bridge site between two neighboring carbon atoms.

Conclusions

In this contribution, we have performed DFT calculations about different-sized Pt_n clusters with diameters range of 0.3–1.0 nm interacted with PG, VG and 5775. In order to exclude other factors and mainly consider the size effect, all the clusters interact with the substrates through one platinum atom. When Pt_n clusters supported on the PG, VG and 5775, the interfacial Pt–Pt bonds length present different degrees of increase and the top Pt–Pt bonds length shorten especially in the larger

Acknowledgments

This work is supported by the National Basic Research Program of China (973 Program) (2012CB932800f), the National Natural Science Foundation of China (Nos. 21373099, 21303067, 21173097), Science and Technology Research Program of Higher Education of Jilin Province, China ([2011] No. 388).

References (45)

L. Elbaz et al.
J. Electroanal. Chem.
(2013)
D. Zhao et al.
Electrochim. Acta.
(2013)
E. Auer et al.
Appl. Catal., A: Gen.
(1998)
L. Dong et al.
Carbon
(2010)
Y. Okamoto
Chem. Phys. Lett.
(2006)
K. Okazaki-Maeda et al.
Surf. Sci.
(2010)
K. Bhattacharyya et al.
Chem. Phys. Lett.
(2007)
L. Cao et al.
Angew. Chem. Int. Ed.
(2006)
J.S. Yu et al.
J. Am. Chem. Soc.
(2002)
J. Zhang et al.
Chemistry
(2013)

G. Wu et al.

Chem. Commun.

(2013)

C. Zhang et al.

ACS Catal.

(2014)

Y. Mu et al.

J. Phys. Chem. B

(2005)

Y. Xing

J. Phys. Chem. B

(2004)

N. Kulshrestha et al.

ACS Appl. Mater. Interfaces

(2013)

K.S. Novoselov et al.

Science

(2004)

A.A. Balandin et al.

Nano Lett.

(2008)

M. Orlita et al.

Phys. Rev. Lett.

(2008)

A.K. Geim et al.

Nat. Mater.

(2007)

R. Kou et al.

J. Am. Chem. Soc.

(2011)

Y. Gan et al.

Small

(2008)

B.H. Morrow et al.

Nanotechnology

(2008)

Cited by (22)

Atom hybridization of metallic elements: Emergence of subnano metallurgy for the post-nanotechnology
2023, Coordination Chemistry Reviews
Citation Excerpt :
Pt SNPs interact with the substrates and the adsorption energy of Pt SNPs decreased with the increase of size. This system was further studied by the partial density of states, spin density distribution, Mulliken population analysis and electron density difference of interfacial Pt and C atoms, demonstrating that the size of Pt SNPs has a significant effect on the interfacial interaction[187]. Energies, structural and vibrational characteristics of isomers of step-wise hydrogenated aluminum SNPs AlnHm (n = 13, 144, and 189) were calculated using the DFT B3LYP method.
Miniaturizing metallic materials from the nanoscale to the atomic scale transforms the structures and physical properties. Among these sub-nanoscale materials, some catalysts exhibit activity superior to conventional nanoparticle catalysts. Furthermore, some subnanoparticles become superatoms that exhibit unusual physical properties that deviate from the properties of the original elements. However, conventional metallurgy and chemistry often do not apply to synthetic strategies, structural analysis, and functional design of materials because crystal structures disappear at the sub-nanoscale. This review focuses on the synthesis, structural analysis, and functionality at the sub-nanoscale, overviewing a series of materials research, so-called “atom hybridization,” in which atoms are hybridized to create desired materials for their applications.
Adsorption behavior and sensing properties of toxic gas molecules onto Pt<inf>n</inf>Be (n = 5, 7, 10) clusters: A DFT benchmark study
2022, Materials Today Communications
The adsorption characteristics of some toxic gas molecules (CO, SO, NO, NH₃, PH₃, CH₄, and O₃) onto Pt_nBe (n = 5, 7, 10) cluster were scrutinized using density functional theory (DFT) calculations. Adsorption energy analysis revealed that the adsorption energy value of all gas@Pt_nBe complexes decreases with cluster size increasing. It was discovered that CH₄ gas molecule is physically adsorbed onto Pt₅Be cluster, whereas other gas molecules are highly chemisorbed. Based on natural bond orbital (NBO) analysis, the maximum charge transfer upon adsorption of gas molecules was observed for SO@Pt_nBe complex. The band gap variation in Pt₅Be cluster as a result of adsorption of NO gas molecule was found to be the highest $(-$ 31.37%). However, much shorter recovery time (9 $\times$ 10⁻⁴ s) disclosed that Pt₅Be cluster seems available for monitoring CH₄ gas molecule. Finally, quantum theory of atoms in molecules (QTAIM) and electron location function (ELF) analyses illustrated the main presence of electrostatic with partially covalent interactions in the studied complexes.
Combining metal nanoclusters and carbon nanomaterials: Opportunities and challenges in advanced nanohybrids
2022, Advances in Colloid and Interface Science
Citation Excerpt :
Copper nanoclusters (CuNCs) are the most investigated non-noble metal NCs, as they have similar optical properties to noble metal NCs and are equally excellent catalysts, but at much lower cost [63,64]. Clusters of other noble metal beyond Au and Ag have also been approached in recent years, namely platinum (PtNCs) [17,65–67] and palladium (PdNCs) [68,69] ones, owing to their excellent catalytic properties. Although there is growing interest on new types of MNCs and bimetallic clusters [70], nanoclusters of Au, Ag and Cu are still significantly more investigated than all others.
The development of functional materials with uniquely advanced properties lies at the core of nanoscience and nanotechnology. From the myriad possible combinations of organic and/or inorganic blocks, hybrids combining metal nanoclusters and carbon nanomaterials have emerged as highly attractive colloidal materials for imaging, sensing (optical and electrochemical) and catalysis, among other applications. While the metal nanoclusters provide extraordinary luminescent and electronic properties, the carbon nanomaterials (of zero, one or two dimensions) convey versatility, as well as unique interfacial, electronic, thermal, optical, and mechanical properties, which altogether can be put to use for the desired application. Herein, we present an overview of the field, for experts and non-experts, encompassing the basic properties of the building blocks, a systematic view of the chemical preparation routes and physicochemical properties of the hybrids, and a critical analysis of their ongoing and emerging applications. Challenges and opportunities, including directions towards green chemistry approaches, are also discussed.
High-efficiency water purification for methyl orange and lead(II) by eco-friendly magnetic sulfur-doped graphene-like carbon-supported layered double oxide
2021, Journal of Hazardous Materials
Traditional disposal techniques for the spent layered adsorbents after capturing organics suffer from intractable obstacles, such as resource waste and secondary pollution. To address this diploma, we here developed the “resource-utilization” strategy, i.e., converting the organic layered double hydroxide (as representative) to magnetic sulfur (S)-doped graphene-like carbon-supported layered double oxide (MG/S-LDO) to be reutilized in water purification. The as-prepared MG/S-LDO exhibited outstanding remediation ability toward methyl orange (MO) and lead(II), with the adsorption capacity of 1456 and 656 mg g⁻¹, respectively. Specifically, the residue concentration of Pb²⁺ was reduced to 0.15 mg L⁻¹ within 1 h, which met the discharge limit of the secondary industrial wastewater. MG/S-LDO could also maintain the preeminent adsorption capability under various interferences (such as wide pH and co-existing ions), even in the authentic water matrices. The removal mechanisms were systematically investigated to unveil that MO removal was dominated by metal-complexation, “memory effect”, and π-π electron donor–acceptor (EDA). While for Pb²⁺ removal, besides the released OH^- from LDO as precipitate agent, the vacancy defect resulting from the S doping played a crucial role in electron interaction between Pb²⁺ and S-doped graphene. Additionally, the MG/S-LDO was further confirmed as an eco-friendly adsorbent with excellent reusability via the acute toxicity tests using green algae and multiple cycle experiments. This work provides a novel resource-utilization strategy for organic layered wastes to construct the functional eco-friendly materials in wastewater purification realm.
Theoretical insight into the role of nitrogen in the formic acid decomposition over Pt<inf>13</inf>/N-GNS
2021, Applied Surface Science
Citation Excerpt :
Thus, we choose this calculation level for the CO adsorption in the present work. Former researches studied the adsorption of Pt clusters with different sizes on GNS and results denote that binding energy (the energy released after binding) decreases when the size of Pt cluster goes up [60,61]. In most of experimental works [18,21,62], dispersed Pt nanoparticles exhibit a distribution of size between 0.75 nm and 2 nm.
Catalytic decomposition of formic acid is regarded as one of the most promising hydrogen source conversion technologies. Nitrogen doped carbon supported metal catalyst emerges in recent years and delivers excellent performance in formic acid hydrogenation. However, there is not a well-recognized explanation about the real role of the nitrogen dopant in carbon support. In this work, density functional theory-based calculations were used to individually study the ligand effect and catalytic effect from the nitrogen dopant. Ligand effect mainly tunes the electronic properties of metal active center by shifting d-band center far away from Fermi level. The result unravels that CH scission path is more favorable compared with OH scission path. Catalytic effect is originated from the lower electrostatic potential of nitrogen active site compared with platinum, making N site an efficient capturer for hydrogen atom. Though activation energy for cleaving OH bond is higher than CH bond, nitrogen site can efficiently cleave the OH bond. Microkinetic simulations are performed to obtain the best nitrogen doping concentration in the carbon support. It implies that the optimal nitrogen concentration is a function of temperature, according to the optimized curve. This work will improve the understanding of mechanism of formic acid decomposition and provide new method in modifying metal/carbon support catalysts.
CO<inf>2</inf> hydrogenation to formic acid over platinum cluster doped defective graphene: A DFT study
2020, Applied Surface Science
Citation Excerpt :
Electrons of these Pt atoms prefer to transfer to the surface, resulting in a positive charge transfer of 9.54 e in the Pt4 cluster. According to our calculations, the binding energy of Pt4 cluster and graphene system is −7.09 eV, consistent with the value of −7.11 eV in the work of Qi et al. [52]. And the cohesive energy of Pt4 cluster is −2.64 eV, also consistent with −2.68 eV in the work of Munieswaran et al[53].
The hydrogenation of CO₂ to formic acid is an important reaction in environmental catalysis, which can both alleviate the greenhouse effect and produce useful chemicals. Platinum element catalysts need to be investigated to realize large-scale development due to the expensive and scarce features. In this work, CO₂ hydrogenation to formic acid over Pt₄ cluster doped single-vacancy graphene was investigated using density functional theory. Catalyst configuration was optimized to perform corresponding gas adsorption and reduction reaction. Four reaction pathways were explored according to the reaction mechanism of Langmuir-Hinshelwood (L-H), Eley-Rideal (E-R) and termolecular Eley-Rideal (TER). Kinetic analysis was used to evaluate the reaction rate of different processes under the given temperature range, and the potential effect of CO molecule was also considered to better understand the feasibility. Results showed that Pt₄/SV was a stable and high activity catalyst. The minimum activation energy among different pathways was 0.56 eV and TER could be the dominant reaction mechanism of CO₂ hydrogenation. This work not only provides a promising catalyst but also gives a more deep understanding of CO₂ reduction technology and the future applicability of platinum metal catalysts.

View all citing articles on Scopus

View full text

Theoretical investigation on the interaction of subnano platinum clusters with graphene using DFT methods

Highlights

Abstract

Graphical abstract

Introduction

Section snippets

Computational details

Geometries of Ptn (n = 4, 6, 18 and 27) clusters on the PG, VG and 5775

Conclusions

Acknowledgments

J. Electroanal. Chem.

Electrochim. Acta.

Appl. Catal., A: Gen.

Carbon

Chem. Phys. Lett.

Surf. Sci.

Chem. Phys. Lett.

Angew. Chem. Int. Ed.

J. Am. Chem. Soc.

Chemistry

Chem. Commun.

ACS Catal.

J. Phys. Chem. B

J. Phys. Chem. B

ACS Appl. Mater. Interfaces

Science

Nano Lett.

Phys. Rev. Lett.

Nat. Mater.

J. Am. Chem. Soc.

Small

Nanotechnology

Geometries of Pt_n (n = 4, 6, 18 and 27) clusters on the PG, VG and 5775