Synthesis and Characterization of Mn₃O₄-Graphene Nanocomposite thin Film by an ex situ Approach

Abstract

In this study, we report a new approach for $Mn_3O_4$-graphene nanocomposite by ex situ method. This nanocomposite shows two-dimensional aggregation of nanoparticle, and doping effect by decorated manganese oxide ($Mn_3O_4$), as well. The graphene film was made through micromechanical cleavage of graphite on the $SiO_2/Si$ wafer. Manganese oxide ($Mn_3O_4$) nanoparticle with uniform cubic shape and size (about $5.47{\pm}0.61$ nm sized) was synthesized through the thermal decomposition of manganese(II) acetate, in the presence of oleic acid and oleylamine. The nanocomposite was obtained by self-assembly of nanoparticles on graphene film, using hydrophobic interaction. After heat treatment, the decorated nanoparticles have island structure, with one-layer thickness by two-dimensional aggregations of particles, to minimize the surface potential of each particle. The doping effect of $Mn_3O_4$ nanoparticle was investigated with Raman spectra. Given the upshift in positions of G and 2D in raman peaks, we suggest that $Mn_3O_4$ nanoparticles induce p-doping of graphene film.

Introduction

Graphene, a two-dimensional nanosheet of sp2-hybridized carbon, has attracted tremendous interest as a key material in various scientific areas, because of its unique physical, chemical, and mechanical properties.1-3 In particular, graphene possesses high surface area, superb electron transport, visible transparency, high thermal conductivity, and outstanding chemical tolerance. Because of these advantages, this carbon nanosheet has been investigated with significant interest for its use as a base material for energy storage devices, such as pseudocapacitors,4 or lithium ion batteries,5 heterogeneous catalysis,6 and electrochemical sensors.7 In particular, graphenebased nanocomposites are expected to display enhanced performance in these applications, due to their improved and synergetic physicochemical properties, which cannot be achieved by graphene itself.4,8,9

In constructing graphene based nanocomposites, two synthetic approaches are generally used.10,11 One is that nanoparticles directly grow on the surface of the graphene nanosheets, which is called an in situ approach. The second involves the post synthesis of nanoparticles with the desired size and morphology, followed by modification and subsequent connection to the surface of graphene, which is named an ex situ approach. Although the in situ approach is more widely used in the synthesis of graphene-based nanocomposites, due to its simplicity of reaction, and not needing the additional modification of organic molecules, this approach has limited performance for application. For example, in the case of metal nanoparticle-graphene nanocomposite, the reduction of metals on the surface of graphene usually leads to a partially collapsed structure, due to the destruction of the regular stacks of graphene sheets, resulting in a lower device performance. Also, this approach leads to localized attachment, irregular size distribution, and difficult control of the amount of the nanoparticle on the graphene nanosheet.

Compared with in situ growth of nanoparticles on graphene, the ex situ approach is excellent at overcoming these incompatibilities between nanoparticle syntheses and the formation of nanocomposites. The resulting nanocomposites present better distribution and uniform shape, and the desired size of attached nanoparticle can be obtained more easily. To fabricate nanocomposites by an ex situ approach, nanoparticle components were synthesized in advance, and then attached to the surface of graphene, by linking agents that utilize covalent or noncovalent interactions between graphene and the desired nanoparticles (van der Waals interactions, hydrogen bonding, π-π stacking, or electrostatic interactions).11

Among various nanoparticles that can act as a building block for the construction of nanocomposties, manganese oxides have useful or potentially useful properties and applications, owing to their excellent electric, magnetic, and catalytic properties, low price, and environmental compatibility. In particular, Mn3O4 is known to be an effective catalyst for de-NOx systems,12 and the raw materials for electronics and information devices. Very recently, nanosized Mn3O4 was reported as a high-capacity anode material for rechargeable lithium batteries.13

In this regard, we fabricated Mn3O4-graphene nanocomposite by self-assembly between monodispered Mn3O4 nanocrystalline and single-layer graphene film, via an ex situ approach. The physico-chemical characterizations of the obtained nanocomposite films were carried out as a function of the number of graphene layers, along with the doping effect by attachment of Mn3O4 nanoparticles.

 

Experimental

Sample Preparation.

Synthesis of Uniform sized Mn3O4 Nanoparticle: Uniform sized Mn3O4 nanoparticle was prepared, by thermal decomposition of manganese (II) acetate tetrahydrate in the presence of surfactant molecules, such as oleic acid and oleylamine.14 0.25 g (1 mmol) of manganese (II) acetate tetrahydrate (≥ 99%, Aldrich), and a mixture of 0.63 mL (2 mmol) of oleic acid (≥ 99%, Aldrich) and 4.7 mL (10 mmol) of oleylamine (70%, Aldrich) were dissolved in 15 mL of xylenes (≥ 98.5%, Aldrich). Then, the solution was heated to 90 oC, with a heating rate of 1 ℃/min, under vigorous stirring. When the temperature reached 90 ℃, 1 mL of deionized water was quickly injected into the solution. The solution was maintained at 90 ℃ for 3 h, to induce sufficient growth. Finally, the resulting solution was cooled to room temperature, and absolute ethanol was slowly added, to precipitate Mn3O4 nanoparticles. The precipitate was separated by centrifugation at 8000 rpm for 10 min, and washed twice with absolute ethanol. The resulting product was redispersed in n-hexane.

Synthesis of Mn3O4-graphene Nanocomposite: Monolayer graphene film was obtained, by micromechanical cleavage from the highly oriented pyrolytic graphite (HOPG) on SiO2 (300 nm)/Si (100) wafer (P-type, Namkang Hi-Tech).15 The obtained graphene film was rinsed with n-hexane, and dried under N2 stream. To decorate Mn3O4 nanoparticles on the graphene surface, the substrate was dipped into a known concentration of Mn3O4 nanoparticle (160 ppm in n-hexane) for 20 min, and then washed three times with n-hexane, to remove unattached excess nanoparticles. Finally, the Mn3O4- graphene composite film was dried in air, and calcined at 500℃ for 3 h under Ar atmosphere, with 100 mL/min of gas flow.

Sample Characterization.

The crystal structure of the obtained Mn3O4 nanoparticles was studied by powder X-ray diffraction (XRD) measurement, using Ni filtered Cu Kα radiation (λ = 1.5418 Å, Rigaku D/MAX-2000), with a graphite diffracted beam monochromator. The pattern was recorded at an operating voltage of 40 kV, and a current of 30 mA. The thermal behavior of nanoparticles was obtained by thermogravimetric analysis (TGA, PerkinElmer STA 6000). TGA was performed for powder sample, under N2 with 20 mL/min of gas flow. The temperature range was from 30 to 800 ℃, with a heating rate of 5 ℃/min. Fourier transform-infrared spectroscopy (FT-IR, Varian FTS 1000 FT-IR) was performed in the frequency range of 400-4000 cm−1 at a resolution of 4 cm−1, with the standard KBr disk method. The morphology and size of the obtained Mn3O4 nanoparticles were investigated, using high resolution-transmission electron microscopy (HRTEM, JEOL JEM-3010) with an accelerating voltage of 300 kV. Before and after annealing, the morphological change of Mn3O4 nanoparticle on graphene film was measured by Atomic Force Microscopy (AFM, Bruker-Nano N8 NEOS). The AFM was operated in tapping mode, using a Tap 190AIG probe (Nanosensor). The doping type of Mn3O4-graphene nanocomposite film was investigated by Raman spectroscopy (Senterra Raman Microscope), with a 532 nm excitation laser

 

Results and Discussion

Figure 1 shows the XRD pattern of the prepared Mn3O4 nanoparticle. The XRD pattern is matched and indexed to Hausmannite Mn3O4 (JCPDF file no. 80-0382).14 All peaks of sample present broad features, due to the nanocrystalline nature of the obtained Mn3O4 nanoparticle. These peak broadenings are used to indirectly calculate the nanoparticle size, by using ‘Scherrer‘s equation’, d = Kλ/B cos θ, where d is the nanoparticle size in Å, K is the shape factor, B is the peak broadening of the diffraction line measured at half of its maximum intensity in ‘radians’, λ is the wavelength of Xrays source in Å, and θ is the Bragg’s diffraction angle.16 The peak broadening was determined through computer-fit of the XRD pattern, using the pseudo-Voigt function. Considering the broadening of the (211) diffraction line (2θ = 36.05°) of Mn3O4 nanoparticle, the calculated average nanoparticle size was about 5.2 nm. This calculated size is similar to the mean size (about 5.47 ± 0.61 nm), calculated statistically from TEM image. All the diffraction peaks of the obtained Mn3O4 nanoparticle could be refined as unit-cell parameters of a = 5.756 Åand b = 9.438 Å, which are consistent with those of the reference Hausmannite Mn3O4 (a = 5.765 Åand b = 9.442 Å).

Figure 1.(a) JCPDF file of Hausmannite Mn3O4 (JCPDF file no. 80-0382), and (b) Powder XRD patterns of prepared Mn3O4 nanoparticle.

Figure 2.Fourier transform infrared spectra of (a) oleic acid, (b) oleylamine, and (c) Mn3O4 nanoparticles.

In order to characterize the organic moiety of surfactants on the surface of nanoparticles, FT-IR spectra were collected for the prepared Mn3O4 nanoparticle, as shown in Figure 2 (which also contains the spectra of oleic acid (Figure 2(a)) and oleylamine (Figure 2(b)). In all spectra, the intense peaks at 2924 and 2854 cm−1 are associated with asymmetric and symmetric stretching of the CH2 existing in long chains of oleic acid and oleylamine molecules, while a weak peak at 3007 cm−1 is assigned cis-HC=CH- arrangement in surfactants.17 In Figure 2(c), the broad band at ∼ 3416 cm−1 corresponded to the OH stretching vibration of the absorbed ethanol on the nanoparticle surface, and NH vibration derived from oleylamine.1819 Furthermore, the C=O stretch band of the carboxyl group, which was present at 1710 cm−1 in pure oleic acid [Figure 2(a)], disappeared in Figure 2(c). After absorption of oleic acid on the surface of Mn3O4 nanoparticle, two new peaks appeared at 1546 and 1404 cm−1, which indicated the asymmetric νas(COO-) and symmetric νs(COO-) stretching mode, respectively.20 This wavenumber separation, Δ, of carboxylate can be used to distinguish the type of interaction between the carboxylate head and the metal atom, such as monodentate Δ (200-300 cm−1), bridging bidentate Δ (110-200 cm−1), and chelating bidentate Δ (< 110 cm−1). In this work, the Δ (142 cm−1) indicated that the chemisorption between the carboxylate group of oleic acid and Mn atom has a type of bridging bidentate. The sharp peaks at 628 and 524 cm−1 were assigned to the characteristic of coupling modes between Mn-O in tetrahedral and octahedral site for Mn3O4 nanoparticle.21

Figure 3.TEM images (the upper inset is HR-TEM image and the bottom inset is FFT pattern of nanoparticle) of Mn3O4 nanoparticle, (b) a line profile about the upper inset of (a), and (c) histogram of size distribution of Mn3O4 nanoparticles and fitting gaussian curve.

The morphology and size of the obtained Mn3O4 nanoparticle was observed by high resolution transmission electron microscopy (HR-TEM). Figure 3(a) reveals that the nanoparticles were successfully synthesized with uniform size and cubic shape. The HR-TEM image of a single nanoparticle, depicted in the upper inset of Figure 3(a), shows well an atomic lattice fringe, demonstrating the single crystalline nature of the obtained Mn3O4 nanoparticle. The interplanar distance is about 0.254 nm, shown more clearly by the interval between the peaks of the line profile in Figure 3(b), which is consistent with the (211) planes of tetragonal Mn3O4 nanoparticle. In the bottom inset of Figure 3(a), the fast Fourier transform (FFT) pattern also corresponded with the above result. The size distribution of nanoparticle was determined using computing-fitting based on Gaussian function, to analyze more than 100 particles, as shown in Figure 3(a). The nanoparticles had a mean size of 5.47 nm with standard deviation of 0.61 nm, as displayed in Figure 3(c).

Figure 4.TGA curve of prepared Mn3O4 nanoparticles in the temperature range of 30-800 ℃, in N2 atmosphere.

Thermogravimetric analysis (TGA) of the prepared Mn3O4 nanoparticles represents three weight loss transition steps in the temperature range of 30-800 ℃ under N2, as shown in Figure 4. The first step with a 2.1% weight loss, in the temperature range of 30-180 ℃, is ascribed to the evaporation of adsorbed solvents. The second weight loss of 14.3%, in the range of temperature of 180-460 ℃, is derived from the desorption and decomposition of surfactant molecules, which existed on the surface of nanoparticles. In detail, the gradual weight loss of 1.9% between 180 and 270 ℃ is attributed to the desorption of physisorbed surfactants, and the next transition of 12.4% in temperature of 300-460 ℃ is mainly due to the decomposition of chemisorbed oleic acid.22 The weight loss curve of chemisorbed molecules shows two distinct transitions. These different desorption processes may be because of different bonding forces of surfactant with nanoparticle. It was reported that there were two possible reasons. First, the different facets of nanoparticles represent different adsorption affinity.23 Second, the oleic acid may be bonded with two different oxidation states of manganese ion (Mn2+ and Mn3+).24 The final stage with 5.3% weight loss, is attributed to the further decomposition of residual organic moiety, and crystallization of manganese oxide.

Figure 5.AFM image of (a) before, and (b) after calcined decorated Mn3O4 nanoparticle on a single layer graphene film at 500 ℃, under Ar atmosphere. AFM section profiles along the three different lines intersect, in (a) and (b).

To confirm the thermal behavior of nanoparticles on graphene, atomic force microscopy (AFM) images are measured by noncontact mode. Figure 5shows a representative set of AFM images (before and after the heating treatment at 500 ℃ under Ar atmosphere), for decorated Mn3O4 nanoparticle on single layer graphene sheet. As illustrated in Figure 5(a), Mn3O4 nanoparticles represent uniform single layer coverage on graphene surface. Similar phenomenon reported hydrophobic interaction between the long chain tail part of surfactant molecules covering the surface of gold nanoparticles, and the aromatic carbon surface of HOPG substrate.25 In our case, Mn3O4 nanoparticles are assembled on the graphene surface at around 80 nm size (that means around 70 numbers of nanoparticles were aggregated through the drying process) with single layer thickness, through analysis of the cross-section view of the AFM image of Figure 5(a) [inset]. The overall features of the assembled nanoparticle, shown in Figure 5(a), display an apparent lack of resolution at the edges of each particle, due to a combination of the tip-sample convolution,26 and surfactantsurfactant interaction.25 The size of each nanoparticle was indicated to be around 10 nm, through the width of each fine peak in the cross-section view [inset of Figure 5(a)]. This resulting value was composed of the length of the surfactant molecules (around 2.0 nm), and the Mn3O4 nanoparticle core size (around 5.5 nm) (9.5 nm = 5.5 nm + 2.0 nm + 2.0 nm). Also, height analysis of the cross-section view reveals that the average height of decorated particles is around 6.5 nm. This means that Mn3O4 nanoparticles were decorated and aggregated as only one layer on the surface of graphene. The obtained height is lower than that of the calculated one, because the recorded height was derived from the sum of the topography, and the relative amplitude damping between surface and particle.27 After annealing, the decorated nanoparticles appear to be much better resolved, because of the removal of the surface attached surfactant [inset of Figure 5(b)].25 The size of assembled nanoparticles was significantly shrunk at around 45 nm, with isolated disk shape, maintaining the monolayer thickness. This aggregate size is well matched with the calculated area, by about 70 numbers of nanoparticle 2D assembly, with only the Mn3O4 core. It is worth noting that the assembled nanoparticles on the single layer graphene maintain their monolayer arrangement, even after thermal treatments at 500 ℃. However, the assembled nanoparticles on few layers graphene show different thermal behavior, compared with that of the single layer one, shown in Figure 6. After thermal treatments, the assembled Mn3O4 nanoparticles represent worm-like feathers that consist of nanoparticle aggregations of around 150 nm size, with mono or higher layer thickness. This means that the size and shape of aggregated nanoparticles are strongly influenced by the number of graphene layers. Similar phenomena are also reported in the previous literature that contains the aggregation behavior of Pt nanoparticles on single layer, or few layer graphene.28 So, we can expect that the state of the graphene layer plays an important role, not only in respect of the morphology of the attached nanoparticles, but also of the physicochemical properties of graphene, in graphene-nanoparticle composite systems.

Figure 6.AFM images of (a) Mn3O4-graphene composite including single-layer, few-layer, and SiO2 surface morphology, after annealing at 500 ℃ under Ar atmosphere. (b) and (c) show stacking patterns of Mn3O4 nanoparticle on few layer graphene. (d) Line profile analysis of Mn3O4-graphene composite of few layer graphene.

Raman spectroscopy is a powerful non-destructive technique for identifying the number of layers, structure, doping and disorder of graphene. Figure 7 compares the Raman spectra of single-layer graphene and annealing Mn3O4- graphene nanocomposite on only a single layer. In the case of single-layer graphene, the two most intense features are the G peak at 1581.5 cm−1, and the 2D peak at 2667 cm−1, in Figure 7(a). The G peak is due to the doubly degenerate zone E2g mode. On the other hand, the 2D peak is induced by the second order of zone-boundary, which is related to the A1g breathing mode.29 The type of doping can be determined from the position shift of the G and 2D peaks,30 after hybridization with nanoparticles. As reported in the literature, the upshift of the G peak position and the downshift of the 2D peak position means n-doping of graphene. Otherwise, the upshift of the G peak position and the upshift of the 2D peak were represented as the p-doping of graphene. In our case, the position of the G peak has upshift from 1581.5 cm−1 (pristine graphene) to 1597.5 cm−1 (after annealing Mn3O4- graphene nanocomposite); and the position of the 2D peak has upshift from 2667 cm−1 (pristine graphene) to 2686.5 cm−1 (after annealing Mn3O4-graphene nanocomposite), in Figure 7(b). These results indicate that Mn3O4 nanoparticles play a role in the p-doping dopant of single-layer graphene film. Generally, electronic property of single-layer graphene is strongly affected by surrounding environments in nanocomposite system. As discussed previously, silver nanoparticle deposition induced n-doping of graphene, whereas Au nanoparticle deposition induced p-doping due to the interaction between the metal and the graphene via an electron transfer driven by the work function difference.31 Graphene is doped with electrons if the work function of graphene larger than that of dopant nanoparticle and doped with holes in the opposite of condition. The work functions of graphene and bulk Mn3O4 are 4.24 eV and 4.40 eV, respectively.3233 So, our samples are doped by electron transfer from graphene to Mn3O4 after heat treatment.

Figure 7.Raman spectra of (a) single layer graphene, and (b) after calcined Mn3O4 nanoparticle on single-layer graphene: G peak and 2D peak.

 

Conclusion

facile synthesis route to Mn3O4-graphene nanocomposite has been developed. Mn3O4 nanoparticles with 5.47 ± 0.61 nm were decorated on single-layer graphene sheets with one layer thickness. After annealing, the Mn3O4 nanoparticles show the two-dimensional aggregation of nanoparticles, in particular. These results were confirmed through the AFM measurement. Also, the interaction between Mn3O4 nanoparticle and graphene was investigated by Raman spectra, and the shift of the G peak and the 2D peak has the nature of p-doping dopant on graphene. Furthermore, we hope that these results will help in the thin film application of graphene nanoparticle composite systems.