Chapter 9 Controlled Synthesis and Growth of Perfect Platinum Nanocubes by Fasten Silicon
9.3 Structural characterization of Pt nanostructures
The synthesis and growth of Pt nanocubes were affected by physical and chemical conditions. For example, stirring time of the solution and dipping time of FS sample affects physically while the concentration of H2PtCl6, H2SO4 and HCl affect chemically. In our perfect Pt nanocubes synthesis, we optimized the concentration of HCl, H2PtCl6, H2SO4 and stirring time of aqueous solution.
Scanning electron microscopy (SEM) study was carried out to investigate the surface morphology of the synthesized Pt nanostructures onto the surface of fasten silicon (FS)
sample. Figure 9-2 (A to C), show the plane-view SEM images of the truncated Pt (cubes + tetrahedron), obtained by varying the dipping times 10 min, 13 min and 15 min respectively, in the aqueous solution of 2 M H2PtCl6 + 1 M H2SO4 at NTP. As shown in Fig. 9-2 (A to C), the Pt nanostructure sized is increased as the dipping time of FS sample increased. The Fig. 9-2 (C) clearly indicates that the products were mainly composed of the truncated Pt nanocubes with the truncated Pt tetrahedron, after 15 min dipping in an aqueous solution. The morphology of the truncated Pt tetrahedron nanostructures was further investigated by transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM). Figure 9-2 (D) shows a bright field TEM image of individual truncated Pt tetrahedron nanostructure. It can be clearly seen that the truncated Pt tetrahedron had a size smaller than ~80nm. Figure 9-2 (E) shows a HRTEM image taken from the corner of truncated Pt tetrahedron. The measured atomic spacing from the HRTEM image (Fig. 9-2 (E)) is ~0.23 nm, which corresponds to the set of {111} lattice plane. The above results confirm the truncated Pt tetrahedron grew along the {111} direction. The inset of Fig. 9-2 (E) shows the fast Fourier transform (FFT) image and confirmed that the truncated Pt tetrahedron is single crystalline.
In addition, to investigate the influence of stirred time on the shape of the Pt nanostructure, 5 h stirred solution (2 M H2PtCl6 + 1 M H2SO4) was used for the experiment. Figure 9-3 (A to C), display the plane-view SEM images of the truncated Pt nanocubes deposited on the FS sample in 13, 15, and 20 min, respectively. The obtained Pt nanostructured were 98%
truncated nanocubes (Fig. 9-3, A and B). In the case of 20 min dipping, the nanostructure obtained was well dispersed truncated Pt nanocubes with nanorods (Fig. 9-3 (C)). The morphological evolution of the nanocubes indicates that the larger nanocubes or nanorods originated because of the surface energy minimization and Ostwald repining (Fig. 9-5).
Figure 9-2 shows the SEM images of truncated Pt (cubes + tetrahedron) deposited on the FS sample at NTP for (A) 10, (B) 13, and (C) 15 min; (solution stirring for 3 hr). (D) Low-magnification TEM image of individual truncated Pt tetrahedron synthesized with the FS sample. (E) High-resolution TEM lattice image of the square region in single truncated Pt tetrahedron. The d-spacing between the fringes was ~0.23 nm, which was identified {111} plane of Pt; the inset shows the FFT of the lattice image gives the optical diffractogram, which clearly shows the presence of {111} plane.
100 nm 100 nm
100 nm
C
A B
D E
100 nm 100 nm
B A
1 μm
C
100 nm 100 nm
B A
1 μm
C
Figure 9-3 shows the SEM images of truncated Pt nanocubes deposited on the FS sample at NTP for (A) 13, (B) 15, and (C) 20 min; (solution stirring for 5 hr).
Figure 9-4 (A) Low-magnification TEM image of individual truncated Pt nanocubes and (B) High-resolution TEM lattice image of the square region in single truncated Pt nanocubes. The d-spacing between the fringes was ~0.20 nm, which was identified {200} plane of Pt.
20 nm
A B
Figure 9-5 Shows the SEM images of Pt nanostructures formed by immersing FS samples into aqueous solutions of H2SO4- H2PtCl6 for 20 min at NTP (stirring time = 5 h).
The images clearly show the presence of Pt nanocrystals with cube and rod shapes. The mechanism of rod growth is Ostwald ripening, define as the dissolution of the fine particles and their redeposition on large particles [198].
According to the well-known Gibbs–Thomson law, there is energy difference between large particles and small particles result in vanishing of smaller particles and formation of longer nanorods with the reaction in progress [199].
Figure 9-6 FFT image obtained from fig. 3E. The FFT pattern exhibits only a diffraction spots, revealing a good single-crystal property.
Figure 9-7 SAED image obtained from fig. 9-5 (A). SAED pattern also confirms the single crystal structure of the truncated Pt nanocubes.
As shown in Fig. 9-2 (C), when the stirring time was 3 h, a large number of the truncated Pt tetrahedron with truncated Pt nanocubes was formed. While on the other hand, when the stirring time was 5 h, only the Pt nanocubes were observed (Fig. 9-3(B)). This might be partly due to the homogeneity of aqueous solution controlled by stirred time. Figure 9-4 (A and B), shows the representative TEM and HRTEM images of an individual truncated Pt nanocube.
The size of truncated Pt nanocubes from TEM image was estimated ~80nm. The HRTEM image in Fig. 9-4 (B) recorded from a Fig. 9-3 (A) shows continuous lattice fringes with lattice spacing of ~0.20 nm, which corresponds to the {100} planes of Pt. The FFT image (Fig.
9-6) indicates that each truncated Pt nanocubes was a single crystalline. Furthermore, the selected area electron diffraction (SAED) pattern obtained by directing the electron beam perpendicular to individual truncated Pt nanocube (Fig. 9-7). The SAED pattern also confirmed that the truncated Pt nanocube was a single crystalline.
However, for the synthesis and growth of perfect Pt nanocubes, we used the same aqueous solution (2 M H2PtCl6 + 1 M H2SO4) mixed with 1 M HCl. To explore the synthesis and growth mechanism of the Pt nanocubes, again time-dependent experiments were carried out at NTP.
Figure 9-8 shows the SEM images of perfect Pt nanocubes deposited on the FS sample at NTP for (A) 10, (B) 13, and (C) 15 min; (solution stirring for 5 hr).
Figure 9-9 shows the typical XPS survey spectrum of perfect Pt nanocubes deposited on the surface of FS sample. XPS clearly show the presence of Pt nanocube catalyst.
100 nm 100 nm
100 nm
B
C
A
800 600 400 200 0
Si(2p) C(1s)
Pt(4d) Si(2s) Pt(4f) O(1s)
Rel at iv e XP S I nt en si ty
Binding Energy (eV)
Pt Nanocubes/Si
A series of SEM images taken at different dipping times (t) are presented in (Fig. 9-8, A to C). As shown in Fig. 9-8 (A), at t = 10 min, the Pt nanocubes were very hard to observed by SEM image. When t = 13 min, the Pt nanocubes were start to develop on the both side of the FS sample (Fig. 9-8 (B)). At t = 15 min, the Pt nanocubes were start to grow in all directions, which is considered as evidence for a growth of Pt nanocubes (Fig. 9-8 (C)).
The x-ray photoelectron spectroscopy (XPS) data were collected to further confirm the deposition of perfect Pt nanocubes onto the silicon. Figure 9-9 shows an XPS survey analysis of a Pt nanocube deposited onto a FS sample. Two distinct Pt (4f7/2) and Pt (4f5/2) peaks at 71.0 and 74.2 eV, respectively, were clearly observed (Fig. 9-10). These are typical values for zero valent Pt, [193] indicating that the perfect Pt nanocubes were zero valence. Furthermore, no obvious shoulders at higher binding energies, representing Pt2+ and Pt4+, were found, and no peek of chlorine from the Pt precursor was
Figure 9-10 XPS deconvoluted Pt(4f) doublet of perfect Pt nanocubes deposited on the surface of FS sample. The peaks can be attributed to Pt(4f7/2) and Pt(4f5/2) of metallic Pt, respectively.
68 70 72 74 76 78
Pt(0)4f
5/2Pt(0)4f
7/2Rel at iv e XP S I nt en si ty
Binding Energy (eV)
Figure 9-11 Cl(2p) XPS narrow scan spectra of the perfect Pt nanocubes deposited on the surface of FS sample. This figure clearly shows the complete absence of chloride ion at the perfect Pt nanocubes.
detected by XPS (Fig. 9-11). From XPS analyses, we conclude that the Pt nanocubes were composed of metallic platinum. The morphology of the perfect Pt nanocubes was further investigated by HRTEM. Figure 9-12 (A) shows the HRTEM images of individual Pt nanocube. From the TEM image, it is clear that the Pt nanocubes have perfect shape, with a sized of ~10 nm. Figure 9-12 (B) displays a representative electron diffraction pattern recorded by directing the electron beam perpendicular to the {100} facets of an individual nanocube and confirms that the platinum particles are single crystals. The FFT of the atomic lattice fringing apparent in the inset of Fig. 9-12 (A), corresponding to the perfect individual platinum nanocube, further demonstrating the single crystallinity of the nanocubes. All of these results also confirm the perfect platinum nanocubes grew along the {100} direction. We also found that the nanocubes are formed as a result of fast growth along {111} directions, and the surfaces of the final nanocubes correspond to {100} planes [194].
195 196 197 198 199 200 201 202 Cl(2p)
Rel at iv e XPS I nt en si ty
Binding Energy (eV)
Figure 9-12 (A) High-resolution TEM image of individual perfect Pt nanocube. The d-spacing between the fringes was ~0.20 nm, which was identified {200} plane of Pt; the inset shows the FFT of the lattice image gives the optical diffractogram, which clearly shows the presence of {200} plane and (B) Corresponding SAED pattern, showing the single crystal structure of the perfect Pt nanocubes.
To compare the geometrical structure of Pt, kinetic effect and the effect of fasten pair silicon, Pt nanoparticles were also deposited on the single Si substrate by electroless deposition. The etched native SiO2 layer from the Si sample dipped in a solution of 1 M HCl + 1 M H2PtCl6 + 1 M H2SO4, prepared by vigorously stirred for 5 h at NTP. Figure 9-13 (A and B) shows the growth of Pt nanoparticles on the single silicon substrate with an immersion time of 6 h.The deposition times were too longer so the kinetic factor has negligible effect on the synthesis and growth of Pt nanoparticles. The Fig. 9-13 (A and B) also conformed that the Pt nanoparticles were nanospherical.This is might be due to low charge density on the single silicon wafer compare to that of a pair of FS sample when they were in very close proximity in aqueous solution. The charges were developed on the silicon surface by etching with HF.
A B
Figure 9-13 Plane views SEM images of the silicon after electroless deposition of Pt for 6 h.
Compositions of solutions were (A) 2 M H2PtCl6 + 1 M H2SO4 + 1 HCl and (B) 2 M H2PtCl6 + 1 M H2SO4. Both the aqueous solution was stirred continuously for 5 h.
A
B
The idea originated from faraday’s first law of electrolysis, the mass of a substance altered at an electrode during electrolysis is directly proportional to the quantity of charge transferred at that electrode. The negative potential and the electrostatic force of interaction developed in between two silicon wafer is comparatively higher than that of the single silicon wafer resulting in fast deposition of Pt nanoparticles on the FS sample.
From the results above, we can conclude that the separate role played by 1 M HCl and 1 M H2SO4 aqueous solution. In the absence of HCl in aqueous solution of 2 M H2PtCl6 +1 M H2SO4, SO4– ions from H2SO4 have a negligible effect on the synthesis of Pt nanocubes because the concentration of H2SO4 was lower compared to the concentration of H2PtCl6. Figure 9-3 (C) shows a typical SEM image of the truncated Pt nanocubes at t = 20 min, that consisted a mixture of small Pt cubes and rods. This observation implies that sulfuric acid alone is able to induce the etching and dissolution of seeds and thus channel the product into truncated single crystal nanocubes. The potential reasons for occurrence a truncated shape may be due to enough H+ and Cl– ions needed for etching and dissolutions. On the other hand, when HCl was added to the aqueous solution of 2 M H2PtCl6 +1 M H2SO4, Cl– ions from HCl should have a strong effects on the synthesis of perfect Pt nanocubes because chloride ions can reduce the surface energies of the {100} facets of the seeds by binding strongly to them, thus leading to the formation of perfect nanocubes [195, 196]. The presence of Cl– ions in the aqueous solution stabilize the single crystal and therefore Cl– ions also can prevent the nanocubes from aggregating and sedimentation by providing electrostatic repulsion forces between the Pt nanocubes [197]. Moreover, we found that the morphology and dimensions of Pt nanocubes were greatly influenced by concentration and homogeneity of solution, dipping time and substrate surface energy. For example, the number of Pt nanocubes formed is proportional to the concentration of solution. With increase in the homogeneity of solution >
98% cubes formed were sharp edge. Prolong dipping times of FS sample results in formation of cubes with rods. Surface energy accelerates the deposition of Pt cubes as evident in FS
surface.
9.4 Electrocatalytic activity
The synthesis of sizes, shapes and structure control of platinum nanoparticles allows us to study surface dependent properties such as catalytic activity. Figure 9-15 (A and B) shows cyclic voltammograms for methanol and ethanol oxidation on the perfect Pt nanocubes (Fig.
9-8 (B)), the truncated Pt nanocubes (Fig. 9-3 (B), the truncated Pt (cubes + tetrahedron) (Fig.
9-2 (C)), and the spherical Pt nanoparticles (Fig. 9-14) at NTP. As shown in Fig. 9-15 (A and B), the onset potential for methanol and ethanol electrooxidation of the perfect Pt nanocubes was lower than that of truncated Pt nanocubes, truncated Pt (cubes + tetrahedron), and spherical Pt nanoparticles, indicating the faster electrodes kinetics [174]. In addition, the perfect Pt nanocubes show a peak current density is higher than that of truncated Pt nanocubes, truncated Pt (cubes + tetrahedron), and spherical Pt nanoparticles.
Furthermore, the chronoamperometry study was carried out to determine the stability of electrocatalytic activity toward methanol oxidation reduction and ethanol oxidation reduction.
Figure 9-15 (C and D) shows the chronoamperograms of the four electrodes, which were obtained by measuring the steady-state reaction current density at the electrode potential of 0.4 V. After the polarization of 200 s, the current density of methanol electrooxidation for the perfect Pt nanocubes, the truncated Pt nanocubes, the truncated Pt (cubes + tetrahedron), and the spherical Pt nanoparticles are ~1.45, ~0.29, ~0.18, and ~0.16 mA/cm2 respectively. The perfect Pt nanocubes indicate much higher activity than the truncated Pt nanocubes, truncated Pt (cubes + tetrahedron), and spherical Pt nanoparticles, whose steady-state currents are over five times, eight times and nine times higher than that of truncated Pt nanocubes, truncated Pt (cubes + tetrahedron), and spherical Pt nanoparticles, respectively.
Figure 9-14 (A) FE-SEM image of the electrodeposited polycrystalline Pt nanosphere on silicon substrate. (B) HRTEM image recorded from a region at the edge of a sphere, showing that the surfaces are covered by Pt nanoparticles of a few nanometers in size, and the Pt nanocrystallites dominated by {111}-type exposed surface. (C) SAED pattern showing that the formed nanoparticles are polycrystalline. The as-labelled diffraction rings 1–4 correspond to {111}, {200}, {220}, and {311} reflections, respectively, all of which reveal the fcc crystal structure.
(B) (C)
(A)
Figure 9-15 Comparison of catalytic activity and stability of the Pt nanoshapes. (A) and (B) Cyclic voltammograms for (a) perfect Pt nanocubes, (b) truncated Pt nanocubes, (c) truncated Pt (cubes + tetrahedron), and (d) spherical Pt nanoparticles were obtained at a scan rate 25 mVs-1 in 1 M H2SO4 + 1 M methanol and 1 M H2SO4 + 1 M ethanol respectively. (C) and (D) Chronoamperometric curves for (a) perfect Pt nanocubes, (b) truncated Pt nanocubes, (c) truncated Pt (cubes + tetrahedron), and (d) spherical Pt nanoparticles were recorded at 0.40 V in oxygen free 1 M H2SO4 + 1 M methanol and 1 M H2SO4 + 1 M ethanol, respectively. The results demonstrate that the perfect Pt nanocube electrode showed excellent catalytic activity and stability for oxidation of methanol and ethanol.
On the other hand, the current density of ethanol electrooxidation for the perfect Pt nanocubes, the truncated Pt nanocubes, the truncated Pt (cubes + tetrahedron), and the spherical Pt nanoparticles are ~2.01, ~0.66, ~0.27, and ~0.016 mA cm-2 respectively. Again, the perfect Pt nanocubes indicate much higher activity than that of truncated Pt nanocubes,
truncated Pt (cubes + tetrahedron), and spherical Pt nanoparticles, whose steady-state currents are over three times, seven times, and 125 times higher than that of truncated Pt nanocubes, truncated Pt (cubes + tetrahedron), and spherical Pt nanoparticles, respectively. The above result implies that the perfect Pt nanocubes have a better electrocatalytic toward both methanol and ethanol electrooxidation, which are promising alternative fuels for liquid feed fuel cells.
9.5 Summary
We have demonstrated, for the first time simple and efficient method for the synthesis and growth of well-dispersed perfect platinum nanocubes, employing a pair of low-resistivity fasten silicon (FS) sample at normal temperature and pressure (NTP) in the absence of surfactant, additives and capping materials and so on. The perfect Pt nanocubes deposited on the FS sample exhibited an excellent catalytic activity and stability for methanol and ethanol oxidation, as compared to the truncated Pt nanocubes, the truncated Pt (cubes + tetrahedron) and the spherical Pt nanoparticles. We believed that these new and simple concepts will open the door for the synthesis of the well-controlled shape, sizes and facets of not only Pt nanocubes but also other metal as well (for instance silver, gold and copper), for optical, electrical, thermal, magnetic and catalytic applications. Study of the mechanism behind the origin of perfect Pt nanocubes from fasten silicon wafer is still underway in our laboratory.
Chapter 10
Conclusions and Future Works 10.1 Conclusions
In this study, ordered arrays of nanostructures including Pt/Si nanocone, and Pt/α-C/Si nanocone are successfully fabricated by using the nanoporous anodic aluminum oxide (AAO) membrane as a template. Well-ordered nanomask arrays of TiOX are constructed by anodizing the Al/TiN bilayered films. The TiOX nanodots were then used as the nanomask for etching the TiN layer and the underlying layer in an inductively coupled plasma reactive-ion-etch (ICP-RIE) system. On the other hand, the Pt nanostrucutred materials were synthesized by direct electrochemical deposition and fasten silicon. In addition, we have synthesized the nanoporous graphitic carbon (g-C), which was utilized as a support for Pt-Ru alloy catalysts.
The electro-catalytic characteristics of the nanostructures have been investigated. The primary results of this thesis are summarized as follows:
(1) We have fabricated well-ordered Si nanocones (SNCs) using AAO templation, and electrodeposited Pt nanoparticles on the SNC arrays by potentiostatic pulse electrodeposition. The pulse electrodeposition led to the formation of well-dispersed Pt nanoparticles on the SNCs. The Pt nanoparticles were predominantly spherical in shape and had a size of < 5 nm. Because of uniform dispersion of Pt nanoparticles and the high surface area, the Pt-SNC electrode exhibited superior electrocatalytic properties toward the methanol electro-oxidation, with the onset potential of 0.08 V (vs. SCE). The catalyst mass activity was several times higher than that of the blanket Pt film/flat Si. Moreover, chronoamperometric analyses and CO stripping cyclic voltammetric (CV) study indicated that the Pt-SNC electrode had a stable electrooxidation activity with a very good CO tolerance. The Si surface oxide surrounding the Pt nanoparticles on the SNCs was
suggested to play a key role in improving the CO tolerance via the bifunctional mechanism.
(2) This study pulse-electrodeposited Pt nanoparticles on amorphous carbon coated silicon nanocones (ACNCs) and explored them as the electrocatalyst for methanol oxidation reaction (MOR) and oxygen reduction reaction (ORR) for direct methanol fuel cell applications. The work prepared silicon nanocones on the Si wafer using porous anodic aluminum oxide as the template and then deposited the amorphous carbon layer on the nanocones by microwave plasma chemical vapor deposition. According to Raman scattering and x-ray photoelectron spectroscopies, the surface of the ACNC support is composed of a nanocrystalline graphitic structure, and rich in oxygen containing adspecies. The Pt nanoparticles pulse-electrodeposited on the highly ordered ACNC support dispersed well with a large electrocatalytic surface area. The Pt-ACNC electrode exhibited excellent electrocatalytic activity and stability toward both MOR and ORR.
This study suggests the abundant oxygen containing surface species and the nanocrystalline graphitic structure as the two major factors enhancing electrocatalytic performance of Pt catalyst nanoparticles.
(3) A three-dimensional (3D) nanoporous graphitic carbon (g-C) was synthesized by flame of adamantane (C10H16), which was utilized as the support for Pt50–Ru50 alloy catalyst. The electrochemical investigations show the Pt50-Ru50 supported on 3D nanoporous g-C have excellent activity towards methanol oxidation, and better CO tolerance due to the presence of Ru nanoparticles, in which Ru was proposed to be able to promote the oxidation of the strongly adsorbed CO on Pt by supplying an oxygen source (Ru-OHad).
Moreover, the presence of 3D nanopores in g-C supports may also contribute to the observed higher current density because of the easy transport of methanol and the
oxidation products in these nanopores.
(4) Two-dimensional (2D) continuous Pt island network was successfully synthesized by pulse-potentiostatic electrodeposition on the flat silicon substrate, and electrochemical investigations confirm that this catalyst structure on the silicon substrate has a better electroactivity toward methanol oxidation than the blanket Pt/Si and the Ru decorated Pt film/Si electrodes. The electrooxidation performance is better than that of the other two electrodes, due to the synergistic effect of the Pt island catalyst and surrounding SiO2
(4) Two-dimensional (2D) continuous Pt island network was successfully synthesized by pulse-potentiostatic electrodeposition on the flat silicon substrate, and electrochemical investigations confirm that this catalyst structure on the silicon substrate has a better electroactivity toward methanol oxidation than the blanket Pt/Si and the Ru decorated Pt film/Si electrodes. The electrooxidation performance is better than that of the other two electrodes, due to the synergistic effect of the Pt island catalyst and surrounding SiO2