J. Phys. Chem. B 2001, 105, 7211-7215
7211
The Preparation and Structure of Platinum Metal Nanosheets between Graphite Layers Masayuki Shirai,* Koichi Igeta, and Masahiko Arai† Institute of Multidisciplinary Research for AdVanced Materials, Tohoku UniVersity, Sendai 980-8577, Japan ReceiVed: March 12, 2001; In Final Form: May 19, 2001
Platinum chloride-graphite intercalated compounds (PtCl4-GICs) with a stage three structure were produced by mixing platinum tetrachloride and graphite at 723 K under 0.3 MPa of chlorine atmosphere. Two-dimensional platinum nanosheets with a thickness of 2-3 nm and a width of 5-300 nm were formed between graphite layers by hydrogen reduction of PtCl4-GICs at 573 K. The sheets had a number of hexagonal holes and edge angles of 120°, owing to the limited aggregation of the intercalated platinum chloride molecules to platinum sheets along regular hexagonal nets of carbon atoms between graphite layers.
Introduction The production of metal particles having a unique structure is important because the structure (shape and electronic state) of metal particles is closely related to their surface properties. Loadings, reduction conditions, and structural features of supported metal catalysts determine the size and shape of metal particles formed. When metal particles are formed in pores of supports, the morphology of the metal particles becomes the same as that of the pores (a template method). For example, platinum nanowires are formed in the uniform mesotubes of folded-sheet mesoporous material FSM-16.1 One-dimensional metal clusters (platinum nanorods,2 iron nanoparticles,3 nickel nanowires,4 and copper nanoparticles5) are produced in the channels of carbon nanotubes. Graphite has a layered structure, and each layer is a regular hexagonal net of carbon atoms. The interlayer spacing is equal to 0.335 nm with the layers interacting via van der Waals forces. Because the interaction between graphite layers is fairly weak, a variety of chemical substances can be inserted into the interlayer space to produce graphite intercalated compounds (GICs).6 Transition metal particles intercalated in graphite layers (M-GICs) are formed via the insertion of transition metal chlorides into graphite and subsequent reduction.7,8 Commercially available M-GICs (M: Fe, Co, Cu, Ni, Pd) (graphimet) are obtained by treating the graphite-metal chloride with lithium biphenyl at 223 K under a helium atmosphere, and small metal particles (1-10 nm) are obtained.9-12 Graphimets have numerous applications for catalytic reactions.13-17 Walter et al. also reported that two-dimensional palladium nanoparticles encapsulated into graphite were obtained by reduction of palladium chloride-graphite intercalated compounds (PdCl2GICs) in a hydrogen atmosphere.18-22 Small or platelike metal particles were thus far obtained using the two-dimensional space of graphite layers because only the two-dimensional growth was permitted for metal particles by the steric hindrance of graphite layers. However, there are few reports on the formation of metal particles having an interaction with a regular hexagonal net of carbon atoms in graphite layers. * To whom correspondence should be addressed. Tel/Fax: +81-22-2175631. E-mail:
[email protected]. † Present address: Division of Materials Science and Engineering, Graduate School of Engineering, Hokkaido University, Sapporo 060-8628, Japan.
We have recently reported that platinum nanosheets are obtained between graphite layers by reduction of platinum tetrachloride-graphite intercalated compounds under hydrogen atmosphere.23 In the present work, we have studied in detail the insertion of platinum tetrachloride molecules into graphite layers, their reduction behavior, and also the formation and unique shape of platinum metal nanosheets containing hexagonal holes. Experimental Section Sample Preparation. Platinum tetrachloride and graphite (KS6, Lonza) were mixed in a thick-walled Pyrex reactor under nitrogen atmosphere, and the resultant sample (PtCl4-G mix) was dried in vacuo at 423 K for 2 h. The intercalation reaction was performed in the reactor at 723 K for 2 weeks under 0.3 MPa of chlorine (Takachiho, 99.999%) to obtain the platinum chloride intercalated compounds (PtCl4-GIC). The PtCl4-GIC and PtCl4-G mix samples were reduced at several temperatures (573 and 673 K) for 1 h under 30 kPa hydrogen to produce the platinum metal intercalated compounds (Pt-GIC) and the platinum metal particles on graphite (Pt-G mix). Characterization. X-ray Diffraction. The structure of the PtCl4-GIC samples prepared was determined by an X-ray diffraction method on a Shimadzu XD-D1 instrument (30 kV, 20 mA) using a Cu KR source. Temperature-Programmed Reduction (TPR). The stability of PtCl4-GIC and PtCl4-G mix was examined by heating a sample under hydrogen and monitoring the consumption of hydrogen with a thermal conductivity detector. The TPR experiments were performed with the same apparatus as described by Arai et al.24 A mixture of hydrogen (7.5 mL min-1) and argon (45 mL min-1) was used for the reduction. The temperature program in TPR experiments usually involved heating from 300 to 573 K at 10 K min-1, followed by evacuation for 2 h at 423 K. Transmission Electron Microscopy (TEM). TEM and X-ray microanalysis (XMA) measurements of the samples prepared were performed on JEOL JEM-2000ExII (200 keV) and JEM3010 (300 keV) instruments using copper grids. Results and Discussion The Insertion of Platinum Chloride into Graphite Layers. The diffraction peaks derived from platinum tetrachloride were
10.1021/jp010940l CCC: $20.00 © 2001 American Chemical Society Published on Web 07/07/2001
7212 J. Phys. Chem. B, Vol. 105, No. 30, 2001
Shirai et al.
Figure 2. The intensities of graphite (002) peak (b) and PtCl4-GIC peak at 10.0° (9) as a function of platinum loadings.
Figure 1. X-ray diffraction patterns of platinum tetrachloride (a) and PtCl4-GIC samples (Pt loading: 15 wt % (b), 10 wt % (c), and 5 wt % (d)).
not observed, and the peaks from graphite were observed in the X-ray diffraction (XRD) pattern for the 1 wt % PtCl4-GIC sample. The intensities of the diffraction peaks from graphite were smaller than those of the diffraction peaks from the graphite KS6 sample. The peaks ascribed to platinum tetrachloride and graphite were observed on XRD patterns of a reference mixture of platinum tetrachloride and graphite (PtCl4-G mix) (the platinum loading of this sample was 1 wt %). These results indicate that platinum chloride is well dispersed in the graphite matrix in the PtCl4-GIC sample. Figure 1 shows XRD patterns of 5-15 wt % PtCl4-GIC (platinum loading: 5-15 wt %), indicating that the peaks ascribed to platinum tetrachloride are not seen and the (002) diffraction peaks of graphite at 2θ ) 26.57° are weak but new peaks at 2θ ) 10.0°, 14.6°, and 20.3° can be seen. The positions of these three new diffraction peaks were almost the same for each of these 5-15 wt % PtCl4GIC samples. The diffraction peak positions calculated for (002), (003), and (004) reflections from the repeat distance along the c axis (c ) 1.76 nm) were 10.0°, 15.1°, and 20.2°, respectively, in good agreement with those of the three peaks observed. The distance of 1.76 nm corresponds to the sum of three graphite layers and one intercalated layer (0.75 nm). This result shows that platinum tetrachloride is intercalated in every three graphite layers (stage three structure) in the PtCl4-GIC samples. Figure 2 shows the intensities of a peak at 26.57° (graphite (002)) and a peak at 10.0° of the PtCl4-GIC samples as a function of platinum loadings. With an increasing amount of platinum tetrachloride inserted, the three peaks become stronger while the diffraction for (002) of graphite becomes weaker. No other peaks except for the three peaks and graphite peaks were observed. Although the three peaks at 10.0°, 15.1°, and 20.2° were not observed for the 1 wt % PtCl4-GIC sample, platinum chloride should be intercalated in graphite layers with the stage three structure, similar to the samples of higher Pt loadings. Platinum tetrachloride has a layered structure (Figure 3).25 The van der Waals distance of the layer (Cl-Pt-Cl distance) is 0.611 nm. The distance (0.75 nm) of the intercalated layers calculated from the third stage structure of the PtCl4-GIC samples is 0.14 nm larger than the layer distance of bulk
Figure 3. The structure of platinum tetrachloride. The configuration of [PtCl4/2Cl2] molecules is shown in part a, and a side view drawing of a layer is shown in part b.
platinum tetrachloride. X-ray photoelectron spectroscopy (XPS) analysis showed that the peak intensity ratios of Pt 4f and Cl 2p (Pt(4f)/Cl(2p)) for platinum tetrachloride, 5 wt % PtCl4GIC, and 5 wt % Pt-G mix were 1:3.3, 1:5.6, and 1:3.2, respectively, indicating that the amount of chlorine in the GIC sample was larger than that of platinum tetrachloride. The excess amounts of chlorine atom in the GIC samples may enlarge the intercalated layer by 0.14 nm. Taking into account the surface area of KS6 (20.5 m2 determined with a nitrogen adsorption method) and the molecular size of platinum tetrachloride, the maximum amount of platinum tetrachloride that can be intercalated between graphite layers with the stage three structure is 38 wt %. However, our experimental maximum was 15 wt %. Platinum chloride molecules would not be closely packed in graphite layers in the PtCl4-GIC samples. It is then assumed that the PtCl4-GIC samples have two domains; one is “pure” graphite matrix and the other PtCl4-intercalated graphite with the stage three structure. The latter should increase with an increase in the amount of PtCl4 inserted. The Stability of Platinum Chloride between Graphite Layers and on the Outer Surfaces. The three new peaks of PtCl4-GIC remained unaltered after 12 days of exposure to air; however, the platinum tetrachloride peaks of PtCl4-G mix changed after only a few days. The PtCl4-GIC samples were stable in air because platinum chloride molecules were intercalated between graphite layers and water molecules do not enter into graphite layers. Figure 4 shows TPR spectra for 5 wt % PtCl4-GIC and 5 wt % PtCl4-G mix samples. The 5 wt % PtCl4-GIC sample had a peak maximum at 470 K ascribable to the reduction of platinum chloride. The other sample showed peaks at 360 and 440 K. Thus, higher temperature is required to reduce platinum chloride intercalated between graphite layers, indicating that intercalated platinum chloride molecules are harder to reduce to metal particles than those on the graphite surface. The
Platinum Metal Nanosheets
J. Phys. Chem. B, Vol. 105, No. 30, 2001 7213
Figure 4. TPR spectra for 5 wt % PtCl4-GIC (a) and 5 wt % PtCl4-G mix (b) samples.
difference of the reduction temperature would explain their aggregation behavior during hydrogen reduction. Platinum chloride agglomerates to metal particles while interacting with both the top and bottom planes of two graphite layers in the PtCl4-GIC sample; however, platinum chloride is aggregated while interacting with only a graphite surface in the PtCl4-G mix sample. Excess amounts of hydrogen were consumed at lower temperatures in the PtCl4-G mix. Small amounts of hydrogen were consumed at 440 K to reduce the platinum tetrachloride molecules at the edge sites of graphite flakes in the 5 wt % PtCl4-G mix sample. The Structure of Platinum Metal Particles between Graphite Layers. The TEM images of the 5 wt % Pt-GIC sample reduced at 573 K showed the presence of dark images in a parallel rodlike arrangement and large sheets (Figure 5). Figure 5a shows the presence of dark images in a parallel rodlike arrangement, and Figure 5b,c shows the presence of large sheets with a number of hexagonal holes and edge angles of 120°. XMA scans obtained from regions a, b, and 1-3 showed that the dark material corresponded to platinum metal while those from regions c, d, and 4-6 indicated the absence of platinum metal (Figure 6). XMA results indicate that the dark features in the images correspond to platinum sheets. The rodlike image shown in Figure 5a is a side view and the sheet images shown in Figure 5b,c are top views of the structure of platinum nanosheets, because the graphite (KS6) used in this study has a powder form. The observation of a number of sheets in parallel arrangement presumably indicates that platinum sheets exist between graphite layers (Figure 5a). The platinum nanosheets observed were 2-3 nm thick, corresponding to 10 layers of platinum atoms. Platinum chloride can move and aggregate between a graphite layer during hydrogen reduction. The steric hindrance of graphite layers would cause the two-dimensional structures of platinum metal particles in the Pt-GIC sample. The platinum nanosheets have a number of hexagonal holes and edge angles of 120°. The edges of the hexagonal holes (AB) are straight and parallel to each other (Figure 7). The BC and AC edges are also parallel to each other. Moreover, the angles between two straight lines from three AB, BC, and AC directions are oriented at 120°. The structure of the graphite layers would reflect on the peculiar structure (holes and edges) of the platinum nanosheets. Platinum atoms or platinum chloride molecules or both would migrate along a regular hexagonal net of carbon atoms during hydrogen reduction at 573 K. Movement during the reduction will determine the morphology of the resulting platinum nanosheets, which have hexagonal holes. The TEM images would show such a metal-support interaction between platinum and graphite layers during the reduction of the PtCl4-GIC sample.
Figure 5. TEM images of 5 wt % Pt-GIC sample reduced at 573 K.
The TEM images of the 15 wt % Pt-GIC sample reduced at 573 K also showed the presence of platinum nanosheets which have a number of hexagonal holes and edge angles of 120° (Figure 8). Similar images of platinum sheets were observed in 10 wt % Pt-GIC samples reduced at 573 K. The size of platinum nanosheets did not depend on the amount of platinum intercalated in the Pt-GIC samples. Platinum nanosheets were also observed in the 5 wt % PtGIC sample reduced at a higher temperature of 673 K. The edge angles and holes of the platelike platinum particles became round, and many sheets did not have holes in the Pt-GIC sample reduced at 673 K (Figure 9). Although platinum sheets would be broken during 673 K reduction, the platinum particles still exist between graphite layers, so the particles were not spherical but sheetlike. For comparison, the 5 wt % PtCl4-G mix yielded spherical platinum particles upon reduction at 573 K (Figure 10). The size of the spherical particles was not uniform. For the PtCl4-G mix sample, platinum tetrachloride molecules can aggregate in a three-dimensional manner and there is no interfering factor
7214 J. Phys. Chem. B, Vol. 105, No. 30, 2001
Shirai et al.
Figure 8. TEM image of 15 wt % Pt-GIC sample reduced at 573 K.
Figure 9. TEM image of 5 wt % Pt-GIC sample reduced at 673 K.
Figure 6. XMA results of 5 wt % PtCl4-GIC sample at points a-d and 1-6 as indicated in Figure 5a,b.
Figure 10. TEM image of 5 wt % Pt-G mix sample reduced at 573 K.
materials would show interesting properties owing to their structural features and metal-graphite interactions compared with platinum particles on graphite layers. Conclusions
Figure 7. TEM image and drawing of platinum nanosheets.
for the growth of metal particles on the graphite surface and so large platinum particles can be formed. The structure of metal particles is closely related to their properties. Platinum nanosheets can be prepared between graphite layers for the Pt-GIC samples, and such complex
(1) Platinum chloride-graphite intercalated compounds (PtCl4GICs) with 1-15 wt % Pt loading were produced by mixing platinum tetrachloride and graphite under 0.3 MPa chloride at 723 K. (2) Platinum chloride is intercalated in graphite layers with the stage three structure for the PtCl4-GIC samples. (3) Platinum nanosheets with 2-3 nm thickness and 5-300 nm width were formed between graphite layers by hydrogen reduction of the PtCl4-GIC samples.
Platinum Metal Nanosheets (4) Platinum nanosheets have a number of hexagonal holes and edge angles of 120°. Acknowledgment. We thank Dr. E. Aoyagi and Dr. Y. Hayasaka (High Voltage Electron Microscope Laboratory of Tohoku University) for the TEM and XMA analysis. This work was partially supported by The Japan Securities Scholarship Foundation. References and Notes (1) Sasaki, M.; Higashimoto, N.; Fukuoka, A.; Ichikawa, M. Microporous Mesoporous Mater. 1998, 21, 597. (2) Kyotani, T.; Tsai, L.; Tomita, A. Chem. Commun. 1997, 701. (3) Pradhan, B. K.; Toba, T.; Kyotani, T.; Tomita, A. Chem. Mater. 1998, 10, 2510. (4) Pradhan, B. K.; Kyotani, T.; Tomita, A. Chem. Commun. 1999, 1317. (5) Chen, P.; Wu, X.; Lin, J.; Tan, K. L. J. Phys. Chem. B 1999, 103, 4559. (6) Herold, A. In Intercalated Layered Materials; Levy, F., Ed.; Physics and Chemistry of Materials with Layered Structures, Vol. 6; Reidel: Dordrecht, The Netherlands, 1979; p 323. (7) Volpin, M. E.; Novikov, Y. N.; Lapkina, N. D.; Kasatochkin, V. I.; Struchkov, Y. T.; Kazakov, M. E.; Stukan, R. A.; Povitskij, V. A.; Karimov, Y. S.; Zvarikina, A. V. J. Am. Chem. Soc. 1975, 97, 3366.
J. Phys. Chem. B, Vol. 105, No. 30, 2001 7215 (8) Tilquin, J.-Y.; Cote, R.; Veilleux, G.; Guay, D.; Dodelet, J. P.; Denes, G. Carbon 1995, 33, 1265. (9) Lalancette, J. M. U.S. Patent 3,847,963, November 12, 1974. (10) Smith, D. J.; Fisher, R. M.; Freeman, L. A. J. Catal. 1981, 72, 51. (11) Chii, S. I.; DeVera, A. L.; Hawley, M. C. Fuel 1986, 65, 1432. (12) Sirokman, G.; Mastalir, A.; Molnar, A.; Bartok, M.; Sachay, Z.; Guczi, L. J. Catal. 1989, 117, 558. (13) Furstner, A.; Hofer, F.; Weidmann, H. J. Catal. 1989, 118, 502. (14) Siroknam, G.; Mastalir, A.; Molnar, A.; Bartok, M.; Schay, Z.; Guczi, L. Carbon 1990, 28, 35. (15) Notheisz, F.; Mastalir, A.; Bartok, M. J. Catal. 1992, 134, 608. (16) Mastalir, A.; Notheisz, F.; Bartok, M.; Haraszti, T.; Kiraly, Z.; Dekany, I. Appl. Catal., A 1996, 144, 237. (17) Kiraly, Z.; Mastalir, A.; Berger, F.; Dekany, I. Langmuir 1997, 13, 465-468. (18) Walter, J. AdV. Mater. 2000, 12, 31. (19) Walter, J.; Shioyama, H. Phys. Lett. A 1999, 254, 65. (20) Walter, J. Philos. Mag. Lett. 2000, 80, 257. (21) Mendoza, D.; Morales, F.; Escudero, R.; Walter, J. J. Phys.: Condens. Matter 1999, 11, L317. (22) Walter, J.; Heuermann, J.; Dyker, G.; Hara, S.; Shioyama, H. J. Catal. 2000, 189, 449. (23) Shirai, M.; Igeta, K.; Arai, M. Chem. Commun. 2000, 623. (24) Arai, M.; Guo, S.-L.; Shirai, M.; Nishiyama, Y.; Torii, K. J. Catal. 1996, 161, 704. (25) Pilbrow, M. F. Chem. Commun. 1972, 270.
