SYNTHESIS OF AMORPHOUS SiO X NANOSTRUCTURES. 1. Introduction

Transcription

1 SYNTHESIS OF AMORPHOUS SiO X NANOSTRUCTURES J. Y. LAO, J. G. WEN, D. Z. WANG, Z. F. REN Department of Physics, Boston College, 140 Commonwealth Avenue Chestnut Hill, Massachusetts 02467, United States of America Various amorphous SiO x nanotube structures nucleated by GeO x nanoparticles were synthesized by thermal evaporation method. The presence of Ge does not only nucleate the growth of the SiO x nanomaterials, but also dopes them. The nanostructure morphology is affected by the substrate temperature, source temperature and GeO x vapor density through their effect on the size and lifetime of the nucleation center. In general, low substrate temperature promotes the formation of the nanotube bundle structure with 2--3% atomic ratio of Ge doping, and high temperature produces Ge-free much less bundled nanotubes. Keywords: Silicon oxide; Nanotube; Germanium; Amorphous. 1. Introduction The discovery of carbon nanotubes 1 has stimulated great interest in synthesizing various nanostructure materials such as BN nanotubes, 2,3 and GaAs, InAs, 4 GaN, 5 GeO 2, 6 and silicon 7 nanowires as well as various carbide nanowires. 8 These one-dimensional nanomaterials have novel physical and chemical properties due to the nanoscale size, 4,9,10 and offer great potential for applications such as nanodevices, mesoscopic research, etc. For silicon oxide, researchers at Mobil first successfully synthesized the mesoporous silicate molecular sieves (MCM-41) for the application in catalysis. 11 Silica mesoporous material and amorphous nanowires synthesized using Fe catalyst showed strong photoluminescence properties and have potential applications in high-resolution optical heads of scanning near-field optical microscopy or nano-interconnection in integrated optical device. 12,13 Zhu et al. has grown silica nanowire flowers using Co as the catalyst. 14 Satishkumar et al. prepared silica nanotubes by using carbon nanotube as template. 15 Wang et al. has grown silicon trapped silica nanotubes by pyrolysis of a mixture of Si and SiO Ge doping in silica glass can modify the silica optical properties in both the visible and UV regions of the spectrum and can be used in opto-electronics and fiber optics to generate refraction index grating. 17 Ge nanocrystals in SiO 2 films have shown interesting visible photoluminescence, and its microstructure and photoluminescence strongly depends on the annealing conditions But there has no report about Ge doping of silica nanostructure and study of temperature effect on the formation of SiO x nanostructure. Here we report the synthesis of various interesting SiO x amorphous nanotube structures by using a low melting point GeO x as the nucleation site. GeO x does not only nucleate the SiO x nanotube structures, but also dopes the nanotube walls slightly. We als o studied the synthesis temperature effect on nanotube morphology and Ge doping level in these novel Ge -doped amorphous SiO x nanotube structures.

2 GeO2 Powder Si Wafer Quartz Tube Source Heating Element Coating Substrate Vacuum Tube Furnace Heating Element Fig. 1 Schematic drawing for growth of the SiO x nanostructures. The substrate temperature is controlled by its location in the tube furnace. 2. Experimental The experimental set up is schematically illustrated in Fig. 1. GeO 2 powder was put at the sealed end of one end sealed quartz tube. Silicon wafer pieces were placed next to the GeO 2 powder as Si source. Fresh silicon wafer instead of silicon powder is used due to its easy handling. Then the opening end of the quartz tube was covered by a graphite foil as substrate. The whole assembly was put into the ceramic tube furnace, pumped to below 1.0x10-1 Torr, and heated at high temperatures for two to three hours. The evaporation source temperature is maintained at 1100 o C except noted. The graphite foil substrate temperature is controlled by its location in the ceramic tube due to the temperature gradient along the ceramic tube. The GeO x and Si vapor produced from the heated source will deposit on the graphite foil substrate due to the temperature gradient and the pumping direction. After the experiment, fine powder was usually found on the graphite foil surface facing the vapor indicated by coating in Fig. 1. The powder was analyzed using a JEOL JSM -6340F scanning electron microscope (SEM), a Bruker Analytical X- ray System, and a JEOL 2010 transmission electron microscope (TEM). 3. Results The morphology of the synthesized SiO x nanomaterial is greatly affected by the substrate temperature. Fig. 2 and Fig. 3 show the SEM images and corresponding TEM images of the SiO x nanostructures synthesized at the substrate temperature of 700, 860, 960 and 1100 o C, respectively. Initially we put the substrate in a very low temperature zone. There is no deposition observed on the graphite foil surface because the substrate is too far away from the source. Instead, many aligned nanotube flowers are formed on the inner wall of the quartz tube where the temperature is around 700 o C. As shown in Fig. 2a, the nanotubes on the quartz tube inner wall formed into one directional bundle with a central 2

3 Fig. 2 SEM images of the SiO x nanostructures grown at substrate temperatures of a) 700 o C, b) 860 o C, c) 960 o C, and d) 1100 o C. particle at the top of each bundle facing the vapor coming direction. These nanotubes are only several microns long and the nanotube diameter varies along the tube from 100 nm at the bottom to a few tens nanometers at the tip. TEM image at Fig. 3a shows that the nanotubes can be either internal bamboo-like hollowed or straight-hollowed structures. Energy dispersive x-ray spectroscopy (EDS) and TEM shows that the central particles of the nanotube flowers are Ge nanocrystal, as discussed later. At substrate temperature of 860 o C, two kinds of Ge -doped SiOx nanotube were obtained. One is SiO x nanotube bundles, shown in Figs. 2b and 3b, formed on the center of graphite foil with structures similar to that formed at 700 o C. The other is branched SiO x bamboo-like nanotubes with longer length and big diameter formed at the edge of the graphite foil, as shown in Figs. 4a and 4b. Compared to the structures formed at 700 o C, the centers of the nanotube flowers in Fig. 2b are holes, instead of particles. Probably, these holes are the footprint of Ge particles after evaporation. TEM image of Fig. 3b shows that the tube structure is similar to Fig. 3a. The black and white contrast of 3

4 Fig. 3 TEM images of the SiO x nanostructures grown at substrate temperatures of a) 700 o C, b) 860 o C, c) 960 o C, and d) 1100 o C. Fig. 4 a) SEM image of the SiO x nanotubes grown at 860 o C at the edge of graphite foil substrate. The insert is the higher magnification SEM image of the branches. b) TEM image of the tube structure. the tubes in Fig. 4a is due to the internal bamboo-like structure. The insert of Fig. 4a shows the branched tubes. These branched tubes are curved with diameters from 100 nm to 150 nm. TEM image in Fig. 4b shows that the inside of the nanotubes are again bamboo-like hollowed. The structures formed in the 960 o C zone are also different from that above. Fig. 2c shows the mixture of big SiO x bundles and some branched tubes. The structures formed at the edge of graphite foil substrate have more branched nanotubes. The big bundles have a diameter of 100 to 400 nm, and the branched nanotubes have a diameter of 50 to 200 nm. The big bundles are in fact adhered bundles of small wires, and the small ones are hollowed tubes, as shown in Fig. 3c. 4

5 Fig. 5 EDS of the wall of SiO x nanotubes grown at 700 o C, which shows around 2.3% atomic ratio of Ge doping. With increasing substrate temperature (1100 o C), although some 100 nm diameter branched SiO x nanotubes were also obtained as shown in Figs. 2d and 3d, but the quantity is much less. TEM images show that the tubes are hollowed structures with very small hollowed centers. We have also tried 1160 o C for both the source and substrate temperature, but no growth has resulted. EDS shows that the atomic ratio of O to Si of these nanotubes is close to two, which has been calibrated by pure quartz with known composition of SiO 2. The Ge doping level on the nanotube walls decreases with increasing temperature. At 700 o C, the doping content of Ge in the nanotube wall is around 2--3% atomic ratio as shown in Fig. 5 whereas no Ge is detectable for structures formed at 1100 o C. The effect of GeO 2 and Si source temperature has also been studied while the substrate temperature is fixed at 800 o C zone. It is found that 1050 o C source temperature also gives oriented SiO x nanotube flowers, but the nucleation center is much smaller and only five to ten tubes grow from each nucleation center. If the source temperature is over 1150 o C, instead of nanostructures, big Si-Ge-O chunks are formed on the graphite foil surface. In addition, we have also tried to increase Ge concentration in the nanostructure by adding graphite powder into the GeO 2 powder. In this case, graphite powder will reduce GeO 2 into lower melting point Ge, so that the vapor concentration of Ge will be higher. On the other hand, Si source is moved to the zone where the temperature is about 100 o C lower than that of GeO 2 powder to reduce the SiO x deposition rate, with a higher Ge/Si ratio in the material formed as an expected result. However, we found that the source temperature has to be lowered to 1050 o C because high temperature gives Si-Ge -O coarse powder deposition on the graphite foil. Therefore, we did not see much improvement of Ge doping level on the nanostructure walls by this method. But interestingly, besides the 5

6 Fig. 6 TEM images of the SiO x nanotubes grown with graphite powder added into the GeO 2 source. a) Typical image, b) one of the small branches, c) holes on the tube wall, and d) Ge nanocrystal enclosed in the tube end. The insert is the SAD pattern of the Ge nanocrystal. oriented nanotube flowers, we also found some different SiO x tube structures, as shown in Fig. 6a. These well-defined tubes have diameters of 50 nm to 1000 nm. Some tubes have tapered ends as indicated by the arrows. Only a few of the tubes have branches, and these branches are very short, as shown in Fig. 6b. Fig. 6c shows the very obvious round holes on the tube wall, as marked by arrow. The round clear contrast in the hole is the Fresnel fringe formed in the TEM observation, and the gray contrast is the projection of the layer of tube wall opposite to the hole. We postulate that the holes are the footprint of evaporated small Ge particles. The Ge vapor pressure in this setup is high, so the deposited Ge atoms can segregate out to form Ge particles. The left holes are difficult to be repaired due to low mobility of Si atoms at this temperature range. So these tube internals are connected to the outside by the holes, which is good for catalyst substrate application. Also, some nanocrystals were found at the end of some tubes, as shown in Fig. 6d. EDS spectrum shows that the particle is Ge. The inserted selected-area diffraction (SAD) pattern shows that the Ge exists inside the SiO x nanotubes as single crystal. The Ge nanocrystal could be the nucleation point for the SiO x nanotube growth. It is covered by SiO x layer when the growth terminates. 6

7 Fig. 7 TEM image showing the Ge central particles (black spots) for the growth of the SiO x nanostructures at substrate temperature of 700 o C. 4. Discussions The geometry of the structure of Fig. 2a indicates that the aligned nanotube flowers were probably nucleated from the central particles. Fig. 7 is the TEM image of the SiO x nanotube bundle central particles, the related EDS spectrum shows that the black particles are oxygen free Ge particles. These Ge particles must come from GeO x vapor which is probably reduced by Si vapor at high temperature. Fig. 2b also indicates that the nanotube bundle growth was started from the center holes, which are probably the footprint of evaporated Ge central particles. The geometry of other less dense nanotube structures also shows a common center, although the centers are not Ge rich anymore after the reaction, as measured by EDS. From these results, we believe that the growth of SiO x nanostructure observed here is based on the vapor-liquid-solid (VLS) mechanism. 24 At 1100 o C source temperature, the GeO 2 powder with melting point of 1116 o C will produce significant amount of GeO x vapor which reacts with the Si vapor. The Si reduced GeO x vapor can deposit on lower temperature graphite foil substrate as melting or semi-melting droplets and become the nucleation source. Then the upcoming SiO x vapor can diffuse into the GeO x nucleation particles and segregate out to form SiO x nanostructures. At low substrate temperature such as 700 o C the growth model of SiO x nanotube bundles should be similar to what was proposed by Zhu et al. for the explanation of SiO x nanowire flower growth from Co particles. 14 The formed GeO x droplets have big size and comparably longer lifetime at low substrate temperature. Blisters on the surface of big GeO x droplets serve as the nucleation sites for the SiO x nanotube growth and stay at the tip of nanotube during the 7

8 growth. On the other hand, since those nucleation sites are from the same big GeO x droplet, the bottoms of the formed SiO x nanotubes are also on the same big droplet (nanotube bundle center). The difference between here and the literature 14 is that the GeO x particles at the tip of nanotubes will be finally evaporated so those nanotubes have sharp tips, but the big GeO x droplets at center will probably stay and are reduced into Ge. At high substrate temperature such as 960 and 1100 o C, GeO x droplets formed from the vapor are smaller in size and shorter lifetime compared to that of low temperature, so the SiO x nanotubes are less bundled and more difficult to form. The difference of structures formed between the edge and the center of the graphite foil substrate at 860 and 960 o C substrate temperature is due to the influence of nucleation material vapor aerodynamics in this particular setup. The vapor at the edge is easier to be pumped away, so instead of big particles, only small particles with short lifetime can be formed as nucleation centers which can only support a few nanotubes to grow. In the case of 1100 o C substrate temperature, not much difference was found between the structures formed at the edge and the center. Probably, this temperature is already high enough to prevent the formation of any big nucleation center. On the other hand, higher source temperature such as 1150 o C gives too high deposition rate and Si-Ge-O big chunks are formed. The formation of SiO x nanotube branches from main stem as shown in Fig. 4b and 6b could be due to the splitting of GeO x nucleation source on the side-wall of the tube. Because of the low melting point of GeO x particles and associated particle size effect on melting points, very small GeO x particles can not be formed in large amount, so thin SiO x nanowires are difficult to form in large amount by this method. In fact, based on the effect of the size of nucleation on synthesized nanostructures, we have also grown SiO x nanostructures from nm diameter long amorphous nanowires to similar short aligned SiO x nanotube bundles by using Ag paint as the nucleation source. Similarly, it is also found that big Ag particle size favors the growth of nanotube bundles, and the small Ag particle size favors the growth of thin wires. Comparing to the commonly used catalysts such as Fe, Co and Ni, the formation of SiO x nanotube from GeO x and Ag nucleation centers is probably due to their relatively low melting temperature. 5. Conclusions In summary, different morphology of slightly Ge -doped SiO x nanotubes were synthesized by thermal evaporation method. The nanotube morphology and concentration can be controlled by its substrate temperature and GeO x vapor density, and Ge doping concentration also varies with the substrate temperature. The choice of right nucleation center size is necessary for the formation of right nanostructures. These materials have potential applications in opto-electronics and catalysis. 8

9 Acknowledgments This work is partly supported by DoE under a grant DE-FG02-00ER45805, partly by The US Army Natick Soldier Systems Center under a grant DAAD16-00-C-9227, and partly by NSF under a grant ECS References 1. S. Iijima, Nature (London) 354, 56 (1991). 2. N.G. Chopra, R.J. Luyken, K. Cherrey, V.H. Crespi, M.L. Cohen, S.G. Louie and A. Zettl, Science 269, 966 (1995). 3. M. Terrones, W.K. Hsu, H. Terrones, J.P. Zhang, S. Romas, J.P. Hare, R. Castillo, K. Prassides, A.K. Cheetham, H.W. Kroto and D.R.M. Walton, Chem. Phy. Lett. 259, 568 (1996). 4. K. Hiruma, M. Yazawa, T. Katsuyama, K. Okawa, K. Haraguchi, M. Kogucchi, H. Kakibabyashi, J. Appl. Phys. 77, 447 (1995). 5. W. Han, S. Fan, Q. Li, Y. Hu, Science 277, 1287 (1997). 6. Z.G. Bai, D.P. Yu, H.Z. Zhang, Y. Ding, Y.P. Wang, X.Z. Gai, Q.L. Hang, C.C. Xiong, S.Q. Feng, Chem. Phys. Lett. 303, 311 (1999). 7. A.M. Morales, C.M. Lieber, Science 279, 208 (1998). 8. H. Dai, E.W. Wang, Y.Z. Lu, S.S. Fang, C.M. Lieber, Nature (London) 375, 769 (1995). 9. M.M.J. Treacy, T.W. Ebbesen, J.M. Gibson, Nature (London) 381, 678 (1996). 10. M.F. Yu, O. Lourie, M.J. Dyer, K. Moloni, T.F. Kelly, R.S. Ruoff, Science 287, 637 (2000). 11. C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature (London) 359, 710 (1992). 12. H.J. Chang, Y.F. Chen, H.P. Lin, C.Y. Mou, Appl. Phys. Lett. 78, 3791 (2001). 13. D.P. Yu, Q.L. Houg, Y. Ding, H.Z. Zhang, Z.G. Bai, J.J. Wang, Y.H. Zhou, W. Qian, G.C. Xiong, S.Q. Feng, Appl. Phys. Lett. 73, 3076 (1998). 14. Y.Q. Zhu, W.K. Hsu, M. Terrone, N. Grobert, H. Terrones, J.P. Hare, H.W. Kroto, D.R. Walton, J. Mater. Chem. 8(8), 1859 (1998). 15. B.C. Satishkumar, A. Govindaraj, E.M. Vogl, L. Basumallick, C.N.R. Rao, J. Mater. Res. 12, 604 (1997). 16. Z.L. Wang, R.P. Gao, J.L. Gole, and J.D. Stout, Adv. Mater. 12, 1938 (2000). 17. K.O. Hill, Y. Fujii, D.C. Johnson, B.S. Kawasaki, Appl. Phys. Lett. 32, 647 (1978). 18. M. Fuji, S. Hayashi, and K. Yamamoyo, Jpn. J. Appl. Phys. 30, 687 (1991). 19. Y. Maeda, N. Tsukamoto, Y. Yazawa, Y. Kanemitsu, and Y. Masumoto, Appl. Phys. Lett. 59, 3168 (1991). 20. J.Y. Zhang, X.M. Bao, Y.H. Ye, and X.L. Tan, Appl. Phys. Lett. 73, 1790 (1998). 21. W.K. Choi, S. Kanakaraju, Z.X. Shen, and W.S. Li, Appl. Surf. Sci. 144, 697 (1999). 22. W.K. Choi, V. Ng, S.P. Ng, H.H. Thio, Z.X. Shen, and W.S. Li, J. Appl. Phys. 86, 1398 (1999). 23. W.K. Choi, Y.W. Ho, S.P. Ng, and V. Ng, J. Appl. Phys. 89, 2168 (2001). 24. R.S. Wanger, W.C. Ellis, Appl. Phys. Lett. 4, 89 (1964). 9