CMCに関する原著論文

J. Mater. Res., Vol. 14, No. 11, 4329-4336(1999)

Three-dimensional vapor growth mechanism of carbon microcoils

Xiuqin Chen
Department of Chemical Engineering, Huaqiao University, Fujian 362011, People 's Republic of China

T. Saito and M. Kusunoki
Japan Fine Ceramics Center, 2-4-1 Mutsuno, Atsuta-ku, Nagoya 456-8587, Japan

S. Motojima
Department of Applied Chemistry, Faculty of Engineering, Gifu University, Gifu 501-1193, Japan

(Received 16 April 1999; accepted 4 August 1999)

　Carbon microcoils were grown by the Ni-catalyzed pyrolysis of acetylene. The growth patterns and the tip morphologies of the carbon coils are examined in detail, and a growth mechanism is proposed. Basically, six thin fibers grew from a Ni catalyst grain during the initial growth stage immediately followed by the coalescence of the four fibers to form two fibers and then forming double-helixed carbon coils. A small amount of S and O, as well as C and Ni, was observed on the periphery of the cross section of the Ni catalyst grain. On the other hand, S and O were not observed in the central part. The driving force of the coiling of the straight fibers to form carbon coils is considered to be the strong anisotropy of the carbon deposition between different crystal faces.

I. INTRODUCTION

Many researchers have been trying to prepare materi-als with a 3D-helical/spiral structure. Some researchers have reported the growth of coiled fibers of carbon,1-9 SiC,10~12 and Si3N413~16 from the vapor phase. However, the growth of these coiled fibers is extremely accidental and the reproducibility is very poor. We have prepared regularly microcoiled carbon fibers by the catalytic pyrolysis of acetylene containing a small amount of sulfur or phosphorus impurity, and reported the preparation conditions, morphologies, and some properties of the products.17-26 The microcoiled carbon fibers (referred to as "carbon coils" hereafter) have a 3D-helical/spiral structure, which is the fundamental structure in nature: DNA, proteins, electric waves, growth of vine plants, screw dislocation in solids, etc. The carbon coils may be the candidates for novel electromagnetic wave absorbers, tunable microdevices, absorbers of hydrogen, etc. Furthermore, the growth mechanism of carbon coils is very interesting based on the similarity of the fundamental structure of all objects. Amelinckx et al. proposed a formation mechanism for a catalytically grown helix-shaped graphite nanotube.28 We postulated the 2D and 3D growth mechanism of carbon coils based on the anisotropy of carbon deposition among catalyst crystal faces.26,27,29 However, the growth process and growth mechanism of the carbon coils has not been known in detail until now. In this paper, we propose a 3D growth mechanism of the carbon microcoils based on the growth patterns and morphologies of the coils from the vapor phase, and also show some of its properties.

II. EXPERIMENTAL

The carbon coils were obtained by the Ni-catalyzed pyrolysis of acetylene containing a small amount of thiophene. The apparatus, preparation procedure, and reaction conditions used for obtaining carbon coils are noted in Ref. 21.

III. RESULTS AND DISCUSSION

A. Growth conditions and morphology of the carbon coils

Figure I shows an enlarged view of the used Ni catalyst powder. It can be seen that this Ni catalyst powder has many secondary grains composed of very fine grains averaging 500 nm in diameter. This grain size is comparable with that of the fibers from which the carbon coils were formed. Generally, the carbon coils grew perpendicular on the substrate as shown in Fig. 2. A Ni grain was always observed on the tip of the carbon coils, and the Ni grain, which is a growing point, pointed in the source gas inlet direction (Fig. 2, arrow). There were two kinds of carbon coils: circular carbon coils having a circular or elliptical cross section as shown in Fig. 3(a) and flat carbon coils having a flat cross section as shown in Fig. 3(b). The circular carbon coils generally grew during the relatively initial growth stage and then became flat carbon coils. Accordingly, the carbon coils obtained after a 2-h reaction time have more flat carbon coils. Table I shows the effects of reaction time on the shapes and dimensions of the carbon coils. The carbon coils with an irregularly coiled shape during the initial growth stage became regularly shaped with increasing reaction time.

The coil diameters decreased with increasing reaction time, and attained a constant value of about 5 ｵm above a I -h reaction time. The growth of small amounts of very thin coils was observed among these coils. The smallest diameter of the carbon fibers and coils observed were 50 nm and 300 nm, respectively. The diameter of the carbon fiber, from which carbon coil is formed, is considered to be mainly determined by the diameter of the used catalyst grain, and coil diameter by the difference of the catalytic anisotropy of the respective crystal faces of the catalyst, as will be shown later.

B. Microstructures

Figure 4 shows the enlarged views of the ruptured cross section and surface of the carbon coils shown in Fig. 3(b). Fine carbon grains are observed on the ruptured cross section and surface, and no pore is observed in the central part of the fiber axis. Compared with a vapor-grown carbon fiber (VGCF), which is a straight crystalline fiber and has a continuous fine pore through the fiber axis, the carbon coils have two critical characteristics: a microcoiled morphology composed of almost amorphous fine carbon grains and the absence of pores in the fiber axis. The polished cross sections of the carbon coils are shown in Fig. 5. It can be seen that fine and slender carbon grains are concentrically oriented, and the grain regions can be apparently divided into three parts (A, B, and C). The transmission electron microscopy (TEM) image of the cross section of a bulk carbon coil is shown in Fig. 6. It can also be seen that the cross sections are reasonably divided into three regions. That is, a piece of the carbon coil is formed from three parts of the car-bon grains which were deposited from different Ni cata-lyst crystal faces, as will be shown later. Figure 7 shows the high-resolution TEM image of the cross section of the bulk carbon coils. The apparent lattice image of the graphite layers cannot be observed while the lattice image of very short-range orders are observed. Figure 8 shows a TEM image of the tip part of the carbon coils. A x-ray microanalysis was carried out in the cross section of the Ni catalyst grain observed on the tip part (referred to as "coil tip" hereafter) and the bulk carbon coils, and these results are shown in Fig. 9. The presence of a small amount of sulfur (S) and oxygen (O) as well as nickel (Ni) and carbon (C) is observed on the periphery of the Ni grain [Fig. 9(a)], whereas S or O are scarcely observed in the central part of the Ni grain [Fig. 9(b)]. The presence of small amount of S is also observed in the bulk carbon coils [Fig. 9(c)]. The Ni catalyst observed on the coil tips is a Ni3C single crystal (rhombohedral).25 Accordingly, the Ni catalyst grain observed on the coil tips is a Ni3C single crystal, on the surface of which a small amount of S and O was contained, and this mixed phase of (Ni-C-S-O)/Ni3C single crystal is considered to act as the actual catalyst phase.

C. Morphology of the coil tip obtained during the initial growth stage and growth mechanism

A Ni grain is always observed on the coil tip. There were generally two kinds of growth patterns of the carbon coils from the Ni grain. One kind is that two fibers grew from a Ni grain to form the double coils.25 The other kind is that six fibers, A, B, C, and A', B ', C', grew from a Ni grain (among which A and B and A' and B ' are coalescing together, respectively), the six fibers coalesce to form two fibers, X and Y, and then form the double coils as shown in Fig. 10. An enlarged view of the representative tip morphology showing the six fibers' growth from a Ni grain is shown in Fig. I I . These interesting morphologies were usually observed on the carbon coils obtained during the initial stage of growth within 5 min of reaction time. The fact that fibers X and Y, which built up the coil, are composed of three pieces can be definitely verified by Fig. I I . Some stripes or striations were always observed between the surface of fibers A and B (or A' and B'). Figure 12 shows a schematic drawing of the Ni grain shown in Fig. 10. From the crystal face a-d-h-e, fiber A, from the face a-b-c-d, fiber B, from the c-d-h-g, fiber C, from e-f-g-h, fiber A', from b-c-g-f, fiber B ' and from a-b-f-e, fiber C' are formed. The rhombohedral Ni grain shown by "a-h" is slightly inclined; pointing the apex "a" (or “g” ) to the fiber axis. '' ,, The two fibers, C and C', are not connected straight because of this inclination of the Ni grain, while the two fibers, A and A', are connected straight. From these observations, we show the growth model of the carbon coils in Fig. 13. In this model, the order of the catalytic activity of the three crystal faces is A > B > C. Basically, the carbon fiber is formed from the fine carbon grains de-posited from the three crystal faces of A, B and C, and curls such that the carbon layers deposited from the crystal faces A and B are the outer side that forms the carbon coils. The coil diameters may be determined by the anisotropy of the carbon deposition between the crystal faces of A and C, or B and C, and the coil pitch by that of A and B. Furthermore, the fiber diameters and the cross sections may be determined by the dimensions and shape of the Ni catalyst grain. That is, the circular carbon coils may be grown from the isometric Ni catalyst grain [Fig. 13(a)] and the flat carbon coils from the elongated Ni grain [Fig. I 3(b)]. We reported that the order of the deposition rate of the carbon coils between different Ni single crystal faces was Ni(100) > (111) > (110).26 On the other hand, Yang and Chen reported that the most favored face for the graphite precipitation by the pyrolysis of methane at 700 'C was Ni(111) followed by Ni(111) > (311) > (100) > (110).30 We could not yet identify the respective crystal faces as shown in Figs. 12 and 13. However, it may be reasonably considered that faces A, B, and C are (100), (111), and (110), respectively.

D. Influence of reaction conditions on the catalytic anisotropy

The reaction conditions significantly affected the morphology of the carbon coils, that is, on the anisotropy of the Ni catalyst. The presence of small amounts of impurities such as S or P is indispensable for the growth of the carbon coils. A coil tip obtained after stopping the sulfur impurity during the growing of the carbon coils is shown by the arrow in Fig. 14. The carbon fibers could not normally coil to form irregularly curled fibers. The car-bon coils with larger coil pitches and larger coil diameters than those obtained under standard conditions were obtained at a lower flow rate of acetylene or H21C2H2 ratio and also higher or lower reaction temperatures that those of the standard conditions as shown in Figs. 15 and 16. At a reaction temperature of 700 'C, the carbon coils could not be obtained, and the paired straight fibers formed by the coalescence of two fibers (A + B) and C were obtained. It may be considered that the anisotropy of the carbon deposition cannot be attained at this low temperature. Using an Inconel reaction tube that con-tained iron, the growth of the carbon coils was sup-pressed, and only short and conically grown coils were obtained, as shown in Fig. 17. Carbon layers on the outer part of the carbon coils are fringed, striped, or ruptured, indicating the fast growth rate of the outer part. However, the growth rate of the carbon coils was very low, probably caused by the poisoning effect of the iron metal impurity on the catalytic activity of Ni catalyst.

E. Some properties

The as-grown carbon coils have almost an amorphous phase, which can be graphitized by heat treating at 3000 'C for 5 h. The bulk electric resistivity, surface area, and density of the as-grown carbon coils are 10-10~1 ｽ cm, 60-140 cm2lg and 1.7-1.9 g/cm3, respectively. The carbon coils can absorb about 99% of the electromagnetic waves of GHZ order.

IV. CONCLUSIONS

The carbon coils were obtained by the Ni-catalyzed pyrolysis of acetylene. The growth patterns and the tip morphologies of the carbon coils were examined in detail and the growth mechanism was discussed. Principally, six thin fibers grew from a Ni catalyst grain at the tips, and immediately following four fibers coalesced to form two fibers. Next, the formed four fibers coalesced with each other to form two fibers, and then to form double-helixed carbon coils. A small amount of sulfur and oxy-gen were observed on the periphery of the cross section of the Ni grain, and not observed in the central part of the Ni grain. The driving force of the coiling of straight fibers to form the carbon coils is considered to be the strong anisotropy of the carbon deposition from different crystal faces. This anisotropy was strongly affected by the reaction conditions, such as gases flow rate and ratio, reaction temperatures, and carbon coils with a coil diam-eter greater than 10 ｵm or where very irregular coils are obtained.

REFERENCES

1. W.R. Davis, R.J. Slawson, and G.R. Rigby, Nature 171, 756 (1953).

2. W.R. Davis, R.J. Slowson, and G.R. Rigby, Trans. Br. Ceram. Soc. 56, 67 (1957).

3. R.T.K. Baker, M.A. Barber, P.S. Harris, F.S. Feates, and R.J. Waite, J. Catal. 26, 51 (1972).

4. L.S. Lobo and D.L. Trimm, J. Catal. 29, 15 (1973).

5. R.T.K. Baker and R.J. Waite, J. Catal. 37, 1018 (1975).

6. M. Hillert and N. Lange, Z. Kristallogr. 111, 24 (1958).

7. J. Calszka and M.H. Back, Carbon 22, 141 (1984).

8. M. Audier and M. Coulon, Carbon 23, 317 (1985).

9. R.A. Dalla Metta, A.G. Piken, and M.J. Shelef, J. Catal. 40, 173 (1975).

10. A. Addamiano, J. Cryst. Growth 58, 617 (1982).

11. T-K. Kang, S-D. Park, C-K. Rhee, and II-H.K. Kuk, in Proceed-ings of the 6th Japan-Korea Ceramics Seminar, Kobe, Japan, 1989, p. 249.

12. S. Motojima, T. Hamamoto, H. Iwanaga, J. Cryst. Growth 158, 79 ( I 996) .

13. S. Motojima, S. Ueno, T. Hattori, and K. Goto, Appl. Phys. Lett. 54, 1001 (1989).

14. S. Motojima, S. Ueno, T. Hattori, and H. Iwanaga, J. Cryst. Growth 96, 383 (1989).

15. S. Motojima, T. Yamana, T. Araki, and H. Iwanaga, J. Electro-chem. Soc. 142, 3141 (1995).

16. U. Vogt, H. Hofmann, and V. Kramer, Key Eng. Mater. 89-91, 29 (1994).

17. S. Motojima, M. Kawaguchi, K. Nozaki, and H. Iwanaga, Appl. Phys. Lett. 56, 321 (1990).

18. S. Motojima, M. Kawaguchi, K. Nozaki, and H. Iwanaga, Carbon 29, 379 (1991).

19. S. Motojima, I. Hasegawa, M. Kawaguchi, K. Nozaki, and H. Iwanaga, J. Chem. Vapor Deposition 1, 136 (1992).

20. S. Motojima, I. Hasegawa, S. Kagaya, S. Asakura, M. Kawagu-chi, and H. Iwanaga, J. de Phys. IV (C3), 599 (1993).

21. S. Motojima, H. Hirata, and H. Iwanaga, J. Chem. Vapor Deposi-tion 3, 87 (1994).

22. S. Motojima, I. Hasegawa, S. Asakura, K. Ando, and H. Iwanaga, Carbon 33, 1167 (1995).

23. S. Motojima, Y. Itoh, S. Asakura, and H. Iwanaga, J. Mater. Sci. 30, 5049 (1995).

24. S. Motojima, S. Kagiya, and H. Iwanaga, Mater. Sci. Eng. B 34, 47 (1995).

25. S. Motojima, S. Asakura, M. Hirata, and H. Iwanaga, Mater. Sci. Eng. B 34, L9 (1995).

26. S. Motojima, S. Asakura, T. Kasemura, S. Takeuchi, and H. Iwanaga, Carbon 34, 289 (1996).

27. M. Kawaguchi, K. Nozaki, S. Motojima, and H. Iwanaga, J. Cryst. Growth 118, 309 (1992).

28. S. Amelincks, X.B. Zhang, D. Bernaerts, X.F. Zhang, V. Ivanov, and J.B. Nagy, Science 265, 635 (1994).

29. S. Motojima and X. Chen, J. Appl. Phys. 85, I (1999).

30. R.T. Yang and J.P. Chen, J. Catal. 115, 52 (1989).

原著論文　　　　　総説・解説・資料　　　　　新聞　　　　　雑誌　　　　　テレビ報道
 口絵　　　学会予稿原稿　　　　　セミナー・イベント等　　　　　その他の文献