J O U R N A L O F C H E M I S T R Y Materials Balloon-shaped graphitic-carbon material induced by shockcompression of dehydrochlorinated poly(vinylidene chloride) Tamikuni Komatsu*a and Miho Samejimab aNational Institute of Materials and Chemical Research, High Density Energy Laboratory 1-1 Higashi, Tsukuba-shi, Ibaraki 305, Japan bExplosive Research Laboratory, Asahi Chemical Industry Co., Ltd., 304 Mizushiri, Nobeoka, Miyazaki 882, Japan Received 27th April 1998, Accepted 14th September 1998 Shock-compression of dehydrochlorinated poly(vinylidene chloride) at 15 GPa and 6500 K produced a small portion of balloon-shaped graphitic-carbon material mixed with a large proportion of graphite and diamond.The size and thickness of the various balloon particles were ca. 40–300 nm and 4–6 nm, respectively. The balloonshaped material is presumably a by-product from a carbon source fragmented under shock-compression and would not be related to the high yield of diamond. Carbon forms fundamentally three types of allotropes resulting a trace of a new crystalline carbon mixed with diamond and graphite.16 The new carbon was assigned to the hexagonal from its available three bond modes, sp, sp2 and sp3.In nature, crystal system with cell dimension a0=0.338 nm and was only the sp2-type allotrope (graphite) and sp3-type (diamond) compared with carbyne in terms of the carbon geometry exist. The search for a sp-type has originated from a discovery projected onto the 001 plane. Here, we further investigate of two resonance hybrids, a-carbyne1 (polyyne form) prepared the possibility of producing unknown carbon-based substances by high temperature treatment of oxidative condensates of under a shock pressure of 15 GPa, the medium pressure copper acetylide and b-carbyne2 (polycumulene form) derived required for the shock-induced graphite-to-diamond trans- by coupling of carbon suboxide and sodium acetylide, which formation, using the same starting material as previously.was made by a Russian group between 1961 and 1967. This was followed by the preparation of eight or more polymorphs of carbon from vaporized carbon,3 de(hydro)halogenated polymers having long polyyne segments (carbynoids),4–6 and Experimental a coil form of polyyne by laser vaporization of graphite.7 Sample preparation However, the presence of carbyne as a high temperature material is unusual considering the fragility of conjugated The starting material was prepared by dehydrochlorination of polyyne compounds.8 poly(vinylidene chloride) (Asahi Chemical Industry Co.Ltd., The world of carbon allotropes seemed to have been Saran polymer MW=105) using 1,8-diazabicyclo[5.4.0]undeccompleted with diamond, graphite and carbyne.However, 7-ene (DBU) as a base and N,N-dimethylacetamide as solvent. between 1985 and 1992, an outstanding discovery of carbon Determination of the chemical structure using elemental analyclusters by Kroto et al.,9 Iijima,10 and Ugarte,11 the so-called sis, IR, Raman, UV–VIS, and NMR spectroscopy was carried fullerenes, nanotubes and carbon onions, which form topologi- out in the same way as described previously.16 The material cally closed-shell sp2-carbon networks completely diVerent which was obtained as a fine black powder was an amorphous from planner graphite networks, were synthesized upon high block polymer linked with conjugated polyene and polyyne energy irradiation of graphite such as laser vaporization, arc segments, –(CHLCH)n–(CMC)m–, containing small amounts evaporation, and electron irradiation.This finding along with of carbonyl and nitrogen and trace amounts of chlorine, and the recent prediction of an unexpected variety of carbon this structure was confirmed to be the same as previously. isomers12 like cyclo-C18 and graphyne, etc. has encouraged the A homogeneous mixture of the starting material (2 equiv.) search for other new carbon-based substances. On the other and coarse-grained copper balls (98 equiv.) was filled in a steel hand, the carbon clusters including carbyne were found to sample holder of 30 mm diameter and 15.5 mm depth, and show surprisingly high diamond conversions under much pressed at 28 MPa for 30 min.The product was then dried at milder physical conditions in comparison with the graphite- 80 °C for 20 min under ordinary pressure, and vacuum-dried to-diamond transformation, for example, C60-to-diamond (90 at 120 °C for 2 h.The packing density of the starting material mass%) conversion occurs under a non-hydrostatic pressure in the pressed sample was 0.44 g cm-3 (20% of the theoretical of 20±5 GPa at room temperature,13 and carbyne-to-diamond density of graphite).Shock-compression of this sample was (99 mass%) occurs under a static pressure of 8 GPa at carried out using the plane shock-compression apparatus 1400–1700 °C without catalyst.14 shown in Fig. 1. The apparatus was constructed with a deton- Knowledge of the isomerisation between two carbon ator, an explosive lens composed of a set charge of hydrazine allotropes suggests the possibility of producing hitherto nitrate (63.5 mass%)–hydrazine hydrate (36.5 mass%) and unknown carbon allotropes with or from diamond.Shock- nitromethane, a high melting point explosive (HMX, detoncompression15 used for diamond synthesis owing to the ation speed 9.1 km s-1), a copper flyer, a brass vessel containextremely high dense energy supply by shockwaves and success- ing the sample holder, and a steel momentum trap surrounding ive rapid cooling in microseconds is expected to be a very the brass vessel.The average shock pressure applied to the useful method to search for new metastable carbon-based sample was estimated to be 15 GPa. The recovered sample substances. We studied previously shock-compression of was machined, immersed in 10% HNO3 to remove the copper dehydrochlorinated poly(vinylidene chloride) (a carbynoid) matrix, washed with distilled water, and vacuum-dried at 200 °C.A fine black powder was obtained in a yield of ca. 25%. under several GPa for the preparation of carbyne and found J. Mater. Chem., 1998, 8, 2725–2728 2725Fig. 1 Section of the plane shock-compression apparatus: 1, detonator; 2, explosive lens consisting of (a) hydrazine nitrate–hydrazine hydrate and (b) nitromethane; 3, high melting point explosive; 4, copper-flyer; 5, brass vessel containing the sample; 6, steel momentum trap.Measurements The C, H, N contents of the sample were determined on a Holiba CHN analyzer. Raman spectra were measured at an excitation wavelength of 514.5 nm generated with a 1.0 mW argon laser using a MST foundation Ga-505 spectrometer.The crystal structures were investigated using a Rigaku RINTFig. 2 Raman spectra of the starting material and the shock- 2500 X-ray powder diVractometer equipped with a position- compacted sample: (a) the starting material; (b) the shocksensitive proportional counter and a graphite monochromator compacted sample.on the detector. Ni-filtered Cu-Ka radiation generated at 50 kV and 200 mA was used as the X-ray source. The micro- X-Ray diVraction scopic morphologies were observed using a JEOL JEMFig. 3 shows the X-ray powder diVraction of the 4000FX high-resolution transmission electron microscope shock-compacted sample. The diVraction consisted of two (HRTEM) equipped with an energy-dispersive X-ray analyzer patterns.The observed d-values, intensities, and lattice con- (EDX), an electron energy-loss spectrometer (EELS) and an stants of the patterns are given in Table 1 in addition to the electron diVractometer (ED). The nanoscaled observation was data of the known carbon materials. Pattern 1, a cubic complemented using a Hitachi HF-2000 instrument for field structure with a=0.35597 nm, agreed with that of diamond emission transmission electron microscopy (FETEM) (JCPDS 06-0676) except for a broad FWHM (2h=1.664° for equipped with an ED, a parallel electron energy-loss specthe 111 peak).Pattern 2, a hexagonal structure with the trometer (PEELS) and an EDX. The sample powder was diVused supersonically in a methanol–distilled water (151) solution for 5 min, a drop of the solution was then placed on a microgrid coated with a carbon–collodion membrane, air-dried, and measured at 400 kV accelerating voltage.Results and discussion Elemental analysis and Raman spectrum The C/H/N content of the shock compacted sample was 98.1% C, 0.3% H and 0.3% N. The elements of the same sample detected by EDX were almost entirely carbon except for oxygen and silicone impurities due to the carbon–collodion membrane and copper due to the microgrid used for the TEM measurement.Fig. 2 shows the Raman spectra of the starting material and the shock-compacted sample. The spectra indicate that the starting material was entirely transformed into diVerent materials. The bands at 1592 and 1352 cm-1 of the shockcompacted sample were similar to those of glassy carbon17 in terms of the peak position, and were assigned to G- and Dbands of the graphite structure, respectively.A broad line Fig. 3 X-Ray powder diVraction of the shock-compacted sample. width and a relatively comparable intensity of the peaks For comparison, the standard patterns of graphite (JCPDS 25-0284) indicate the material is coagulated with poorly crystalline and diamond (JCPDS 06-0676) are shown together with their superposition (syn).fine grains. 2726 J. Mater. Chem., 1998, 8, 2725–2728Table 1 X-Ray diVraction pattern of the shock-compacted sample and assignment of the structures by comparison with patterns of standard references Pattern 1 Cubic Diamond (JCPDS 06-0676) a=0.35597 nm a=0.35667 nm 2h/degrees dobs/nm dcalc/nm 100 I/I0 hkl d/nm 100 I/I0 hkl 44.022 0.20553 0.20552 100 111 0.20600 100 111 75.479 0.12585 0.12585 24 220 0.12610 26 220 91.734 0.10733 0.10733 13 311 0.10754 16 311 120.416 0.08876 0.08899 4 400 0.08916 8 400 141.221 0.08166 0.08166 8 331 0.08182 16 331 Pattern 2 Graphite (JCPDS 25-0284) Hexagonal c=0.6683 nm a=0.2458 nm, c=0.6696 nm 2h/degrees dobs/nm dcalc/nm 100 I/I0 hkl d/nm 100 I/I0 hkl 26.657 0.33413 0.33413 100 002 0.33480 100 002 0.21270 3 100 0.20270 15 101 0.17950 3 102 53.953 0.16981 0.16707 5 004 0.16740 6 004 lattice constant c=0.6683 nm, fitted well with that of graphite (JCPDS 25-0284).The 100 peak of pattern 2 was apparently absent because of overlapping with the 111 peak of pattern 1. As a guide to indexing, the standard patterns of the graphite and diamond are included in Fig. 3.The average size of diamond crystallites estimated from the FWHM is ca. 5 nm, and the formation of such nanosized diamond particles is also a characteristic of the shock-compression synthesis. The existence of diamond was confirmed by XRD, however, it could not be detected by Raman spectroscopy, probably because of its very small size.The mass ratio of diamond to graphite contained in the shock-compacted sample, which was estimated by calibration with the relative intensity of known composition, was ca. 70%. As already stated, diamond conversion of the carbon clusters exceeds 90% under even much lower pressures. A characteristic common feature of these starting materials is an abundance of active unsaturated bonds.The bonds would be easily fragmented even under mild conditions, which may be related to such a high diamond conversion. TEM observation Fig. 4 shows the HRTEM image of the shock-compacted sample. This reveals a large portion of graphitic domains and microdiVused fine crystalline particles showing a lattice image. Fig. 4 High resolution TEM image of the shock-compacted sample.The FETEM-ED patterns of the graphitic domains and crystal- The areas enclosed within squares indicate microdiVused fine line particles showed a diVuse ring pattern and a spot pattern, crystalline particles having a lattice image. respectively. The interplanar spacings determined from the spot pattern were 0.208 nm 111 , 0.127 nm 220 and spectrum showed both 1s�p* and 1s�s* transitions at the 0.0826 nm 331 .The values are in accord with those of carbon K-edge similar to those of graphite, which indicates diamond within an experimental error of 2% using the silicon the material is composed of graphitic sp2 carbon. The size and pattern. Therefore, the crystalline material was assigned to thickness of the various balloon particles estimated from the diamond. The graphitic domains were assigned to graphite TEM observation were ca. 40–300 nm and ca. 4–6 nm, from the similarity to the ED pattern of low crystalline respectively. This form of carbon material is probably new. graphite. These assignments were also supported by the Raman The shock pressure, P, applied to the sample and the particle and XRD results described above.In addition to these mate- velocity, Up, estimated from the Hugoniot data by Marsh,18 rials, balloon-shaped particles as shown in Fig. 5 were are 15 GPa and 5.2 km s-1 respectively. The internal energy occasionally seen. In this figure, the various balloon particles increase, DE, and a temperature increase, DT , of the sample cluster together locally near the graphite domains.The outside under shock-compression are derived from the well known of the particles is covered with a dense wall as a round closed Rankin–Hugoniot equations:19 shell in contrast to the inside which suggests a bright non- DE=Up2/2 (1) structured space. The shell structure is seen as laminated with graphitic thin layers oriented along the walls.The EELS DT=E/3R (2) J. Mater. Chem., 1998, 8, 2725–2728 2727Conclusions Creation of a new type of carbon-based substance was attempted by shock-compression of dehydrochlorinated poly (vinylidene chloride). Codeposition of a small quantity of balloon-shaped graphitic-carbon material and a large quantity of graphite and of diamond was confirmed by TEM analysis. The size and thickness of the various balloon particles were 40–300 nm and 4–6 nm, respectively. The pressure and temperature applied to the sample by shock-compression were estimated to be 15 GPa and 6500 K, respectively.The balloonshaped material was presumably deposited as a by-product from a carbon source fragmented under shock-compression and would not be related to the high yield of diamond. References 1 V.V. Korshak, V. I. Kasatochikin, A. M. Sladkov, Yu. P. Kudryavtsev and K. Usenbaev, Dokl. Akad. Nauk SSSR, 1961, 136, 1342. 2 V. I. Kasatochkin, T. M. Babchinitsev, Yu. P. Kudryavtsev and A. M. Sladkov, Dokl. Akad. Nauk SSSR, 1967, 177, 358. 3 A. G. Whittacker, Science, 1978, 200, 763. 4 K. Akagi, M. Nishiguchi, H. Shirakawa, Y. Fulukawa and I. Harada, Synth. Met., 1987, 17, 557. 5 Yu. P. Kudryavtsev, S. E. Evsyukov, V. G. Babaev, M. B. Guseva, V. V. Khvostov and L. M. Krechko, Carbon, 1992, 30, 213. 6 J. Kastner, H. Kuzmany, L. Kavan, F. P. Dousek, and J. Kurti, Macromolecules, 1995, 28, 344. Fig. 5 High resolution TEM image of balloon-shaped particles 7 R. J. Lagow, J. J. Kampa, H. C. Wei, S. L. Battle, J. W. Genge, contained in the shock-compacted sample.D. A. Laude, C. J. Harper, R. Bau, R. C. Stevens, J. F. Haw and E. Munson, Science, 1995, 267, 20. 8 R. Eastmond, T. R. Johnson and D. R. M. Walton, Tetrahedron, where R is the gas constant. From these, DE=13.5 MJ kg-1 1972, 28, 4601. and DT=6500 K are obtained. This temperature is suYciently 9 H. W. Kroto, J. R. Heath, S. C. O’Brien, R. F. Curl and high to make the sample fragment, therefore the composition R.E. Smalley, Nature (London), 1985, 318, 162. of the shock-compacted sample would originate from carbon- 10 S. Iijima, Nature (London), 1991, 354, 56. aceous fragments, that is, the carbonaceous fragments are 11 D. Ugarte, Nature (London), 1992, 359, 707. presumably transformed into diamond as a high pressure 12 F. Diederich and Y. Rubin, Angew.Chem., Int. Ed. Engl., 1992, 31, 1101. phase, graphite as the thermodynamically the most stable 13 M. N. Regueiro, P. Monceau and J. L. Hodeau, Nature (London), phase, and the balloon-shaped graphitic carbon possibly as a 1992, 355, 237. high temperature phase. However, the previous crystalline 14 V. V. Korshak, Yu. V. Korshak, Yu. P. Kudryavtsev, carbon16 was not observed in this work in spite of the same M.B. Guseva, V. G. Babaev, V. V. Khvostov, S. E. Evsyukov, starting material. This is probably due to the diVerence in the a. D. Varfolomeeva, USSR SU, 1,533,221, 1993; Chem. shock conditions. The pressure and temperature in this work Abstr., 1994, 120, 110831a. 15 E. I. du Pont de Nemours, USP., 3401019, 1968; G. R. Cowan, were much higher than those (several GPa, 600–800 K) in the B.W. Dunnington and A. H. Holtman, Chem. Abstr., 1969, 70, previous study. The high pressure increased the yield of 699. diamond, while the high temperature (6500 K) would com- 16 T. Komatsu, M. Nomura, Y. Kakudate, S. Fujiwara and pletely decompose the previously observed crystalline material R. B. Heimann, Macromol. Chem. Phys., 1995, 196, 3031. even if it were formed. Also, codeposition of known carbon 17 H. Hiura, T. W. Ebbessen and K. Tanigaki, Chem. Phys. Lett., clusters such as fullerenes, nanotubes and carbon onions is 1993, 202, 509. 18 S. P. Marsh, in LASL Shock Hugoniot Data, Univ. California, unknown. This exclusive tendency among carbon allotropes Berkeley, 1980, p. 43. other than graphite and diamond (we do not know any 19 R. G. McQueen, S. P. Marsh, J. W. Taylor, J. N. Fritz and examples of codeposition for the carbon clusters) suggests W. J. Carter, in High-Velocity Impact Phenomena, ed. R. Kinslow, that the balloon-shaped material would be a kinetically con- Academic Press, New York, 1970, p. 244. trolled by-product but not a thermodynamically controlled intermediate leading to diamond. Paper 8/03131B 2728 J. Mater. Chem., 1998, 8, 2725–2728