J O U R N A L O F C H E M I S T R Y Materials Attempted preparation of diamond-like carbon nitride by explosive shock compression of poly(methineimine) Tamikuni Komatsu National Institute of Materials and Chemical Research, High Density Energy Laboratory, 1-1 Higashi, Tsukuba-shi, Ibaraki 305, Japan Received 2nd June 1998, Accepted 6th August 1998 Explosive shock compression of poly(methineimine) produced a large portion of amorphous graphitic carbon nitride of composition CN0.2 and small amounts of diamond-like carbon containing a slight amount of nitrogen.Codeposition of the two materials suggests the possibility of cubic phase transformation of heterocyclic C–N compounds under well-designed shock-compression conditions. Introduction Experimental Sample preparation Diamond and related materials combined or doped with hetero elements such as B, N, Si and P are expected to be promising The starting material was prepared by ring opening polymerizmaterials in mechanics and electronics for their potential uses ation of s-triazine with excess ZnCl2 at 250 °C for 5 h in an due to their outstanding hardness, thermal conductivity and autoclave, according to ref. 16. The product, which was wide bandgap semiconductivity. The presentations of cubic obtained as a black powder, was heated with conc. HCl in a C–BN by Badzian1 and b-C3N4 by Liu and Cohen2 have water bath, filtered with a glass filter, washed with distilled stimulated numerous investigators to search for diamond-like water and vacuum-dried at 200 °C. The starting material thus hetero compounds.Cubic C–BN compounds were synthesized obtained was mixed with small copper balls in a polymer:copin small amounts by Wedlake,3 Badzian,1 Nakano,4 and per=2:98 mass ratio, filled into a steel capsule, and pressed Knittle5 under static high pressure and high temperature into a disc. The bulk density of the disk was set at 70% of the conditions from 1979 to 1995, and in the next year a consider- theoretical value.The disc was shock compressed using a able amount of purified heterodiamond of composition BC2.5N shock compression apparatus shown in Fig. 1. The apparatus was synthesized by Komatsu et al.6 using an explosive shock- was constructed with a detonator, a sheet explosive discharger compression technique. For b-C3N4, a great number of papers (the shock wave velocity is 6 km s-1), a high melting point have been reported, but only a few studies claim the existence explosive (the shock wave velocity is 9 km s-1), a copper of b-C3N4 so far.The crystalline carbon nitrides which were flying plate, a brass-vessel containing the sample disc, and a synthesized by the groups of Niu,7 Yu,8 Li,9 and Sung10 using steel momentum trap surrounding the sample vessel.The net various chemical and physical vapor deposition methods shock pressure and temperature applied to the sample through agreed fairly well with the predicted interplanar spacings of b- the sample vessel compacted by the explosively accelerated C3N4, however in a strict sense the materials were not com- copper plate was estimated to be 15 GPa and 3500 K, respectpletely proved because of unidentified chemical structures and ively.The recovered sample was machined, immersed in 30% stoichiometry. This essential problem seems to arise from the HNO3 to remove the copper matrix, heated with conc. HCl thermodynamically more unstable b-structure compared to the to remove trace amounts of metallic contaminants, washed a-structure.11,12 There is room for reviewing whether the CVD with distilled water, and vacuum-dried at 200 °C.A fine black and PVD methods are suitable for the synthesis of b-C3N4 powder was obtained at a yield of 36.5%. The material and whether the deposition is quantitatively enough for identification and purification. An explosive shock-compression technique usually used for diamond synthesis is able to supply an extremely high pressure and high temperature to the sample in microseconds and therefore is desirable to obtain non-equilibrium materials like heterodiamond as a bulk form.Wixom attempted to prepare b-C3N4 by shock compression of several pyrolyzed C–H–N compounds under a shock pressure of 30–50 GPa, however he obtained a mixture of amorphous graphitic material and small amounts of diamond.13 Previously, we studied the phase transformation of poly(aminomethineimine) under an explosive shock pressure of about 40 GPa, but unexpectedly no diamond-like materials were produced except for sp2-bonded amorphous carbon nitrides having several diVerent morphologies. 14 As one of the causes, a high atomic ratio of hydrogen included as amino groups of the starting material was anticipated, because a large amount of hydrogen seems to terminate a growth of active species into diamond.15 In this work, the preparation of diamond-like carbon nitride Fig. 1 Section of a plain shock-compression apparatus: 1, detonator; was attempted by shock compression of poly(methineimine) 2, sheet explosive discharge; 3, HMX explosive; 4, copper flying plate; 5, brass vessel containing the sample disk; 6, steel momentum trap.having no functional groups except for theMCHNNMchains. J. Mater. Chem., 1998, 8(11), 2475–2478 2475Table 1 Elemental analysis of the starting material and shock-compression product Starting material Shock-compression product Elemental analysis (%) Atomic ratio Elemental analysis (%) Atomic ratio C H N C H N C H N C H N 49.2 3.9 41.2 1.0 0.9 0.7 41.9 0.1 11.5 1.0 0.0 0.2 obtained was checked in terms of the presence or absence of molecular reconstruction has been induced by shock compression.The following IR and XRD results suggest that this a diamond-like material with an analytical transmission electron microscope. change is possibly due to complicated cyclization of the starting material accompanying elimination of p-bonding nitrogen atoms.Measurements The chemical compositions of the samples were determined IR spectra using a Holiba CHN analyzer. The crystalline structures were determined using a Rigaku RINT-2500 X-ray powder Carbon nitrides are IR active owing to an increased polarity diVractometer (XRD) equipped with a position-sensitive pro- and breaking of symmetric vibrations by incorporation of portional counter and graphite monochromater on the detec- nitrogen with carbon. The materials show generally two tor.Ni-filtered Cu-Ka radiation generated at 50 kV and vibrational modes in the range 1600–1300 cm-1 which are 200 mA was used. The crystalline structure, elemental composi- related to the Raman G and D bands of amorphous carbon.tion, and chemical bonding nature in the microscopic regions Fig. 2(a) shows the IR spectrum of the starting material. The of the samples were investigated using a Hitachi HF-2000 spectrum was simple and broad and resembled the IR spectrum field-emission transmission electron microscope (FETEM) of paracyanogen17 and those of several amorphous CNx thin equipped with an electron diVractometer (ED), a parallel films.18–20 The middle absorption band at 3322 cm-1 was recording electron energy-loss spectroscope (PEELS), and an assigned to NMH stretching of imine, the middle band at energy-dispersive X-ray analyzer (EDX).These measurements 3125 cm-1 to NCMH stretching, the strong bands at were made at 200 kV accelerating voltage.The infrared (IR) 1612 cm-1 and 1549 cm-1 to CNN and CNC stretching spectra of the samples were measured in KBr disk form using (related to the G band), the middle plateau band ranging from a Perkin-Elmer FTIR-1640 spectrometer. 1260 cm-1 to 1450 cm-1 (showing a peak at 1387 cm-1) to ring stretching and CMN stretching (related to the D band), and the small band at 927 cm-1 to NCMH bending.The Results and discussion small peak at 616 cm-1 was unassigned. The chemical structure Chemical analysis of the starting material is an incomplete poly(methineimine) structure including sp3 carbon and nitrogen possibly Table 1 gives the elemental analysis of the starting material and the shock-compression product. The chemical composition of the starting material was approximately CH0.9N0.7 and that of the shock-compression product was close to carbon nitride of composition CN0.2 in which a large portion of hydrogen atoms and significant amounts of nitrogen atoms of the starting material were lost.Such a remarkable change in composition after shock compression indicates that some Fig. 2 IR spectra showing the chemical structure change before and Fig. 3 FETEM image and ED pattern of a large portion of particles after shock compression: (a) the starting material, (b) the shockcompression product. included in the shock compression product. 2476 J. Mater. Chem., 1998, 8(11), 2475–2478cross-linked with adjacent polymer chains, as depicted on the next page. CH C N N) n N CH N (CH CH On the other hand, the shock-compression product showed a more simple broad spectrum as several peaks of the starting material disappeared, as shown in Fig. 2(b). The broad band centered at 1592 cm-1 was assigned to a superposition of CNN stretching and CNC stretching and the plateau band around 1344 cm-1 was assigned to ring stretching. The band at 3442 cm-1 is due to moisture contained in the IR specimen. The simple IR bands and decreased IR activity of the shockcompression product must be related to the graphitic ring formation and the remarkable decrease in nitrogen content compared to the starting material.Fig. 6 PEEL spectrum of the particle observed in Fig. 5. Note the presence of a large amount of carbon and a small amount of nitrogen and the carbons having a large s-bonding feature relative to the p feature.An expanded N-K edge shows only a s* peak. X-Ray diVraction The XRD patterns of the samples before and after shock compression were almost identical: the pattern of the starting material consists of two broad peaks centered at 2h=26.08° (strong peak, d=0.350 nm) and 42.63° (weak, 0.227 nm), and that of the shock-compression product consists of two broad peaks at 2h=26.21° (strong, 0.349 nm) and 42.75° (weak, 0.227 nm) but no patterns of diamond-like materials.These patterns, which are similar to the pattern of amorphous carbon, indicate amorphous structures. TEM analysis Fig. 3 and 4 show a representative TEM image and ED pattern and a PEEL spectrum of the shock-compression product, Fig. 4 PEEL spectrum of the particle observed in Fig. 3. Note the presence of carbon and nitrogen both having p features. respectively.The TEM image and ED pattern show a dis- Fig. 5 FETEM image and ED pattern of small amounts of nanoparticles microdiVused in the matrix of the shock-compression product. J. Mater. Chem., 1998, 8(11), 2475–2478 2477ordered structure peculiar to amorphous carbon, and the Conclusions PEEL spectrum indicates the material is composed of a large Explosive shock compression of poly(methineimine) was amount of carbon and a small amount of nitrogen both having carried out under 15 GPa, 3500 K conditions in order to clear 1s�p* and 1s�s* transitions at the carbon and nitrogen prepare diamond-like carbon nitride.A large portion of K-edges. This means that the grain is combined with sp2- amorphous graphitic CN0.2 and small amounts of diamond- bonding carbon and nitrogen.Mixed with this grain, small like carbon combined with a slight amount of nitrogen were amounts of microdiVused nanoparticles showing a diVerent confirmed by analytical TEM. The low nitrogen concentration bonding nature were observed. The particle was amorphous of the shock-compression product compared to the starting or slightly crystalline, from the TEM image and ED pattern material may be due to phase separation of an unidentified shown in Fig. 5.Considering the influence from the graphitic intermediate material. The finding of diamond-like and gra- material underneath, the measurements of the TEM and PEEL phitic carbon nitrides in the shock-compression product sug- spectrum of this particle were made as nearly as possible on gests the possibility of producing diamond-structured carbon such thin fringes.From the PEEL spectrum shown in Fig. 6, nitrides by shock compression of heterocyclic C–N it was found that the material is composed of a large portion compounds. of carbon and trace amounts of nitrogen and that the carbon K-edge shows a large s* feature relative to the p* feature.This indicates a mixture of sp3- and sp2-bonding carbons. The References ratio of sp3-carbon to total carbons, sp3/(sp2+sp3), was 1 A. R. Badzian, Mater. Res. Bull., 1981, 6, 1385. estimated from the following equation according to ref. 21: 2 A. Y. Liu and M. L. Cohen, Science, 1989, 245, 841. 3 De Beers Industrial Diamond Division Ltd., Ger.Pat., 2806070, sp3/(sp2+sp3)=3(cstd-cexp)/(3cstd+cexp) 1979; R. J. Wedlake and A. L. Penny, Chem. Abstr., 1979, 90, 42865 Z. cstd=(Ip*/Is*)std, cexp=(Ip*/Is*)exp 4 S. Nakano, M. Akaishi, T. Sasaki and S. Yamaoka, Chem.Mater., 1994, 6, 2246. 5 E. Knittle, R. B. Kaner, R. Jeanloz and M. L. Cohen, Phys. Rev. where (Ip*/Is*) is the intensity ratio of the p* and s* peaks B, 1995, 51, 12149.at the carbon K-edge, and the subscripts std and exp, respect- 6 T. Komatsu, M. Nomura, Y. Kakudate and S. Fujiwara, J. Mater. ively, refer to the standard and experimental, and graphite Chem., 1996, 6, 1799. was used as the standard. The result ranged from 40% to 60% 7 C. Niu, Y. Z. Lu and C. M. Lieber, Science, 1993, 261, 334. for many particles. In order to clarify the diVerence in the 8 K.M. Yu, M. L. Cohen, E. E. Haller, W. L. Hansen, A. Y. Liu and I. C.Wu, Phys. Rev. B, 1994, 49, 5034. bonding nature of nitrogen in this material and the earlier 9 Y. A. Li, S. Xu, H. S. Li and W. Y. Luo, J. Mater. Sci. Lett., 1998, graphitic material, the nitrogen K-edges in Fig. 6 and 4 were 17, 31. expanded and compared with each other. The p* and s* peak 10 S.L. Sung, T. G. Tsai, K. P. Huang, J. H. Huang and H. C. Shih, positions in the N–K edge of the graphitic material appeared Jpn. J. Appl. Phys., 1998, 37, L148. at 401.7 and 408.4 eV. On the other hand the noted material 11 Y. Gou and W. A. Goddard III, Chem. Phys. Lett., 1995, 237, 72. showed an appreciable s* peak but an insignificant p* peak 12 D. M. Bhusari, C. K. Chen, T.J. Chuang, L. C. Chen and M. C. Lin, J. Mater. Res., 1997, 12, 322. in these positions. The nitrogen content was too small to 13 M. R. Wixom, J. Am. Ceram. Soc., 1990, 73, 1973. estimate the C/N ratio of the material by PEELS analysis. 14 T. Komatsu and M. Samejima, J. Mater. Chem., 1998, 8, 193. The PEEL spectrum resembled those of amorphous CNx thin 15 R. H. Wentorf, J. Phys.Chem., 1965, 69, 3063. films22,23 and diamond-like carbon.21 Therefore, the material 16 D. Wohrle, Tetrahedron Lett., 1971, 22, 1969. was assigned as amorphous diamond-like CNx. 17 L. Maya, J. Polym. Sci., A: Polym. Chem., 1993, 31, 2595. 18 J. H. Kaufman, S. Metin and D. D. Saperstain, Phys. Rev. B, Although the content of the diamond-like CNx was slight 1989, 38, 13053. in comparison with the formation of graphitic carbon nitride 19 X.A. Zhao, C. W. Ong, Y. C. Tsang, Y. W. Wong, P. W. Chan CN0.2, codeposition of the two materials suggests the synthesis and C. L. Choy, Appl. Phys. Lett., 1995, 66, 2652. of diamond-structured carbon nitrides to be possible by shock 20 J. Hartmann, P. Siemroth, B. Schultrich and B. Raushenbach, compression of heterocyclic C–N compounds. The large nitro- J. Vac. Sci. Technol. A, 1997, 15, 2983. gen loss after shock compression is possibly due to phase 21 D. L. Pappas, K. L. Saenger, J. Bruley, W. Krakow, J. J. Cuomo, T. Gu and R. W. Collins, J. Appl. Phys., 1992, 71, 5675. separation of an unidentified intermediate material because 22 L. A. Bursill, P. Julin, V. N. Gurarie, A. V. Orlov and S. Prawer, the high nitrogen ratio of heterocyclic C–N compounds means J. Mater. Res., 1995, 10, 2277. they tend to lose nitrogen at mild temperatures.17,24 Balance 23 J. Hu, P. Yang and C. M. Lieber, Phys. Rev. B, 1998, 57, R3185. of the applied shock pressure and temperature to the sample, 24 L. Maya, D. R. Cole and E. W. Hagamo, J. Am. Ceram. Soc., preferably high pressure and low temperature conditions, must 1991, 74, 1686. be important for creation of a C–N heterodiamond which is supposed to be kinetically much more unstable than diamond. Paper 8/04 2478 J. Mater. Chem., 1998, 8(11), 2475–2478