A facile mechanochemical way to prepare g-C3N4

A facile mechanochemical way to prepare g-C3N4

Materials Science and Engineering B 122 (2005) 90–93 A facile mechanochemical way to prepare g-C3N4 Huaizhou Zhao a , Xiaolong Chen a,∗ , Chengchang ...

234KB Sizes 7 Downloads 54 Views

Materials Science and Engineering B 122 (2005) 90–93

A facile mechanochemical way to prepare g-C3N4 Huaizhou Zhao a , Xiaolong Chen a,∗ , Chengchang Jia b , Ting Zhou b , Xuanhui Qu b , Jikang Jian a , Yanping Xu a , Tang Zhou a a

Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, P.O. Box 603, Beijing 100080, PR China b State Key Laboratory for Advanced Metals and Materials, School of Materials Science and Engineering, University of Science and Technology Beijing, Beijing 100083, PR China Received 9 March 2005; received in revised form 20 April 2005; accepted 2 May 2005

Abstract Under the conditions of high-energy ball milling, the reaction of C3 N3 Cl3 and Li3 N has progressed in a high-efficient way. Carbon nitride with the atomic ratio of 1.23–1.30 for N to C that is very similar to the stoichiometry of C3 N4 has been obtained. Moreover, XRD reveals that with the prolonging of the milling time, the products can be soon transformed to comparatively well-crystallized graphitic carbon nitride (g-C3 N4 ). SEM investigations show the products have the glassy morphology with different dimension from micron to nano scales. The XPS analysis shows the products have the presumable element compositions and the chemical bond state. © 2005 Elsevier B.V. All rights reserved. Keywords: Carbon nitride; g-C3 N4 ; Mechanochemical preparation

1. Introduction During the last decades, carbon nitrides have long been the focus of interests [1–13] due to their novel predicted mechanical, optical, and tribological properties since ␤-C3 N4 was firstly provided in theory by Liu and Cohen [1] in 1989. To date, there have been tens of C3 N4 phases postulated [2,3,8,12], several dense, hard sp3 -hybridised carbon nitride structures such as ␤-C3 N4 (P 63 /m), ␣-C3 N4 (P31 c), c-C3 N4 (I-43 d), pc-C3 N4 (P-42m) have been examined with various theoretical approaches. Meanwhile, a large number of soft, sp2 -hybridised carbon nitride variants like g-C3 N4 have been also suggested. Kroke and Schwarz [12] summarized the sequence of stability of these C3 N4 modifications under ambient conditions, and by taking the shear modulus G0 and the bulk modulus B0 into concerned, he suggested the order of hardness for these phases in which pc-C3 N4 was considered on the top. Besides of the much effort devoted on hard C3 N4 phases, more attentions are still focused on the graphitic forms of carbon nitride, g-C3 N4 , which is believed ∗

Corresponding author. Tel.: +86 10 82649039; fax: +86 10 62556598. E-mail address: [email protected] (X. Chen).

0921-5107/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.mseb.2005.05.007

to be the potential precursors of the superhard phases [2], just like the phase relation between graphite and diamond under high temperature–pressures conditions. The synthesis of bulk solid carbon nitrides has been a definitely challenging work for researchers due to the less thermodynamic stability of C3 N4 phases than that of the separate nitrogen molecules. Many approaches have been provided: Kroll and Hoffmann [4] investigated theoretically novel experimental polymer route for C3 N4 materials. Several nitride-rich carbon nitrides that can be promising precursors for C3 N4 phases have been synthesized [11–13]. By simple solvent-thermal routes, Zhu and co-workers [7] successfully prepared crystalline carbon nitride powder mainly composed of ␣- and ␤-C3 N4 , Xie and co-workers [14] synthesized C3 N4 nanotubes via low temperature benzene-thermal process. Recently, Yin et al. [10] reported the preparation of single-crystalline ␤-C3 N4 nanorods via reaction of graphite and NH3 under high-energy ball milling conditions. Besides, many groups have synthesized amorphous and g-C3 N4 through various approaches [5,6,15–25]. Among of those approaches mentioned above, mechanochemical route has the potential to obtain metastable phases under appropriate conditions [26]. There have been reports

H. Zhao et al. / Materials Science and Engineering B 122 (2005) 90–93


on the synthesis of nitride nanopowders by mechanochemical processing [26,27]. As to the synthesis of carbon nitrides, there are only a few works reported [10,28,29]. In this work, we introduce the solid-state displacement reaction of C3 N3 Cl3 (1,3,5-trichlorotriazine) and Li3 N into the highenergy ball milling system in order to obtain metastable C3 N4 phases with the desired stoichiometry.

2. Apparatus and experimental The mechanochemical process was operated using the SPEX 8000 mixer/mills, the ball, lid and vial were in a steel type, and the gasket was adapted properly. In a typical procedure, 1.5 g C3 N3 Cl3 and 0.33 g Li3 N were put together into milling vial with volume of 100 ml, 100 g steel balls with the radius of 2.0 mm were also put into the vial to the half of its volume. Then the vial was sealed tightly. The above manipulations were performed in a glove box with flowing nitrogen gas. Three samples were milled for different times, two for 10 h (sample A and B), and other for 25 h (sample C). After that, the as-milled powder products A were kept in dry box. Samples B and C were thoroughly washed with hydrochloric acid, distilled water, and ethanol. Then, the brown products were dried in vacuum at 100 ◦ C for 3 h. For sample B the obtained powder weighted 0.43 g (about 87% of that theoretically expected for C3 N4 ). For sample C, the obtained powder weighted 0.44 g (about 89% of that theoretically expected for C3 N4 ). Field emission scanning electron microscopy (SEM, Philips XL 30 FEG) was used to investigate the morphology and structure of the obtained samples B and C. The microstructure of samples was observed by high-resolution TEM (Philips, CM 200-FEG). X-ray powder diffraction was conducted with Rigaku D/max-2400, Cu K␣ radiation. X-ray photoelectron spectroscopy (XPS, VG MK II, Al K␣) was used to investigate the element composition of the samples. IR spectrum was obtained on a Nicolet FTIR760 infrared spectrometer.

3. Result and discussion Elemental analysis for the samples was performed by normal combustion method. For the as-milled sample A (C: 14.70, N: 22.29, O: 0.11, H: 0.06 wt.%, most of the other compositions in the powders would be LiCl), the composition of the synthesized carbon nitrides was C3 N3.9 O0.02 H0.18 , the N/C atomic ratio was 1.30. The small amount of O and H elements in sample A was avoidless because the powder was extraordinarily hygroscopic. For the sample B (C: 34.66, N: 51.92, O: 10.43, H: 2.99 wt.%), the composition was C3 N3.84 O0.67 H3.11 , the N/C atomic ratio was 1.28. For sample C (C: 35.12, N: 50.04, O: 12.21, H: 2.53 wt.%), the composition was C3 N3.69 O0.78 H2.58 , the N/C atomic ratio was 1.23. These results indicated that the displacement reaction

Fig. 1. SEM images of products: (a) sample B; (b) sample C, inset shows the TEM image of sample C.

between C3 N3 Cl3 and Li3 N under high-energy ball milling conditions had been progressed almost completely. The O and H elements in the compositions of sample B and C were obviously introduced in the process of washing. Fig. 1 shows the SEM images of sample B and C, from Fig. 1(a), we can see sample B has the typical glassy morphology of carbon nitrides which can be found in others reports [5,6,13], the size dimension is in the range of 5–10 ␮m. As the melting point of C3 N3 Cl3 is only 154 ◦ C, we believe this large dimension being formed in melting-state reactions. With the prolonging of the milling time, the size dimension was reduced distinctly to about 500 nm, see Fig. 1(b), but the glassy like structure was still retained. The inset in Fig. 1(b) reveals the disordered range of the microstructure of the samples in the orientations of a–b dimension. The XRD patterns of three samples were shown in Fig. 2. The lowest line corresponds to the as-milled sample A, for which most of the peaks can be indexed as LiCl (ICSD #44273) and LiCl·H2 O (ICSD #281198). A small peak of Fe, which was introduced by the milling ball, was also observed in the patterns. By the same reason as has been described in the element analysis part, we believe that LiCl·H2 O was made by the moisture absorption of LiCl in the operation


H. Zhao et al. / Materials Science and Engineering B 122 (2005) 90–93

Fig. 2. XRD patterns of three samples: the lowest (a) attributed to sample A, and the others attributed to sample B and C.

courses. The b line corresponds to the amorphous structure of sample B, it has a broad peak centered at 2θ = 26.34◦ , ˚ this indicated that the as-synthesized the d-space is 3.38 A, carbon nitride was in the amorphous structure. For the sample C, the sharp peak positioned at 2θ = 27.51◦ , the d-space

˚ corresponds well to the (0 0 2) lattice plane of is 3.23 A, ˚ reported in Ref. [23], but the crysg-C3 N4 (d0 0 2 = 3.22 A) tallization was still not good enough to show other lattice indices. Considering the amorphous structure revealed by the HRTEM image in the inset of Fig. 1(b), we speculate the sample has the layered structure along the (0 0 2) stacking faces while it is disordered in the a–b dimension. Further additional milling cannot make obvious improvement on the crystallization of sample C; we presume that higher mechanical energies would be favored to comparatively good crystallization of this graphitic carbon nitride. As samples B and C have the very similar composition and chemical bonding, we only did XPS measurement for sample C. Fig. 3(a) is the typical wide-scan XPS spectrum, it reveals the product predominantly consists of C and N elements with small amount of O which is absorbed during the washing procedure. The gross chemical composition estimated from XPS results gave a N/C atomic ratio about 1.2, a little less than that of the element results. The deconvoluted C1s and N1s spectra are shown in Fig. 3(b) and (c). The C1s spectra are deconvulated into three lines which associated three sort of binding energy (as shown in Fig. 3(b)), we assign C1s peak at 284.5 eV to the C C bonds, the peak at 286.6 eV to

Fig. 3. XPS spectrum for sample C: (a) wide-scan XPS spectrum; (b) high-resolution XPS spectrum and the deconvolution of C1s and (c) high-resolution XPS spectrum and the deconvolution of N1s .

H. Zhao et al. / Materials Science and Engineering B 122 (2005) 90–93


crystallization process. We believe that the mechanochemical approaches would be promising for the formations of sp3 or sp2 carbon nitride phases.

Acknowledgement This work is supported by National Natural Science Foundation of China under grant No.: N1MS041J.


Fig. 4. FTIR spectrum of sample A, B and C.

the sp2 C N bonds, and peak at 288.1 eV to the C N bonds [7,14]. In Fig. 3(c), the N1s peak is composed of three components, peaks at 298.5 and 400.25 eV can be attributed to two sorts of sp2 N in C N bond form in the g-C3 N4 structure [14], 401.6 eV may be attributed to the NH2 group. Scarce of C N in C1s peak and absence of C N in N1s peak indicated that sample C has a good layered structure just as the XRD analysis has revealed. The FTIR spectrum (Fig. 4) implies the existence of the graphite-like sp2 -bonded structure for sample A, B and C. The band centered at 3428 cm−1 in sample A revealed the existence of absorbed water molecular. Broadband centered at 3329 and 3144 cm−1 in sample B and C can be attributed to NH2 or NH groups. The trace peak at 800 cm−1 belongs to s-triazine ring modes [14]. For all samples, there have two strong band centered at about 1328 and 1631 cm−1 . The former can be attributed to C N, and 1631 cm−1 came from the C N stretching mode [7,14,23]. The obvious differences between these samples in this band area is that the C N peaks is stronger and sharper than C N peaks in sample C comparing with the similar aptitude for two peaks in sample A and B. We attributed this to the better crystallization of C than that of sample A and B. The smaller peaks at 2167 cm−1 , which belongs to C N in C than that of A and B, can also approve this conclusion. 4. Conclusion In summary, we have successfully synthesized comparatively well-crystallized graphitic carbon nitride (g-C3 N4 ) by facile mechanochemical reaction of C3 N3 Cl3 and Li3 N. The samples with the atomic ratio of 1.23–1.30 for N to C is very similar to the stoichiometry of C3 N4 . All the characterizations have identified the formation of g-C3 N4 and its further

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]

[21] [22] [23] [24] [25] [26] [27] [28] [29]

A.Y. Liu, M.L. Cohen, Science 245 (1989) 841. D.M. Teter, R.J. Hemley, Science 271 (1996) 53. J.B. Wang, J.L. Lei, R.H. Wang, Phys. Rev. B. 58 (1998) 11890. P. Kroll, R. Hoffmann, J. Am. Chem. Soc. 121 (1999) 4696. V.N. Khabashesku, J.L. Zimmerman, J.L. Margrave, Chem. Mater. 12 (2000) 3264. E.G. Gillan, Chem. Mater. 12 (2000) 3906. Q. Lv, C.B. Cao, C. Li, J.T. Zhang, H.S. Zhu, X. Kong, X.F. Duan, J. Mater. Chem. 13 (2003) 1241. P. Mori-S´anchez, M. Marqu´es, A. Beltr´an, J.Z. Jiang, L. Gerward, J.M. Recio, Phys. Rev. B. 68 (2003) 064115. Y. Qiu, L. Gao, Chem. Commun. (2003) 2378. L.W. Yin, Y. Bando, M.S. Li, Y.X. Liu, Y.X. Qi, Adv. Mater. 15 (2003) 1840. D.R. Miller, D.C. Swenson, E.G. Gillan, J. Am. Chem. Soc 126 (2004) 5372. E. Kroke, M. Schwarz, Coord. Chem. Rev. 248 (2004) 493. M.H.V. Huynh, M.A. Hiskey, J.G. Archuleta, E.L. Roemer, Angew. Chem. Int. Ed. 44 (2005) 737. Q.X. Guo, Y. Xie, X.J. Wang, S.Y. Zhang, T. Hou, S.C. Lv, Chem. Commun. (2004) 26. M. Kawaguchi, Y. Tokimatsu, K. Nozaki, Y. Kaburagi, Y. Hishiyama, Chem. Lett. (1997) 1003. Q. Fu, C.B. Cao, H.S. Zhu, Chem. Phys. Lett. 314 (1999) 223. J.L. Zimmerman, R. Williams, V.N. Khabashesku, J.L. Margrave, Nano Lett. 1 (2001) 731. T. Komatsu, T. Nakamura, J. Mater. Chem. 11 (2001) 474. T. Komatsu, J. Mater. Chem. 11 (2001) 799. H.A. Ma, X.P. Jia, L.X. Chen, P.W. Zhu, W.L. Guo, X.B. Guo, Y.D. Wang, S.Q. Li, G.T. Zou, G. Zhang, P. Bex, J. Phys.: Condens. Matter 14 (2002) 11269. Y.L. Gu, L.Y. Chen, L. Shi, J.H. Ma, Z.H. Yang, Y.T. Qian, Carbon 41 (2003) 2653. Y.J. Bai, B. Lv, Z.G. Liu, L. Li, D.L. Cui, X.G. Xu, Q.L. Wang, J. Cryst. Growth 247 (2003) 505. Q.X. Guo, Y. Xie, X.J. Wang, S.C. Lv, T. Hou, X.M. Liu, Chem. Phys. Lett. 380 (2003) 84. M. Kawaguchi, S. Yagi, H. Enomoto, Carbon 42 (2004) 345. C.B. Cao, F.L. Huang, C.T. Cao, J. Li, H.S. Zhu, Chem. Mater. 16 (2004) 5213. C. Suryanarayana, Prog. Mater. Sci. 46 (2001) 1. P.G. McCormick, T. Tsuzuki, J.S. Robinson, J Ding, Adv. Mater. 13 (2001) 1008. L.W. Yin, M.S. Li, Y.X. Liu, J.L. Sui, J.M. Wang, J. Phys.: Condens. Matter 15 (2003) 309. M.D. Alcal´a, J.C. S´anchez-L´opez, C. Real, A. Fern´andez, P. Matteazzi, Diamond Relat. Mater. 10 (2001) 1995.