ChemComm Published on 11 November 2013. Downloaded by Lomonosov Moscow State University on 18/01/2014 12:19:15.
COMMUNICATION
Cite this: Chem. Commun., 2014, 50, 826 Received 15th August 2013, Accepted 8th November 2013
View Article Online View Journal | View Issue
Direct observation of carbon nanostructure growth at liquid–solid interfaces† Lin-feng Fei,a Tie-yu Sun,a Wei Lu,a Xiao-qiang An,b Zhuo-feng Hu,b Jimmy C. Yu,b Ren-kui Zheng,c Xiao-min Li,c Helen L. W. Chana and Yu Wang*a
DOI: 10.1039/c3cc46264a www.rsc.org/chemcomm
Our TEM observation revealed that in a carbon–Pt3Co system, amorphous carbon first crystallized into nanoclusters at step-edges on melting Pt3Co surfaces before merging into graphene layers through a kinetic restructuring via oriented-attachment, leading to the final formation of few-layered graphene nanostructures. The result obtained from density-functional theory calculations further suggested that Co atoms rather than Pt atoms acted as initial nucleation centers.
Disclosing the mechanism of chemical reaction at an atomic level is of great importance for the structural and outlook design of functional inorganic materials. The advancing in situ transmission electron microscopy (TEM) technique now offers an exclusive approach towards direct observation of the real-time process with atomic resolution.1 Diversified in situ TEM techniques have enabled direct observations of solid phase reactions,2–5 and have also made great contributions to solid–gas reactions.6,7 However, applying such in situ TEM techniques in liquid–solid reactions has been long hampered by difficulties in introducing liquids into high-vacuum TEM systems.8 One possible solution to overcome this constraint would be to employ sealed liquid cells as TEM specimens (made up of Si3N4 or SiO2)9–11 yet not only this would involve a complex specimen packaging process to obtain a good cell, the electron transmittance of the resulting cells would also be poor thus, affecting the applicability of this method. An example that may benefit from atomic resolution imaging of liquid–solid reactions could be carbon nanostructures. The most prevalent growth models for carbon materials are the
vapor–liquid–solid (VLS) mechanism and the solid–liquid–solid (SLS) mechanism in which the liquid intermediary (melting catalysts) decisively controls the growth of various nanostructures (such as nanotubes or graphene).12,13 Unfortunately, previous research has only been able to focus on catalytic growth from gas–solid interfaces.7,14–18 Herein, we propose another solution to monitor the liquid–solid interfacial process with high resolution. We demonstrated it by in situ high-resolution TEM (HRTEM) observation of the ordering transition from amorphous carbon (-C) to graphene layers on the melting Pt3Co surface. The Pt3Co– carbon liquid–solid interface was realized by in situ heating of Pt3Co nanoparticles supported on an a-C film upon their melting. To avoid sample drift in conventional high-temperature experiments so as to gain high resolution, the in situ observation was conducted on a JEM-2100F TEM equipped with a Protochips Adurot platform with a heating E-chip specimen support (the a-C film was pre-coated on the E-chip) which is capable of providing ultra-high stability, low drift at high temperature, and accurate temperature control (Scheme 1). The B9.5 nm Pt3Co nanoparticles were synthesized via simultaneous reduction of Pt4+ and Co2+ salts (see Part S1–S3, ESI†). In order to document the formation of carbon nanostructures on melting
a
Department of Applied Physics and Material Research Center, The Hong Kong Polytechnic University, Hong Kong SAR, P.R. China. E-mail: [email protected] b Department of Chemistry and Institute of Environment, Energy and Sustainability, The Chinese University of Hong Kong, Hong Kong SAR, P.R. China c State Key Laboratory of High Performance Ceramics and Superfine Microstructure, Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 200050, P.R. China † Electronic supplementary information (ESI) available: Experimental details, supplementary figures and tables. See DOI: 10.1039/c3cc46264a
826 | Chem. Commun., 2014, 50, 826--828
Scheme 1 Schematic representation and operating principles of the experimental design, showing the heating E-chip, the viewing window and its cross-sectional configuration.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 11 November 2013. Downloaded by Lomonosov Moscow State University on 18/01/2014 12:19:15.
Communication
ChemComm
Fig. 1 Image sequences showing the typical dynamics of formation of carbon nanostructures on melting Pt3Co. (a–d) Illustration of the carbon transformation process from amorphous carbon to quasi-0D nanoclusters, then 2D graphene, and finally 3D few-layered graphite nanostructures with increasing temperatures. The scale bar in (a) also applies to (b–d). (e) A panoramic micrograph of melting Pt3Co clusters surrounded by 3D few-layered graphite at 850 1C. The dashed black line indicates external boundary of as-formed graphite nanostructures. (f) Outlines the evolution.
Pt3Co properly, it would be essential to first study its melting behaviors. A survey on in situ sintering of Pt3Co nanoparticles was therefore conducted in advance and the results are summarized in Part S4 (ESI†). It can be concluded that the initial surface melting of Pt3Co nanoparticles occurred at around 550 1C. Then the particles continued to melt towards the core-part up to B850 1C. The observed eutectic temperature is obviously lower than that of bulk Pt3Co, a result of the nanosized effect (see Part S5, ESI†).19 It should be noted that in the electron diffraction pattern at 850 1C (Fig. S3k, ESI†), an additional halo ring was observed which can be assigned to the graphite (002) planes with 0.336 nm d-spacing, suggesting the localized formation of considerable graphene layers. In situ HRTEM image sequences were recorded against increasing temperatures and the remarkable intervals are shown in Fig. 1a–d, aiming at the Pt3Co–carbon interfaces as indicated by the white dashed box in Fig. 1e. Initially, the HRTEM micrograph at 450 1C in Fig. 1a reveals that well crystallized Pt3Co nanoparticles were surrounded by a-C. The Pt3Co nanocrystals exhibit a faceted shape with clear periodic lattice fringes of 0.223 nm, corresponding to the (111) planes of Pt3Co. The temperature then rose to 550 1C when the surface of the Pt3Co nanoparticles began to melt as shown in Fig. 1b. The outline of the Pt3Co crystals was more or less retained although faint (111) fringes could still be identified in most regions. Meanwhile, the outside a-C was found to be slightly crystallized into quasi-0D nanoclusters with 0.336 nm graphite (002) planes exposed (as indicated by black arrows). More importantly, most carbon clusters were formed at the step-edges on the Pt3Co surfaces, including either mono-atomic or multi-atomic step-edges as indicated by the white arrows shown in Fig. 1b–d. As the temperature increased to 650 1C (Fig. 1c), the Pt3Co nanocrystals
This journal is © The Royal Society of Chemistry 2014
experienced an obvious reshaping towards spherical configurations, and the blurred lattice fringes could hardly be detected across the entire visible region due to further melting. Simultaneously, the carbon clusters around the Pt3Co crystals aligned into larger 2D graphene sheets, with their basal (002) planes oriented parallel to the liquid–solid interface (as indicated by the dashed black ovals). A pronounced bending of the graphene layers towards the Pt3Co step-edges was found repeatedly, confirming the origin of the carbon on this liquid–solid interface. Finally, the temperature went up to 850 1C when the Pt3Co catalyst melted completely as shown in Fig. 1d and e, and a perfect layered graphite nanostructure was formed surrounding the whole Pt3Co cluster which underwent major volume contraction (see Part S4, ESI†). The complete mechanism of the formation of carbon nanostructures on a liquid–solid interface was thus disclosed. The graphene nanostructure was formed on the melting catalyst through edge-mediated nucleation in addition to further growth via oriented attachment (OA)20 as outlined in Fig. 1f. Initially, step-edges were dynamically introduced onto the Pt3Co surface upon melting, serving as nucleation centers for forming carbon nanoclusters because of the high reactivity of the step edges.7,17 In contrast to gas–solid reactions, the as-formed carbon nanoclusters were highly mobile on the liquid–solid interface because of the high transport ability of a surficial melting catalyst. Subsequently, these nanoclusters were more likely to encounter each other, and would eventually interact mutually to coalesce into 2D graphene layers by spontaneously aligning along the graphite (002) planes, known as the typical OA process.11,21 Further growth of carbon would involve transporting C clusters towards the interface and Pt3Co atoms away from the interface. The process would eventually stop when the graphene
Chem. Commun., 2014, 50, 826--828 | 827
View Article Online
Published on 11 November 2013. Downloaded by Lomonosov Moscow State University on 18/01/2014 12:19:15.
ChemComm
Fig. 2 Illustration of the nucleation process on the Pt3Co (111) surface modeled using density-functional theory calculations. (a) The model includes 2 2 units and specific positions of the first carbon atom; the inset shows the top view. (b) Energy change in the diffusion process of the first carbon atom from A–B, A–C and A–D.
layers have encapsulated the Pt3Co surface to form a 3D carbon nanostructure. In fact, defects (such as edges and dislocations) often trigger the growth of carbon crystals. Recent theoretical studies show that the formation of graphene layers should start from the edge of the matrix,22,23 and our observation agrees well with this claim. This process can be viewed as a kind of physical position selection process in growth. When the chemical environment on the edges was taken into account, certain differences between the Co atoms and the Pt atoms in the catalytic process could also be expected. To track the origin of this interfacial process, density-functional theory (DFT) calculations were performed for the nucleation of carbon near the edge. The model structure is shown in Fig. 2a, in which the deep green spheres represent Pt atoms, and the light blue spheres stand for Co atoms. We selected four specific positions A, B, C and D (with A and D being far away from the edge while B and C being connected to the edges) for the first carbon atom on the surface. We then analyzed the diffusion process at these four positions in order to determine the position of nucleation. Fig. 2b shows the energy change in the diffusion of one carbon atom at these four positions (the energy of position A is set to zero). Generally, lower energy suggests a more stable state and lower potential barrier implies more efficient and effective diffusion. It can be seen from Fig. 2b that position B has the lowest energy level of B 0.5 eV, while the energy levels of positions A, C, and D are very similar to each other. The results suggest that if the first carbon is connected to a Co atom, the structure will possess least energy and thus be most stable. The findings from the potential barriers also show that the diffusion from A to B is more efficient than that from A to C. Considering that position B can also minimize the total amount of energy; we deduce that position B should be the nucleation position for the growth of graphite layers. The result suggests stronger chemical binding between Co and C atoms, which is in accordance with the fact that the chemical affinity of an element to C is higher when it possesses a larger number of unfilled d orbitals. In summary, this study presents a combination of in situ HRTEM observations and DFT simulations to resolve graphene nucleation and growth on liquid–solid interfaces for the first
828 | Chem. Commun., 2014, 50, 826--828
Communication
time. Carbon clusters were found to nucleate near Co atoms at the step-edges on the surface of Pt3Co, a result of associated physical and chemical position selections. Further growth of graphene layers from clusters is of kinetic nature, guided by classic OA theory. This unique simultaneous nucleation and merging process of carbon growth differs greatly from those in liquid–gas interactions where carbon fibers/tubes are developed gradually from single defects due to a constant carbon supply from the environment. As the mechanism was thoroughly identified with simple TEM configuration as such, our developed technique should be of general importance for understanding liquid–solid reactions (see Part S6, ESI†), and should inspire further attempt to tailor the growth of carbon nanostructures and other materials in future. This work was supported by the Hong Kong Polytechnic University (A-PK29 & A-PL53). The support from the Shanghai Institute of Ceramics of CAS (SKL201101SIC) and National 863 Program (No.2013AA031903) is also acknowledged.
Notes and references 1 E. Stach, Mater. Today, 2008, 11, 50–58. 2 A. H. Zewail, Science, 2010, 328, 187–193. 3 V. Radmilovic, C. Ophus, E. A. Marquis, M. D. Rossell, A. Tolley, A. Gautam, M. Asta and U. Dahmen, Nat. Mater., 2011, 10, 710–715. 4 J. C. Meyer, C. O. Girit, M. F. Crommie and A. Zettl, Nature, 2008, 454, 319–322. 5 C.-Y. Wen, M. C. Reuter, J. Bruley, J. Tersoff, S. Kodambaka, E. A. Stach and F. M. Ross, Science, 2009, 326, 1247–1250. 6 P. L. Hansen, J. B. Wagner, S. Helveg, J. R. Rostrup-Nielsen, B. S. Clausen and H. Topsøe, Science, 2002, 295, 2053–2055. ´pez-Cartes, J. Sehested, P. L. Hansen, B. S. Clausen, 7 S. Helveg, C. Lo J. R. Rostrup-Nielsen, F. Abild-Pedersen and J. K. Nørskov, Nature, 2004, 427, 426–429. 8 N. De Jonge and F. M. Ross, Nat. Nanotechnol., 2011, 6, 695–704. 9 J. M. Yuk, J. Park, P. Ercius, K. Kim, D. J. Hellebusch, M. F. Crommie, J. Y. Lee, A. Zettl and A. P. Alivisatos, Science, 2012, 336, 61–64. 10 H.-G. Liao, L. Cui, S. Whitelam and H. Zheng, Science, 2012, 336, 1011–1014. 11 D. Li, M. H. Nielsen, J. R. I. Lee, C. Frandsen, J. F. Banfield and J. J. De Yoreo, Science, 2012, 336, 1014–1018. 12 A. J. Page, Y. Ohta, S. Irle and K. Morokuma, Acc. Chem. Res., 2010, 43, 1375–1385. 13 A. Gorbunov, O. Jost, W. Pompe and A. Graff, Carbon, 2002, 40, 113–118. 14 H. Yoshida, S. Takeda, T. Uchiyama, H. Kohno and Y. Homma, Nano Lett., 2008, 8, 2082–2086. 15 J. A. Rodrı´guez-Manzo, M. Terrones, H. Terrones, H. W. Kroto, L. Sun and F. Banhart, Nat. Nanotechnol., 2007, 2, 307–311. 16 M. Lin, J. P. Ying Tan, C. Boothroyd, K. P. Loh, E. S. Tok and Y.-L. Foo, Nano Lett., 2006, 6, 449–452. ¨schner, L. Schultz, ¨ffel, M. Ru ¨mmeli, E. Mohn, C. Ta 17 D. Pohl, F. Scha C. Kisielowski and B. Rellinghaus, Phys. Rev. Lett., 2011, 107, 185501. 18 Z. Peng, F. Somodi, S. Helveg, C. Kisielowski, P. Specht and A. T. Bell, J. Catal., 2012, 286, 22–29. 19 C. L. Chen, J.-G. Lee, K. Arakawa and H. Mori, Appl. Phys. Lett., 2011, 98, 083108. 20 R. L. Penn and J. F. Banfield, Science, 1998, 281, 969–971. 21 J. Zhang, F. Huang and Z. Lin, Nanoscale, 2010, 2, 18–34. 22 H. Chen, W. Zhu and Z. Zhang, Phys. Rev. Lett., 2010, 104, 186101. 23 S. Saadi, F. Abild-Pedersen, S. Helveg, J. Sehested, B. Hinnemann, C. C. Appel and J. K. Nørskov, J. Phys. Chem. C, 2010, 114, 11221–11227.
This journal is © The Royal Society of Chemistry 2014
COMMUNICATION
Cite this: Chem. Commun., 2014, 50, 826 Received 15th August 2013, Accepted 8th November 2013
View Article Online View Journal | View Issue
Direct observation of carbon nanostructure growth at liquid–solid interfaces† Lin-feng Fei,a Tie-yu Sun,a Wei Lu,a Xiao-qiang An,b Zhuo-feng Hu,b Jimmy C. Yu,b Ren-kui Zheng,c Xiao-min Li,c Helen L. W. Chana and Yu Wang*a
DOI: 10.1039/c3cc46264a www.rsc.org/chemcomm
Our TEM observation revealed that in a carbon–Pt3Co system, amorphous carbon first crystallized into nanoclusters at step-edges on melting Pt3Co surfaces before merging into graphene layers through a kinetic restructuring via oriented-attachment, leading to the final formation of few-layered graphene nanostructures. The result obtained from density-functional theory calculations further suggested that Co atoms rather than Pt atoms acted as initial nucleation centers.
Disclosing the mechanism of chemical reaction at an atomic level is of great importance for the structural and outlook design of functional inorganic materials. The advancing in situ transmission electron microscopy (TEM) technique now offers an exclusive approach towards direct observation of the real-time process with atomic resolution.1 Diversified in situ TEM techniques have enabled direct observations of solid phase reactions,2–5 and have also made great contributions to solid–gas reactions.6,7 However, applying such in situ TEM techniques in liquid–solid reactions has been long hampered by difficulties in introducing liquids into high-vacuum TEM systems.8 One possible solution to overcome this constraint would be to employ sealed liquid cells as TEM specimens (made up of Si3N4 or SiO2)9–11 yet not only this would involve a complex specimen packaging process to obtain a good cell, the electron transmittance of the resulting cells would also be poor thus, affecting the applicability of this method. An example that may benefit from atomic resolution imaging of liquid–solid reactions could be carbon nanostructures. The most prevalent growth models for carbon materials are the
vapor–liquid–solid (VLS) mechanism and the solid–liquid–solid (SLS) mechanism in which the liquid intermediary (melting catalysts) decisively controls the growth of various nanostructures (such as nanotubes or graphene).12,13 Unfortunately, previous research has only been able to focus on catalytic growth from gas–solid interfaces.7,14–18 Herein, we propose another solution to monitor the liquid–solid interfacial process with high resolution. We demonstrated it by in situ high-resolution TEM (HRTEM) observation of the ordering transition from amorphous carbon (-C) to graphene layers on the melting Pt3Co surface. The Pt3Co– carbon liquid–solid interface was realized by in situ heating of Pt3Co nanoparticles supported on an a-C film upon their melting. To avoid sample drift in conventional high-temperature experiments so as to gain high resolution, the in situ observation was conducted on a JEM-2100F TEM equipped with a Protochips Adurot platform with a heating E-chip specimen support (the a-C film was pre-coated on the E-chip) which is capable of providing ultra-high stability, low drift at high temperature, and accurate temperature control (Scheme 1). The B9.5 nm Pt3Co nanoparticles were synthesized via simultaneous reduction of Pt4+ and Co2+ salts (see Part S1–S3, ESI†). In order to document the formation of carbon nanostructures on melting
a
Department of Applied Physics and Material Research Center, The Hong Kong Polytechnic University, Hong Kong SAR, P.R. China. E-mail: [email protected] b Department of Chemistry and Institute of Environment, Energy and Sustainability, The Chinese University of Hong Kong, Hong Kong SAR, P.R. China c State Key Laboratory of High Performance Ceramics and Superfine Microstructure, Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 200050, P.R. China † Electronic supplementary information (ESI) available: Experimental details, supplementary figures and tables. See DOI: 10.1039/c3cc46264a
826 | Chem. Commun., 2014, 50, 826--828
Scheme 1 Schematic representation and operating principles of the experimental design, showing the heating E-chip, the viewing window and its cross-sectional configuration.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 11 November 2013. Downloaded by Lomonosov Moscow State University on 18/01/2014 12:19:15.
Communication
ChemComm
Fig. 1 Image sequences showing the typical dynamics of formation of carbon nanostructures on melting Pt3Co. (a–d) Illustration of the carbon transformation process from amorphous carbon to quasi-0D nanoclusters, then 2D graphene, and finally 3D few-layered graphite nanostructures with increasing temperatures. The scale bar in (a) also applies to (b–d). (e) A panoramic micrograph of melting Pt3Co clusters surrounded by 3D few-layered graphite at 850 1C. The dashed black line indicates external boundary of as-formed graphite nanostructures. (f) Outlines the evolution.
Pt3Co properly, it would be essential to first study its melting behaviors. A survey on in situ sintering of Pt3Co nanoparticles was therefore conducted in advance and the results are summarized in Part S4 (ESI†). It can be concluded that the initial surface melting of Pt3Co nanoparticles occurred at around 550 1C. Then the particles continued to melt towards the core-part up to B850 1C. The observed eutectic temperature is obviously lower than that of bulk Pt3Co, a result of the nanosized effect (see Part S5, ESI†).19 It should be noted that in the electron diffraction pattern at 850 1C (Fig. S3k, ESI†), an additional halo ring was observed which can be assigned to the graphite (002) planes with 0.336 nm d-spacing, suggesting the localized formation of considerable graphene layers. In situ HRTEM image sequences were recorded against increasing temperatures and the remarkable intervals are shown in Fig. 1a–d, aiming at the Pt3Co–carbon interfaces as indicated by the white dashed box in Fig. 1e. Initially, the HRTEM micrograph at 450 1C in Fig. 1a reveals that well crystallized Pt3Co nanoparticles were surrounded by a-C. The Pt3Co nanocrystals exhibit a faceted shape with clear periodic lattice fringes of 0.223 nm, corresponding to the (111) planes of Pt3Co. The temperature then rose to 550 1C when the surface of the Pt3Co nanoparticles began to melt as shown in Fig. 1b. The outline of the Pt3Co crystals was more or less retained although faint (111) fringes could still be identified in most regions. Meanwhile, the outside a-C was found to be slightly crystallized into quasi-0D nanoclusters with 0.336 nm graphite (002) planes exposed (as indicated by black arrows). More importantly, most carbon clusters were formed at the step-edges on the Pt3Co surfaces, including either mono-atomic or multi-atomic step-edges as indicated by the white arrows shown in Fig. 1b–d. As the temperature increased to 650 1C (Fig. 1c), the Pt3Co nanocrystals
This journal is © The Royal Society of Chemistry 2014
experienced an obvious reshaping towards spherical configurations, and the blurred lattice fringes could hardly be detected across the entire visible region due to further melting. Simultaneously, the carbon clusters around the Pt3Co crystals aligned into larger 2D graphene sheets, with their basal (002) planes oriented parallel to the liquid–solid interface (as indicated by the dashed black ovals). A pronounced bending of the graphene layers towards the Pt3Co step-edges was found repeatedly, confirming the origin of the carbon on this liquid–solid interface. Finally, the temperature went up to 850 1C when the Pt3Co catalyst melted completely as shown in Fig. 1d and e, and a perfect layered graphite nanostructure was formed surrounding the whole Pt3Co cluster which underwent major volume contraction (see Part S4, ESI†). The complete mechanism of the formation of carbon nanostructures on a liquid–solid interface was thus disclosed. The graphene nanostructure was formed on the melting catalyst through edge-mediated nucleation in addition to further growth via oriented attachment (OA)20 as outlined in Fig. 1f. Initially, step-edges were dynamically introduced onto the Pt3Co surface upon melting, serving as nucleation centers for forming carbon nanoclusters because of the high reactivity of the step edges.7,17 In contrast to gas–solid reactions, the as-formed carbon nanoclusters were highly mobile on the liquid–solid interface because of the high transport ability of a surficial melting catalyst. Subsequently, these nanoclusters were more likely to encounter each other, and would eventually interact mutually to coalesce into 2D graphene layers by spontaneously aligning along the graphite (002) planes, known as the typical OA process.11,21 Further growth of carbon would involve transporting C clusters towards the interface and Pt3Co atoms away from the interface. The process would eventually stop when the graphene
Chem. Commun., 2014, 50, 826--828 | 827
View Article Online
Published on 11 November 2013. Downloaded by Lomonosov Moscow State University on 18/01/2014 12:19:15.
ChemComm
Fig. 2 Illustration of the nucleation process on the Pt3Co (111) surface modeled using density-functional theory calculations. (a) The model includes 2 2 units and specific positions of the first carbon atom; the inset shows the top view. (b) Energy change in the diffusion process of the first carbon atom from A–B, A–C and A–D.
layers have encapsulated the Pt3Co surface to form a 3D carbon nanostructure. In fact, defects (such as edges and dislocations) often trigger the growth of carbon crystals. Recent theoretical studies show that the formation of graphene layers should start from the edge of the matrix,22,23 and our observation agrees well with this claim. This process can be viewed as a kind of physical position selection process in growth. When the chemical environment on the edges was taken into account, certain differences between the Co atoms and the Pt atoms in the catalytic process could also be expected. To track the origin of this interfacial process, density-functional theory (DFT) calculations were performed for the nucleation of carbon near the edge. The model structure is shown in Fig. 2a, in which the deep green spheres represent Pt atoms, and the light blue spheres stand for Co atoms. We selected four specific positions A, B, C and D (with A and D being far away from the edge while B and C being connected to the edges) for the first carbon atom on the surface. We then analyzed the diffusion process at these four positions in order to determine the position of nucleation. Fig. 2b shows the energy change in the diffusion of one carbon atom at these four positions (the energy of position A is set to zero). Generally, lower energy suggests a more stable state and lower potential barrier implies more efficient and effective diffusion. It can be seen from Fig. 2b that position B has the lowest energy level of B 0.5 eV, while the energy levels of positions A, C, and D are very similar to each other. The results suggest that if the first carbon is connected to a Co atom, the structure will possess least energy and thus be most stable. The findings from the potential barriers also show that the diffusion from A to B is more efficient than that from A to C. Considering that position B can also minimize the total amount of energy; we deduce that position B should be the nucleation position for the growth of graphite layers. The result suggests stronger chemical binding between Co and C atoms, which is in accordance with the fact that the chemical affinity of an element to C is higher when it possesses a larger number of unfilled d orbitals. In summary, this study presents a combination of in situ HRTEM observations and DFT simulations to resolve graphene nucleation and growth on liquid–solid interfaces for the first
828 | Chem. Commun., 2014, 50, 826--828
Communication
time. Carbon clusters were found to nucleate near Co atoms at the step-edges on the surface of Pt3Co, a result of associated physical and chemical position selections. Further growth of graphene layers from clusters is of kinetic nature, guided by classic OA theory. This unique simultaneous nucleation and merging process of carbon growth differs greatly from those in liquid–gas interactions where carbon fibers/tubes are developed gradually from single defects due to a constant carbon supply from the environment. As the mechanism was thoroughly identified with simple TEM configuration as such, our developed technique should be of general importance for understanding liquid–solid reactions (see Part S6, ESI†), and should inspire further attempt to tailor the growth of carbon nanostructures and other materials in future. This work was supported by the Hong Kong Polytechnic University (A-PK29 & A-PL53). The support from the Shanghai Institute of Ceramics of CAS (SKL201101SIC) and National 863 Program (No.2013AA031903) is also acknowledged.
Notes and references 1 E. Stach, Mater. Today, 2008, 11, 50–58. 2 A. H. Zewail, Science, 2010, 328, 187–193. 3 V. Radmilovic, C. Ophus, E. A. Marquis, M. D. Rossell, A. Tolley, A. Gautam, M. Asta and U. Dahmen, Nat. Mater., 2011, 10, 710–715. 4 J. C. Meyer, C. O. Girit, M. F. Crommie and A. Zettl, Nature, 2008, 454, 319–322. 5 C.-Y. Wen, M. C. Reuter, J. Bruley, J. Tersoff, S. Kodambaka, E. A. Stach and F. M. Ross, Science, 2009, 326, 1247–1250. 6 P. L. Hansen, J. B. Wagner, S. Helveg, J. R. Rostrup-Nielsen, B. S. Clausen and H. Topsøe, Science, 2002, 295, 2053–2055. ´pez-Cartes, J. Sehested, P. L. Hansen, B. S. Clausen, 7 S. Helveg, C. Lo J. R. Rostrup-Nielsen, F. Abild-Pedersen and J. K. Nørskov, Nature, 2004, 427, 426–429. 8 N. De Jonge and F. M. Ross, Nat. Nanotechnol., 2011, 6, 695–704. 9 J. M. Yuk, J. Park, P. Ercius, K. Kim, D. J. Hellebusch, M. F. Crommie, J. Y. Lee, A. Zettl and A. P. Alivisatos, Science, 2012, 336, 61–64. 10 H.-G. Liao, L. Cui, S. Whitelam and H. Zheng, Science, 2012, 336, 1011–1014. 11 D. Li, M. H. Nielsen, J. R. I. Lee, C. Frandsen, J. F. Banfield and J. J. De Yoreo, Science, 2012, 336, 1014–1018. 12 A. J. Page, Y. Ohta, S. Irle and K. Morokuma, Acc. Chem. Res., 2010, 43, 1375–1385. 13 A. Gorbunov, O. Jost, W. Pompe and A. Graff, Carbon, 2002, 40, 113–118. 14 H. Yoshida, S. Takeda, T. Uchiyama, H. Kohno and Y. Homma, Nano Lett., 2008, 8, 2082–2086. 15 J. A. Rodrı´guez-Manzo, M. Terrones, H. Terrones, H. W. Kroto, L. Sun and F. Banhart, Nat. Nanotechnol., 2007, 2, 307–311. 16 M. Lin, J. P. Ying Tan, C. Boothroyd, K. P. Loh, E. S. Tok and Y.-L. Foo, Nano Lett., 2006, 6, 449–452. ¨schner, L. Schultz, ¨ffel, M. Ru ¨mmeli, E. Mohn, C. Ta 17 D. Pohl, F. Scha C. Kisielowski and B. Rellinghaus, Phys. Rev. Lett., 2011, 107, 185501. 18 Z. Peng, F. Somodi, S. Helveg, C. Kisielowski, P. Specht and A. T. Bell, J. Catal., 2012, 286, 22–29. 19 C. L. Chen, J.-G. Lee, K. Arakawa and H. Mori, Appl. Phys. Lett., 2011, 98, 083108. 20 R. L. Penn and J. F. Banfield, Science, 1998, 281, 969–971. 21 J. Zhang, F. Huang and Z. Lin, Nanoscale, 2010, 2, 18–34. 22 H. Chen, W. Zhu and Z. Zhang, Phys. Rev. Lett., 2010, 104, 186101. 23 S. Saadi, F. Abild-Pedersen, S. Helveg, J. Sehested, B. Hinnemann, C. C. Appel and J. K. Nørskov, J. Phys. Chem. C, 2010, 114, 11221–11227.
This journal is © The Royal Society of Chemistry 2014
