ARTICLE IN PRESS
Journal of Crystal Growth 282 (2005) 105–111 www.elsevier.com/locate/jcrysgro
CuO nanostructures prepared by a chemical method D. Lia, Y.H. Leunga, A.B. Djurisˇ ic´a,, Z.T. Liua, M.H. Xiea, J. Gaoa, W.K. Chanb b
a Department of Physics, The University of Hong Kong, Pokfulam Road, Hong Kong Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong
Received 17 December 2004; received in revised form 19 April 2005; accepted 25 April 2005 Available online 6 June 2005 Communicated by Prof. J.M. Redwing
Abstract We investigated the properties of CuO nanostructures fabricated from copper(II) nitrate hydrate solutions as a function of synthesis temperature, concentration, and pH value of the solution. The properties of the fabricated nanostructures were studied using scanning electron microscopy, X-ray diffraction, and transmission electron microscopy. We found that the morphology of the obtained structures is strongly dependent on both pH value of the solution and the synthesis temperature. Synthesis conditions, such as solution concentration, temperature, and pH value, also affect the adhesion of the fabricated structures to the substrate, which is of importance for practical applications. r 2005 Elsevier B.V. All rights reserved. Keywords: A1. Nanostructures; B1. Oxides
1. Introduction CuO is a narrow band-gap p-type semiconductor. Its interesting properties make it an important material for a variety of practical applications, such as catalysis, batteries, solar energy conversion, gas sensing, and field emission. A number of different fabrication techniques were reported for CuO nanostructures [1–11]. Synthesis of spherical structures consisting of self-assembled CuO naCorresponding author. Tel.: +852 2859 7946;
fax: +852 2559 9152. E-mail address:
[email protected] (A.B. Djurisˇ ic´).
noribbons was reported [1]. CuO nanowires were synthesized by thermal oxidation of Cu substrates [2]. The growth of CuO nanofibers from electrodeposited Cu nuclei was also reported [3,4]. CuO nanotubes and nanorods were synthesized by hydrothermal treatment of Cu(OH)2 precursor 4 [5]. CuO nanorods were also prepared by thermal decomposition [6] or dehydration [7] of Cu(OH)2. CuO nanoribbons obtained from sacrificial templates of Cu2(OH)2CO3 [8,9] and Cu2Cl(OH)3 [10] were also reported. CuO nanorods and nanoribbons were also synthesized at moderate temperature (77–82 1C) in water–ethanol solutions [11]. Among various synthesis methods, hydrothermal
0022-0248/$ - see front matter r 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jcrysgro.2005.04.090
ARTICLE IN PRESS 106
D. Li et al. / Journal of Crystal Growth 282 (2005) 105–111
or chemical reaction methods are of interest because they are safe and environmental-friendly synthesis procedures performed at moderate (o200 1C) temperatures. In particular, low synthesis temperature is a significant advantage for device applications since it enables the use of flexible plastic substrates and easy integration with organic optoelectronic device fabrication. It was shown that the morphology of the obtained ZnO [12], indium tin oxide [13], titania [14], and yttria [15] nanostructures by solution-phase growth strongly depends on the pH value of the solution. The surface electrification of an oxide depends on the solution pH. Decreasing the pH will lead to adsorption of protons, while increasing the pH will lead to the absorption of hydroxyl ions [16]. The presence of surface charges will affect the aggregation and ripening of the nanostructures [14]. The aggregation of the nanostructures is dependent on the presence or absence of electrostatic repulsion, while the negative surface charge may hinder Oswald ripening process [14]. In this work, we investigated the influence of the synthesis parameters, such as temperature, the solution concentration, and pH value on the properties of obtained CuO nanostructures.
are observed by examining the leftover solution. While the characterization of the products deposited on the substrate could be performed, the deposited CuO could be partly washed away by rinsing the substrate with de-ionized water, which would not be suitable for practical applications. For improved adhesion of the synthesized products on the Si substrate, we also investigated the synthesis from a more concentrated (25 mM) solution. Before synthesis, the substrate was seeded with CuO nanoparticles with particle size in the range 50–60 nm obtained from nanostructured and amorphous materials. The seeding was performed by dispersing the nanoparticles in deionized water in an ultrasonic bath for 1–2 h, putting a drop of the dispersion on the substrate and drying the substrate in an oven at 150 1C for 2 h. Finally, in order to clarify the role of hexamethylenetetramine in the fabrication of CuO nanostructures, the experiments were performed in the absence of C6H12N4. The structure of the deposited materials was investigated by Xray diffraction (XRD) using Siemens D5000 X-ray diffractometer, scanning electron microscopy (SEM) using Cambridge 440 SEM, and Leo 1530 FESEM, transmission electron microscopy (TEM), and high resolution transmission electron microscopy (HRTEM) using JEOL 2010F TEM.
2. Experimental details The CuO nanostructures were fabricated by heating a solution of copper(II) nitrate hydrate (Cu(NO3)2 H2O) (Aldrich 99.999% purity) and hexamethylenetetramine (C6H12N4) (Aldrich 99+% purity). Equimolar solution of copper(II) nitrate hydrate and hexamethylenetetramine with a concentration of 4 mM was prepared. The pH of the solution was adjusted by the addition of nitric acid (HNO3) or sodium hydroxide (NaOH). The prepared solution was transferred into a vial and substrates (Si or ITO/glass) were placed at the bottom of the vial. The vial was then heated at 90–120 1C for 3 h. Lastly, the resulting product was dried in an oven at 100 1C. However, the obtained products from a dilute solution did not adhere well to the substrate. Also, the presence of the substrate did not seem to affect the obtained CuO morphology, since very similar morphologies
3. Results and discussion Fig. 1 shows the SEM images of CuO nanostructures synthesized at 120 1C from solutions with different pH values. It can be observed that, for low pH values, we obtained spherical structures similar to those reported by Liu and Zeng [1]. With increasing pH value, the length of the crystallites forming the sphere increases, and so does the space between the crystallites. Finally, when the pH value reaches pH ¼ 7, only very few spherical assemblies are found, and the substrate is sparsely covered by rods attached to round cores. Fig. 2 illustrates the influence of temperature on the obtained morphology for pH ¼ 6. It can be observed that the length of the rods increases and the packing density decreases with increasing temperature. The temperature dependence of the
ARTICLE IN PRESS D. Li et al. / Journal of Crystal Growth 282 (2005) 105–111
107
Fig. 2. SEM image of CuO nanostructures synthesized for pH ¼ 6 at (a) 90 1C, (b) 100 1C, and (c) 110 1C.
Fig. 1. SEM image of CuO nanostructures synthesized at 120 1C for (a) pH ¼ 3, (b) pH ¼ 4, (c) pH ¼ 5, (d) pH ¼ 6, and (e) pH ¼ 7.
obtained morphologies is likely due to the temperature dependence of the relative supersaturation [17]. For the same concentration, relative supersaturation is expected to decrease as the temperature increases. We also performed XRD, TEM, and high resolution transmission
electron microscopy (HRTEM) for the synthesized CuO nanostructures. Obtained results (for pH ¼ 6, 120 1C) are shown in Fig. 3. It can be observed that XRD data show only peaks characteristic for CuO. HRTEM images show that the crystallites in the spherical assembly have a single-crystal core surrounded by a thin amorphous layer. We also investigated the influence of the solution concentration, and we found that it has significant impact on the obtained morphology. For synthesis temperature 120 1C and pH ¼ 6, concentrations 1 mM and 2 mM do not result in formation of spherical ensembles. The obtained structures are irregular and very small (the largest dimensiono300 nm). For 3 mM concentration, some spherical structures can be observed, but
ARTICLE IN PRESS 108
D. Li et al. / Journal of Crystal Growth 282 (2005) 105–111
Fig. 3. (a) TEM image of the CuO nanostructures synthesized for pH ¼ 6 and 90 1C. (b) HRTEM of the CuO rod (c) XRD spectrum.
the individual crystallites appear to be more closely packed than for 4 mM concentration. On the other hand, when the concentration was increased to 25 mM and the substrate was treated with CuO seeds, entirely different morphology and behavior with the pH value change were obtained. It was proposed that the growth of the nanostructures is determined by the chemical potential of OH [15]. The chemical potential is determined by the pH value and monomer concentrations [15,17]. Thus, it is expected that both pH and solution concentration will affect the obtained morphologies. In addition to the change of concentration, the growth is also affected by the presence of CuO nanoparticle seeds, since the number of available nuclei will affect the obtained
morphology [16]. Fig. 4 shows the SEM images of the CuO nanostructures fabricated from a 25 mM solution on a Si substrate with CuO seeds. It can be observed that the pH value of the solution strongly affects the obtained morphology. However, in this case for a low pH value, we did not observe spherical assemblies. With increasing pH value, structures formed by self-assembly of CuO platelets are more frequently found. Finally, for pH values of 6 and 7, some spherical assemblies can be found. From the incomplete assemblies, it can be seen that there is no common core, similar to previously reported results [1]. However, the distance between the ends of CuO platelets is smaller. Also, for pH ¼ 7, thicker platelets with a rectangular cross-section can be observed in the
ARTICLE IN PRESS D. Li et al. / Journal of Crystal Growth 282 (2005) 105–111
Fig. 4. SEM images of CuO nanostructures synthesized from solution with concentration 25 mM at 90 1C for (a) pH ¼ 3, (b) pH ¼ 4, c) pH ¼ 5, (d) pH ¼ 6, and e) pH ¼ 7.
spherical assemblies, which is different from previously reported rhombic crystal strips [1]. It was proposed that the thin rhombic CuO strips self-assemble into spheres due to geometric shape of the building blocks [1]. In this work, we obtained assemblies with some central symmetry for different shapes of the building blocks for the majority of investigated conditions. Therefore,
109
there likely exists an additional mechanism responsible for the self-assembly of these structures. It was proposed previously that the presence of NaOH as a strong electrolyte may neutralize the surface charges in CuO and affect aggregation [11]. NaOH can also induce additional growth anisotropy, affecting the morphology of synthesized product [11]. Thus, since we found that the morphology of individual platelets and their packing density are strongly affected by the pH value of the solution, it is possible that the selfassembly of the CuO nanostructures is electrostatically driven. This is confirmed by the fact that the spherical assemblies are very stable indicating strong binding between individual platelets. A previous study of CuO microspheres synthesized by a different method reported that the microspheres remain stable after 6 h of sonication in an ultrasonic bath [1]. Therefore, it is likely that electrostatic attraction plays a role in the assembly of these structures. Further study is needed to fully explain the assembly of CuO platelets with different shapes and sizes and the dependence of the process on the pH value of the solution. To clarify the role of C6H12N4, the samples from 4 mM solutions without hexamethylenetetramine were prepared. It has been demonstrated that the addition of organic compounds, such as ethylene glycol, significantly affects the obtained CuO morphology [18]. Also, the presence of organic ligands is known to affect the aggregation of the nanoparticles [19]. Therefore, we repeated the experiments of varying the pH in the absence of hexamethylenetetramine. The obtained results are shown in Fig. 5. The differences in the reaction products with and without hexamethylenetetramine are already obvious by naked eyes observation, since without hexamethylenetetramine very small amount of products are deposited on the substrates while solution remains clear. On the other hand, in the presence of hexamethylenetetramine a significant quantity of black deposits is found. From the SEM images, we can see that the pH value of the solution affects the morphology, but in all cases the spherical assemblies of CuO nanostructures are absent. The low quantity of the deposited products is in agreement with the low tendency of divalent
ARTICLE IN PRESS 110
D. Li et al. / Journal of Crystal Growth 282 (2005) 105–111
tetramine and Zn(NO3)2 aqueous solutions [20]. At elevated temperature, hexamethylenetetramine was hydrolyzed into formaldehyde and ammonia, which formed amino complex with Cu2+. Since the hydrolysis reaction occurred slowly, a low supersaturation could be obtained, and a high yield of CuO nanostructures was obtained. We also investigated the adhesion of the fabricated structures to the substrate and found that the adhesion was strongly dependent on the synthesis temperature, solution concentration, and presence of nanoparticle seeds. CuO nanostructures synthesized without CuO seeds and in less concentrated solutions (1–4 mM) do not adhere to the substrate. The samples prepared from 25 mM solution at temperatures of 100 1C and above, do not adhere to the substrate and can be easily washed away with de-ionized water. On the other hand, for samples synthesized at 90 1C from 25 mM solution it is still possible to have uniformly covered substrates after washing with de-ionized water, even though some of the material is removed. For 80 1C, sonicating the sample in ultrasonic bath does not remove the film. Similar results are obtained on Si and indium tin oxide substrates.
4. Conclusions
Fig. 5. SEM image of CuO nanostructures synthesized without C6H12N4 at 120 1C for (a) pH ¼ 3, (b) pH ¼ 4, (c) pH ¼ 5, (d) pH ¼ 6, and (e) pH ¼ 7.
metal ions to precipitate in aqueous solution by hydrolysis–condensation reaction [20]. It is likely that the CuO nanostructures synthesized in the presence of hexamethylenetetramine are formed by aqueous thermal decomposition of Cu2+ amino complex, similar to the well known method for synthesis of ZnO nanorods from hexamethylene-
We investigated the influence of the synthesis conditions on the morphology of the CuO nanostructures. We found that the solution concentration, pH value, synthesis temperature, and presence of seeds significantly affect obtained morphology as well as adhesion to the substrate. In the absence of seed and for low solution concentrations, spherical assemblies of CuO platelets are obtained. The shape of platelets and density of their packing are strongly affected by the pH value and concentration of the solution, as well as synthesis temperature. On the other hand, synthesis from solutions with higher concentrations enables fabrications of nanostructured films on the substrate, whose morphology and adherence to the substrate depend on the pH value and the synthesis temperature.
ARTICLE IN PRESS D. Li et al. / Journal of Crystal Growth 282 (2005) 105–111
References [1] B. Liu, H.C. Zheng, J. Am. Chem. Soc. 126 (2004) 8124. [2] H. Jiang, T. Herricks, Y. Xia, Nano. Lett. 2 (2002) 1333. [3] H.-H. Lin, C.-Y. Wang, H.C. Shih, J.-M. Chen, C.-T. Hsieh, J. Appl. Phys. 95 (2004) 5889. [4] C.-T. Hsieh, J.-M. Chen, H.-H. Lin, H.C. Shih, Appl. Phys. Lett. 82 (2003) 3316. [5] M.H. Cao, C.W. Hu, Y.H. Wang, Y.H. Guo, C.X. Guo, E.B. Wang, Chem. Commun. 15 (2003) 1884. [6] X.P. Gao, J.L. Bao, G.L. Pan, H.Y. Zhu, P.X. Huang, F. Wu, D.Y. Song, J. Phys. Chem. B 108 (2004) 5547. [7] X. Song, S. Sun, W. Zhang, H. Yu, W. Fan, J. Phys. Chem. B 108 (2004) 5200. [8] C.L. Zhu, C.N. Chen, L.Y. Hao, Y. Hu, Z.Y. Chen, Solid State Commun 130 (2004) 681. [9] L. Zhang, J.C. Yu, A.-W. Xu, Q. Li, K.W. Kwong, S.-H. Yu, J. Crystal Growth 266 (2004) 545.
111
[10] C.L. Zhu, C.N. Chen, L.Y. Hao, Y. Hu, Z.Y. Chen, J. Crystal Growth 263 (2004) 473. [11] Y. Chang, H.C. Zeng, Crystal Growth and Design 4 (2004) 397. [12] Z. Wang, X.-F. Qian, J. Yin, Z. Zhu, J. Solid State Chem. 177 (2004) 2144. [13] K.Y. Kim, S.B. Park, Mater. Chem. Phys. 86 (2004) 210. [14] T. Boiadijieva, G. Cappelletti, S. Ardizzone, S. Rondinini, A. Vertova, Phys. Chem. Chem. Phys. 6 (2004) 3535. [15] Q. Li, C. Feng, Q. Jiao, L. Guo, C. Liu, H.B. Xu, Phys. Stat. Sol. (a) 201 (2004) 3055. [16] L. Vayssieres, Int. J. Mater. Prod. Technol. 18 (2003) 313. [17] Z.A. Peng, X. Peng, J. Am. Chem. Soc. 124 (2002) 3343. [18] S. Li, H. Zhang, Y. Ji, D. Yang, Nanotechnology 15 (2004) 1428. [19] Z. Zhang, H. Sun, X. Shao, D. Li, H. Yu, M. Han, Adv. Mater. 17 (2005) 42. [20] L. Vayssieres, K. Keis, S.-E. Lindquist, A. Hagfeldt, J. Phys. Chem. B 105 (2001) 3350.
