Formation of Colloidal CuO Nanocrystallites and Their Spherical

mechanism has been revealed: (i) formation of CuO nanocrystallites; (ii) spherical aggregation of primary. CuO crystallites; (iii) reductive conversio...
0 downloads 0 Views 876KB Size
Download PDF
1074

Langmuir 2005, 21, 1074-1079

Formation of Colloidal CuO Nanocrystallites and Their Spherical Aggregation and Reductive Transformation to Hollow Cu2O Nanospheres Yu Chang, Joong Jiat Teo, and Hua Chun Zeng* Department of Chemical and Biomolecular Engineering, Faculty of Engineering, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260 Received September 18, 2004. In Final Form: November 1, 2004 In this work, we demonstrate that cuprous oxide Cu2O nanospheres with hollow interiors can be fabricated from a reductive conversion of aggregated CuO nanocrystallites without using templates. A detailed process mechanism has been revealed: (i) formation of CuO nanocrystallites; (ii) spherical aggregation of primary CuO crystallites; (iii) reductive conversion of CuO to Cu2O; and (iv) crystal aging and hollowing of Cu2O nanospheres. In this template-free process, Ostwald ripening is operative in (iv) for controlling crystallite size in shell structures and thus for precisely tuning the optical band gap energy (Eg) of resultant semiconductor nanostructures. For the first time, a wealth of colorful Cu2O hollow nanospheres (outer diameters in 100-200 nm), with variable Eg in the range of 2.405-2.170 eV, has been fabricated via this novel chemical route. Considering their unique hollow structure and facile tuning in band gap energy, the prepared Cu2O hollow spheres can be potentially useful for harvesting solar energy in the visible range. Possibility of fabrication of Cu-Cu2O nanocomposites has also been discussed.

Introduction In recent years, there has been an extensive interest in architecture and fabrication of hollow structures (mostly spherical in shape) with diameters in the submicrometer range.1-34 These novel functional materials consist of either organic compounds, inorganic compounds, or their * Corresponding author. E-mail: [email protected]. (1) Caruso, F.; Caruso, R. A.; Mo¨hwald, H. Science 1998, 282, 1111. (2) Go¨ltner, C. G. Angew. Chem., Int. Ed. 1999, 38, 3155. (3) Zhong, Z.; Yin, Y.; Gates, B.; Xia, Y. Adv. Mater. 2000, 12, 206. (4) Caruso, R. A.; Susha, A.; Caruso, F. Chem. Mater. 2001, 13, 400. (5) Yin, Y.; Lu, Y.; Gates, B.; Xia, Y. Chem. Mater. 2001, 13, 1146. (6) Caruso, F.; Spasova, M.; Susha, A.; Giersig, M.; Caruso, R. A. Chem. Mater. 2001, 13, 109. (7) Wang, D.; Caruso, F. Chem. Mater. 2002, 14, 1909. (8) Valtchev, V. Chem. Mater. 2002, 14, 4371. (9) Liang, Z.; Susha, A.; Caruso, F. Chem. Mater. 2003, 15, 3176. (10) Yang, Z.; Niu, Z.; Lu, Y.; Hu, Z.; Han, C. C. Angew. Chem., Int. Ed. 2003, 42, 1943. (11) Mandal, T. K.; Fleming, M. S.; Walt, D. R. Chem. Mater. 2000, 12, 3481. (12) Kim, S.-W.; Kim, M.; Lee, W. Y.; Hyeon, T. J. Am. Chem. Soc. 2002, 124, 7642. (13) Kamata, K.; Lu, Y.; Xia, Y. J. Am. Chem. Soc. 2003, 125, 2384. (14) Kim, J. Y.; Yoon, S. B.; Yu, J.-S. Chem. Commun. 2003, 790. (15) Torimoto, T.; Reyes, J. P.; Iwasaki, K.; Pal, B.; Shibayama, T.; Sugawara, K.; Takahashi, H.; Ohtani, B. J. Am. Chem. Soc. 2003, 125, 316. (16) Dinsmore, A. D.; Hsu, M. F.; Nikolaides, M. G.; Marquez, M.; Bausch, A. R.; Weitz, D. A. Science 2002, 298, 1006. (17) Kulak, A.; Lee, Y.-J.; Park, Y. S.; Kim, H. S.; Lee, G. S.; Yoon, K. B. Adv. Mater. 2002, 14, 526. (18) Naik, S. P.; Chiang, A. S. T.; Thompson, R. W.; Huang, F. C. Chem. Mater. 2003, 15, 787. (19) Hentze, H.-P.; Raghavan, S. R.; McKelvey, C. A.; Kaler, E. W. Langmuir 2003, 19, 1069. (20) Park, J.-H.; Oh, C.; Shin, S.-I.; Moon, S.-K.; Oh, S.-G. J. Colloid Interface Sci. 2003, 266, 107. (21) Yang, M.; Zhu, J.-J. J. Cryst. Growth 2003, 256, 134. (22) Hu, Y.; Chen, J.; Chen, W.; Lin, X.; Li, X. Adv. Mater. 2003, 15, 726. (23) Collins, A. M.; Spickermann, C.; Mann, S. J. Mater. Chem. 2003, 13, 1112. (24) Li, Y.; Shi, J.; Hua, Z.; Chen, H.; Ruan, M.; Yan, D. Nano Lett. 2003, 3, 609. (25) Wu, M.; Wang, G.; Xu, H.; Long, J.; Shek, F. L. Y.; Lo, S. M.-F.; Williams, I. D.; Feng, S.; Xu, R. Langmuir 2003, 19, 1362. (26) Yang, H. G.; Zeng, H. C. Angew. Chem., Int. Ed. 2004, 43, 5206. (27) Wang, C.; Tang, K.; Yang, Q.; Hu, J.; Qian, Y. J. Mater. Chem. 2002, 12, 2426. (28) Xiong, Y.; Xie, Y.; Li, Z.; Wu, C.; Zhang, R. Chem. Commun. 2003, 904.

combined hybrids and often exhibit newer chemicophysical properties that are different from those of their solid counterparts.1-34 Owing to their unique structural, optical, electrical, magnetic, thermal, and surface properties, applications of the materials of this kind are diverse, ranging from drug delivery and protection of biologically active agents and artificial cells to low dielectric constant materials, acoustic insulators, photonic crystals, shapeselective adsorbents, nanocatalysts, and fillers, etc.1-34 Concerning the fabrication of hollow nanomaterials, there have been two main categories of preparative methods: (i) the template-directed synthesis1-15 and (ii) the emulsion synthesis.16-26 The basis of the template-directed synthesis is adsorption of nanoparticles or polymerization on modified polymeric (e.g., polystyrene)1-10 or inorganic (e.g., SiO2)11-15 template surface and subsequent removal of the template by calcinations or dissolution with solvents. In the emulsion synthesis, on the other hand, the solution is emulsified and the adsorption or reaction then takes place on the surface of sol droplets (micelles) to form the hollow spheres. In fact, the latter methods can also be viewed as another version of templating, and after reaction, the “soft” template can be removed directly from the formed hollow spheres.16-26 In addition to the above common methods, there are several special synthetic strategies that can generate spherical hollow structures.27-34 Among them, interestingly, well-known physical phenomena such as “Ostwald ripening”35 and the “Kirkendall effect”32 have been utilized very recently for fabrication of hollow metal oxide nanospheres through direct solid evacuations, or through self-templating.31,32 In this article, we will demonstrate that chemical composition, structure, and photoelectronic properties of (29) Matsuzawa, Y.; Kogiso, M.; Matsumoto, M.; Shimizu, T.; Shimada, K.; Itakura, M.; Kinugasa, S. Adv. Mater. 2003, 15, 1417. (30) Peng, Q.; Dong, Y.; Li, Y. Angew. Chem., Int. Ed. 2003, 42, 3027. (31) Yang, H. G.; Zeng, H. C. J. Phys. Chem. B 2004, 108, 3492. (32) Yin, Y.; Rioux, R. M.; Erdonmez, C. K.; Hughes, S.; Somorjai, G. A.; Alivisatos, A. P. Science 2004, 304, 711. (33) Park, S.; Lim, J.-H.; Chung, S.-W.; Mirkin, C. A. Science 2004, 303, 348. (34) Liu, B.; Zeng, H. C. J. Am. Chem. Soc. 2004, 126, 8124. (35) (a) Ostwald, W. Z. Phys. Chem. 1897, 22, 289. (b) Ostwald, W. Z. Phys. Chem. 1900, 34, 495.

10.1021/la047671l CCC: $30.25 © 2005 American Chemical Society Published on Web 12/24/2004

Colloidal CuO Nanocrystallites

Langmuir, Vol. 21, No. 3, 2005 1075

Figure 1. The synthetic flow chart developed in the present work: (i) formation of primary CuO nanocrystallites; (ii) spherical aggregation of CuO; (iii) reductive conversion of CuO to Cu2O; and (iv) crystallite growth and cavity formation. Inset shows the color change of a series of hollow Cu2O nanospheres formed from the experiments: 30 mL of [Cu2+] ) 0.010 M at 180 °C for 4, 7, 10, and 14 h, respectively.

semiconductor nanoparticles can be stepwise controlled by forming and gathering primary nanocrystallites, solidsolution redox reactions, and subsequent crystallite aging and solid evacuation. In particular, a process mechanism has been revealed for synthesis of cuprous oxide Cu2O (a p-type semiconductor with a band gap of 2.17 eV) nanospheres: (i) formation of CuO nanocrystallites; (ii) spherical aggregation of primary CuO crystallites; (iii) reductive conversion of CuO to Cu2O; and (iv) crystal aging and hollowing of Cu2O nanospheres. For the first time, a wealth of colorful Cu2O hollow nanospheres, with variable Eg in the range of 2.405-2.170 eV, has been fabricated using the current one-pot synthetic approach. Experimental Section In our typical single-step experiments, 30 mL of 0.005-0.010 M Cu(NO3)2 solution (a certain quantity of Cu(NO3)2‚3H2O was dissolved in organic solvent N,N-dimethylformamide (DMF) to make a stated concentration) was sealed in a Teflon-lined stainless steel autoclave with 50 mL capacity and then heated at different temperatures (150, 160, 170, and 180 °C) for different times to form Cu2O nanoproducts. In a typical two-step experiment, the above-mentioned solution was heated at 140-150 °C for 22-40 h and then continuously heated at 180 °C for 8-42 h. After reaction, the autoclave was cooled with tap water and the Cu2O products were washed with pure ethanol three times and then dried in a vacuum system. Details on the experimental parameters can be found in Supporting Information (SI-1). The crystallographic information of the samples was investigated by powder X-ray diffraction (XRD, Shimadzu, model XRD-6000, Cu KR radiation λ ) 1.5406 Å). Morphological investigation was carried out with transmission electron microscopy (TEM, Joel model JEM-2010, 200 kV), and scanning electron microscopy (SEM, Joel model JSM-5600LV). The UV-visible absorption spectra of suspensions of Cu2O hollow nanospheres synthesized with different reaction times were measured by a UV-vis-NIR scanning spectrophotometer (Shimadzu, model UV-3101 PC; with ethanol as a solvent).

Results and Discussion Figure 1 illustrates the synthetic process developed in the present work and the related color changes upon reaction process time. The evolution of the crystal morphology of samples at 150 °C is reported in Figure 2A. In this TEM study, it is found that the product formed within the first 4 h comprised small nanocrystallites in both CuO and Cu2O phases, which was confirmed with powder XRD (CuO space group C2/c, ao ) 4.684 Å, bo ) 3.425 Å, co ) 5.129 Å, β ) 99.47°, JCPDS file no. 05-0661; Cu2O space group Pn3m, ao ) 4.4267 Å, JCPDS file no. 05-0667;36 SI-2), and some of these crystallites (Figure 2A, inset: a (36) Chang, Y.; Zeng, H. C. Cryst. Growth Des. 2004, 4, 273 and the references therein.

selected area electron diffraction (SAED) ring pattern of CuO phase) have attached each other into spherical aggregates (Figure 2B) with diameters in the range of 100200 nm. The inset in Figure 2B also shows some characteristic of Cu2O diffractions superimposed on the CuO phase. Other dispersive nanocrystallites have also eventually gathered, accompanying with a phase transformation from CuO to Cu2O (i.e., reductive transforming reaction; confirmed with XRD, SI-2), after longer reactions (Figure 2C (inset: a fully developed SAED ring pattern of Cu2O) and Figure 2D). The crystallite agglomeration was then followed by a solid core evacuation, as evidenced in Figure 3. The TEM image contrasts show that hollowing takes place gradually in the central cores of Cu2O nanospheres, where the lighter central parts correspond to the void spaces. The selected area electron diffraction (SAED; Figure 3D) rings are relatively sharp and can be assigned perfectly to polycrystalline Cu2O. It is then understood that the nanospheres of Cu2O did not undergo further chemical reactions but went through a recrystallization process (i.e., Ostwald ripening) in the core evacuation. Although the formation of smaller crystallites is kinetically favored during the initial agglomeration, larger crystallites are thermodynamically favored.35 In Figure 3, indeed, the void space in the Cu2O nanospheres is getting bigger with aging time. This is because the smaller crystallites located at central cores have higher surface energy and they tend to relocate themselves to the shell parts during the Ostwald ripening. On the basis of the above results, the process steps can be summarized: (i) formation of CuO nanocrystallites; (ii) spherical aggregation of primary CuO crystallites; (ii) reductive conversion of CuO nanospheres to Cu2O; and (iv) crystal aging and evacuation of Cu2O cores via Ostwald ripening. Since no strong reducing agents were added in our starting solution, it is thought that N,N-dimethylformamide (DMF) solvent has been working as a weak reducing agent in synthesis; the chemical reactions are proposed as follows:

HCON(CH3)2 + H2O f HCOOH + NH(CH3)2 (1) Cu2+ + H2O + 2NH(CH3)2 f CuO(s) + 2NH2(CH3)2+ (2) 2CuO(s) + HCOOH f Cu2O(s) + H2O + CO2 (3) The formations of the solid phases in eqs 2 and 3 have been confirmed in our above XRD/SAED investigations. Apparently, the formic acid HCOOH in eq 1 is a reducing agent generated from hydrolysis of DMF (water molecules

1076

Langmuir, Vol. 21, No. 3, 2005

Chang et al.

Figure 2. Formation of primary CuO nanocrystallites, spherical aggregation of CuO nanocrystallites, and chemical reduction of CuO to Cu2O: TEM images and SAED patterns of the samples prepared after different reactions times (A and B, 4 h; C, 14 h (with the SAED pattern); D, 23 h) at 150 °C; starting solution, [Cu2+] ) 0.010 M, 30 mL.

Figure 3. The core hollowing process in Cu2O nanospheres: TEM images of the samples prepared after different reactions times at 150 °C (A and B, 35 h; C, 50 h; D is the SAED pattern of the sphere shown in C); starting solution, [Cu2+] ) 0.010 M, 30 mL.

from Cu(NO3)2‚3H2O, Experimental Section). As a simple confirmation for this reaction, it is noted that the starting copper solutions were indeed turned to basic and had a odor of amines after hydrolysis of DMF.37-39

The above chemical and physical processes can be accelerated when a higher process temperature is used. For example, the conversion of CuO to Cu2O had been largely completed after 4 h of reactions at 160-170 °C

Colloidal CuO Nanocrystallites

Langmuir, Vol. 21, No. 3, 2005 1077

Figure 4. TEM images of the samples prepared after different reactions times (2-20 h) at 180 °C; starting solution, [Cu2+] ) 0.010 M, 30 mL. White arrows indicate observable hollow spheres in some short reaction time cases (4 and 7 h); all bar scales ) 50 nm.

(SI-3 & SI-4). On the basis of our XRD results, it is noted that the sample synthesized at 180 °C for only 2 h has already been primarily in monovalent phase Cu2O (SI-5). When the reaction was prolonged to 3 h, the remaining divalent phase was transformed almost completely to Cu2O. When the reaction was continued for 20 h, however, a small amount of metallic copper appeared (SI-5), indicating there is a time limit to obtain phase-pure Cu2O, which, interestingly, may also provide a way for fabricating Cu-Cu2O composite nanospheres though this is not the prime effort of our present study. The crystal morphologies of the above samples are further examined in Figure 4 (180 °C). In addition to the early phase conversion to Cu2O, the core hollowing also takes place much faster (4 h, Figure 4), since the Ostwald ripening is more efficient at higher temperatures. In this regard, the phase conversion and core hollowing virtually proceed simultaneously at higher temperatures, and the process boundary between the phase conversion and core evacuation is no longer clear. In good agreement with the XRD findings, the shells of nanospheres indeed become thinner and less compact when some of the Cu2O crystallites were reduced to metallic copper (20 h, Figure 4). Although the reaction rate and mass transfer rate are both increased at higher temperatures, it is found that the hollow nanospheres formed at 180 °C (Figure 4) are not as round as those prepared at 150 °C (Figure 3). To obtain a better shape control, a combined approach has been developed taking advantage of the temperature effect. (37) Yu, J. Y.; Schreiner, S.; Vaska, L. Inorg. Chim. Acta 1990, 170, 145. (38) Pastoriza-Santos, I.; Liz-Marza´n, L. M. Pure Appl. Chem. 2000, 72, 83. (39) Liu, S.; Wang, C.; Zhai, H.; Li, D. J. Mol. Struct. 2003, 654, 215.

This method is based on the following observations: (i) a low process temperature is suitable for forming symmetrical Cu2O nanospheres (Figure 2); and (ii) a high process temperature is suitable for rapid core evacuation (Figure 4). Some hollow nanospheres prepared with this two-step method are presented in Figure 5 (also see SI-6). It can be seen that the longer the cumulative time, the emptier the cores are (Figure 5A (22 h + 8 h) vs Figure 5B (24 h + 10 h)). It is clear that the samples synthesized are emptier than those synthesized at 150 °C (50 h, Figure 3C) while the time can be shortened significantly. With the present methods, high product uniformity and a 100% morphological yield can be easily attained via controlling reaction time, even up to formation of the Cu-Cu2O composites (outer diameters in 100-200 nm, SEM results, SI-7). In parallel to the core hollowing, nanocrystallite size in shell structure can be further controlled and thus optical band gaps of Cu2O hollow spheres can be fine-tuned. To examine this effect, the crystallite size upon aging time was further quantified with the Debye-Scherrer equation.40 It should be mentioned that since relative intensities of XRD peaks for Cu2O phase did not change in all our samples, data deduced with this method should be highly reliable. As complied in Figure 6A, the Cu2O crystallite growth rate rises sharply in the initial stage (2 and 3 h). The rate then becomes much slower over the range of 4-20 h, but an increasing trend is definitely observable. To correlate crystallite size to the optical band gaps, UVvisible absorption spectra of the hollow Cu2O nanospheres prepared with different reaction times were measured (SI-8), and a classical Tauc approach was employed to (40) Cheetham, A. K.; Day, P. Solid-State Chemistry: Techniques; Clarendon Press: Oxford, 1987; p 79.

1078

Langmuir, Vol. 21, No. 3, 2005

Chang et al.

Figure 5. TEM images of the samples prepared with two-step heating routines (A, 150 °C for 22 h + 180 °C for 8 h; B, 150 °C for 24 h + 180 °C for 10 h; C and D, 150 °C for 26 h + 180 °C for 42 h); starting solution, [Cu2+] ) 0.010 M, 30 mL. The inset is a SAED pattern (essentially Cu2O type) from the spheres shown in C. (Note: due to a prolonged reduction reaction in this synthesis, aggregated metallic copper was also detected in other parts of sample (not shown) in addition to the Cu2O hollow spheres.)

based on the direct transition. The extrapolated value (the straight lines to the x-axis) of Ephoton at R ) 0 gives an absorption edge energy corresponding to Eg ) 2.405 eV (4 h), 2.385 eV (7 h), 2.340 eV (10 h), and 2.170 eV (14 h), respectively. When the reaction time is 14 h, the optical band gap obtained herein is essentially equal to that of the natural single-crystal Cu2O reported in the literature (2.1722 eV).42 The quantum confinement threshold of Cu2O nanocrystallites is therefore ca. 14 nm (14 h, Figure 6A) in these nanospheres. In excellent agreement with the above band gap measurements, the colloidal solutions of these Cu2O nanospheres show a series of color changes upon the process time of the present approach (inset, Figure 1).43,44 The observed color varies from bright yellow to orange, brown, and green, when the size of Cu2O crystallites was gradually increased along the simultaneous hollowing process. Conclusion Figure 6. (A) Deduced crystallite size from the DebyeScherrer method (based on (111) reflection of Cu2O phase; data from SI-5). (B) Representative plots of (REphoton)2 versus Ephoton for the direct transition; band gap energies of hollow Cu2O nanospheres obtained by extrapolation to R ) 0 (the samples were diluted in ethanol solvent in these measurements; see SI-8). Except for variation in reaction time, other experimental parameters were kept identical for all these samples: [Cu2+] ) 0.010 M, 30 mL, and 180 °C.

estimate their optical energy band gaps.41It has been found that n ) 1/2 (allowed direct transition) gives the best description for all absorption measurements. Figure 6B shows four representative plots of (REphoton)2 versus Ephoton (41) Tsunekawa, S.; Fukuda, T.; Kasuya, A. J. Appl. Phys. 2000, 87, 1318.

In summary, a new chemical approach has been devised to fabricate cuprous oxide Cu2O hollow nanospheres without the assistance of solid template. On the basis of the various growth experiments and materials characterization, it is understood that the formation mechanism comprises four different steps: (i) generation of primary nanocrystallites of CuO; (ii) spherical gathering of the primary CuO; (iii) reductive conversion of CuO to Cu2O; and (iv) crystal aging of Cu2O and formation of hollow nanospheres. In the final step (iv), Ostwald ripening has (42) Matsumoto, H.; Saito, K.; Hasuo, M.; Kono, S.; Nagasawa, N. Solid State Commun. 1996, 97, 125. (43) Hayashi, M.; Katsuki, K. J. Phys. Soc. Jpn. 1950, 5, 380. (44) Kleinman, L.; Mednick, K. Phys. Rev. 1980, B21, 1549.

Colloidal CuO Nanocrystallites

been utilized in controlling crystallite size of shell structures and thus for effectively tuning the optical band gap energy of Cu2O (in the range of 2.405-2.170 eV). Considering their unique hollow structure and facile tuning in band gap energy, the prepared Cu2O nanospheres can be potentially used for harvesting solar energy in the visible range. The solid-solution redox reactions

Langmuir, Vol. 21, No. 3, 2005 1079

devised in this work may also be extendable to future fabrication of Cu-Cu2O nanocomposites. Supporting Information Available: A list of experiments, XRD results, TEM micrographs, and UV-vis absorption spectra. This material is available free of charge via the Internet at http://pubs.acs.org. LA047671L